The State, Volume 3 (Critical Concepts)

THE STATE: Critical Concepts This page intentionally left blank. THE STATE: Critical Concepts Edited by John A.Hal...

Author: John Hall

24 downloads 1585 Views 4MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

THE STATE: Critical Concepts

This page intentionally left blank.

THE STATE: Critical Concepts

Edited by John A.Hall VOLUME III

London and New York

First published 1994 by Routledge 11 New Fetter Lane, London EC4P 4EE Simultaneously published in the USA and Canada by Routledge 29 West 35th Street, New York, NY 10001 Routledge is an imprint of the Taylor & Francis Group This edition published in the Taylor & Francis e-Library, 2005. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” Selection and Editorial Matter © 1994 John A.Hall All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue reference for this book is available from the British Library. Library of Congress Cataloging in Publication Data has been applied for. ISBN 0-203-41992-8 Master e-book ISBN

ISBN 0-203-72816-5 (Adobe eReader Format) ISBN 0-415-08683-3 (Boxed Set) ISBN 0-415-08682-5 (Vol. III)

Contents VOLUME THREE PART 3: THE POLITICS OF ECONOMICS (Continued) SECTION 4: THE QUESTION OF HEGEMONY (Continued) 42. Will the United States Decline as did Britain? John A.Hall PART 4: STATES AND POLITICAL BEHAVIOUR Commentary SECTION 1: STATES AND SOCIAL MOVEMENTS Commentary 43. Ruling Class Strategies and Citizenship Michael Mann 44. Revolutionary Movements in Central America: A Comparative Analysis Jeff Goodwin 45. Party and Patronage: Germany, England, and Italy Martin Shefter SECTION 2: STATES AND ETHNICITY Commentary 46. Working-Class Formation and the State: Nineteenth-Century England in American Perspective Ira Katznelson 47. Ethnic Competition and Modernization in Contemporary Africa Robert H.Bates 48. Patterns of Ethnic Separatism Donald L.Horowitz

5

35

39 40 57 99

141

142 167 191

SECTION 3: LIBERALISATION AND DEMOCRATISATION Commentary

222

49.

50. 51. 52. 53.

Democratic Transition and Breakdown in Western Europe, 1870–1939: A Test of the Moore Thesis John D.Stephens The Politics of Backwardness in Continental Europe, 1780–1945 Andrew C.Janos Social Democracy as a Historical Phenomenon Adam Przeworski Transitions to Democracy: Toward a Dynamic Model Dankwart A.Rustow Dilemmas of Democratization in Latin America Terry Lynn Karl

SECTION 4: WELFARE STATES Commentary 54. Social Policy and Distributional Conflict in the Capitalist Democracies. A Preliminary Comparative Framework Walter Korpi 55. The Three Political Economies of the Welfare State Gosta Esping-Andersen 56. The Political Origins of America’s Belated Welfare State Ann Shola Orloff SECTION 5: AMERICAN EXCEPTIONALISM Commentary 57. Why No Corporatism in America? Robert H.Salisbury 58. Constitutionalism, Class, and the Limits of Choice in U.S.Foreign Policy Ira Katznelson and Kenneth Prewitt CONCLUSION: THE END OF THE STATE? Commentary 59. Three Models of the Future Robert Gilpin 60. Supranational and the State Susan Strange 61. Who Is “Us”? Robert B.Reich 62. Beyond the Nation-State: The Multinational State as the Model for the European Community M.Rainer Lepsius

224 274 303 332 352

377

379 400 424

465 467 482

496 498 521 535

546

PART THREE: The Politics of Economics (Continued)


SECTION FOUR: The Question of Hegemony (Continued)


The state: critical concepts

5

42 Will the United States Decline as did Britain?* John A.Hall

* Source: M.Mann, ed., The Rise and Decline of the Nation State, 1990, ch. 6, pp. 114–145.

Determining whether the United States is going to decline as did Britain has clearly become one of the questions of the age, with the presumption in the United States distinctively being that the answer is going to be positive. Thus Paul Kennedy’s The Rise and Fall of the Great Powers lends itself to this interpretation, despite many cautionary words of the author.1 Similarly, the first 1988 issue of the trade journal of American international political economists, International Organisation, has several articles whose analyses take for granted that American hegemonic decline has already occurred. If this is believed, it is likely to affect policy in a dramatic manner. One broad thrust of this paper is sceptical of the claim that the United States will decline as did Britain. Two points are made against the Cassandras of decline. First, the extent of American decline tends to be exaggerated. A few others have seen this,2 but it is noticeable that their arguments have failed to dent the selfconfidence (no lesser word will do) of what the New York Times refers to as ‘the school of decline’.3 If evidence of American power is adduced, the characteristic reply tends to be that this does not really weigh much against what is considered to be a long-term secular trend: the United States may merely be at the stage of Britain in, say, 1880 rather than 1931—what is held to matter is that worse is sure to come. My second point may help to resolve this stand-off in debate. Extremely forceful, if banal, considerations suggest that the world polity facing the United States now is nothing like that which faced Britain at the end of the nineteenth century. The differences, moreover, are systematically to the advantage of the United States; in consequence, it is extremely unlikely that it will lose preeminence so completely as did Britain. These points can be encapsulated by saying that the analogy, whether implicit or explicit, between the undoubted decline of Britain and the contemporary situation of the United States is of limited usefulness. But there is another side to my general argument. To stress fundamental differences is not to dispute that there has been some decline in the position of the United States, nor would it be sensible to deny that it might go further. Different processes of decline are identified, with especial focus being given to the situation of ‘hegemonic leaders’. More particular and detailed attention still is devoted to the ways in which adherence to certain aspects of liberalism accounts for decline. An implication of this last point deserves highlighting. It may well be that hegemonic leadership in capitalist society needs to adhere to the liberalism’s insistence on the virtues of free trade, but there is no reason to believe that the Anglo-Saxon preference for well developed equity markets and for limiting state power should be seen as dictated by, or

Will the United States Decline as did Britain?

6

necessary to, hegemonic rule. In general, it will be maintained that the causes of the decline of the contemporary United States do resemble those recognizable from the British case. The processes identified are not those considered important by the most significant theory asserting hegemonic decline’; some credence is lent instead to the view that the particular nature of Anglo-Saxon liberalism is of importance in explaining decline. Drawing up a balance sheet about the extent of American decline is a difficult task, but it is undertaken in the conclusion; this is a necessary preliminary to the final answer to the question posed. But first we must immediately consider different theories of decline, and certain implications which they raise; we can then analyse the British and American cases in turn.

Theories of Decline, Modern and Classical We can start to gain some grasp of the nature of decline as an analytic category by noting and then commenting on three general causes of decline usefully identified by Mann.4 First, economically powerful nation states which swim inside the sea of capitalist society are prone to suffer decline as the practices which account for their ascendancy diffuse throughout the larger society. Secondly, leading states tend to decline as the result of geopolitics, either because of over-extension or because of actual participation in war. Finally, societies tend to institutionalize the moments of their success, and thereby make it difficult to be as endlessly flexible as the demands of capitalist society necessitate; differently put, distributional coalitions are created which make social adaptation difficult.5 The most obvious comment needing to be made about these categories is that decline is seen as having two sources, either internal or external. The diffusion of practices throughout capitalist society is thus a more or less inevitable external cause of decline given that comparative advantage in general and the advantages of backwardness in particular have always allowed developing states faster growth paths than those of mature economies; in the same spirit, it must be said very clearly that some decline in the position of the United States in the last half century was made inevitable through the recovery of key economic competitors from a situation of considerable internal destruction—a recovery that the United States actively sought, largely for geopolitical reasons. Geopolitics can be as much a force affecting states from the outside, as when heavy expenditure for defence is made necessary by the presence of a ruthless, aggressive and powerful competitor; equally, however, it can be ascribed to internal factors, as when a rash elite foolishly and unnecessarily over-extends commitments. Social blockages, of course, are by definition to be considered entirely an internal matter. A more subtle point follows from this. Decline has two connotations which need to be clearly distinguished. On the one hand, decline is normal and inevitable, the result, as noted, of factors beyond the power of any single nation state. On the other hand, decline is seen as being linked to degeneracy and corruption; here the implication is of failure to do as well as one could. In consequence, we can say that the key question about recent British history is not relative decline per se, but the fact that this relative decline was so steep—becoming in fact


7

absolute between 1979 and 1984. All that has been said to this point refers to the decline of nation states in general. Let us now turn to an important recent theory—in order to advance an argument rather than to add, at least at this stage, to the history of ideas that adds a further element to the picture. The theory in question has usefully been dubbed ‘hegemonic stability theory’, and it has gained general intellectual ascendancy in explaining the fateful careers of Britain and the United States.6 The most obvious tenet of the theory is that the stability of capitalist society as a whole depends upon the presence of a leading power prepared to exercise hegemony—a term with rather different connotations, to be noted in due course—so that certain public goods can be assured. For Kindleberger the most important such goods are a stable medium of exchange, the insistence on free trade, the export of capital for development and the ability to absorb world surplus capacity.7 The cogency of this claim seems much reinforced by considering the other side of the coin. The absence of a single hegemon before 1914 when Imperial Germany challenged Britain, and in the inter-war vacuum when the latter was too weak and the United States unwilling to provide leadership—is held to have contributed to chaos, and then to war. All this is striking, if somewhat tautologous, and much of it has been accepted even by critics of hegemonic stability theory. Thus a liberal institutionalist such as Keohane,8 whose somewhat implausible argument—derived, it should be noted, far more from theory than from evidence—is that cooperation between leading capitalist powers is possible in the absence of a hegemon, also fears the wars that can result from the way in which changes in relative national economic power affect the conduct of states; it is simply his prescription for world order that differs. In passing, however, it is worth noting that a genuinely radical critique, that of George Liska, is available.9 He argues that war tends to result when a balance of power is disrupted by a single state, notably Britain in the nineteenth century, which gains too much power and thereby calls forth no-holds-barred rivalries. Interestingly, a follower of Liska, David Calleo,10 has been led by the logic of this theory to resist frequent cries to prop up failing hegemonic power; it would instead be better to knock the hegemon off its perch as quickly as possible so that we can return to normal balance of power politics. This paper is not centrally concerned with these key elements of hegemonic stability theory. Concentration is instead on the claim, not always spelt out clearly in the literature but distinctively a part of current political debate, that hegemonic power is necessarily self-liquidating, that leadership is such as to sap the strength of the power that provides it. Two key examples of this process are often given. First, hegemonic powers often allow industrializing states to protect their infant industries whilst continuing to offer access to their own rich and well-developed markets.11 In the long run, this lack of genuine multilateralism undermines the hegemon’s domestic industry, a claim made most strikingly by Representative Richard Gephardt in the US presidential campaign of 1988. A second service that the hegemon is seen as providing is that of defence against common enemies. Here too the hegemon is held to suffer as the result of carrying a disproportionate share of the burden; Olson and Zeckhauser12 are typical here in using the language of collective goods theory in claiming that it is in the rational interest of smaller states to free ride on their leader. The economic consequences of this are held to be catastrophic. Thus, William Grieder13 is a characteristic voice in claiming that the


8

United States spends a much larger share of GNP on defence than its main trading partners, the presumption being that this plays a major part in explaining differential growth rates within capitalist society. The most discussed example here is NATO, perhaps curiously given that the situation of Japan makes the case much better. But in either case, the analysis results in calls for a renewed round of ‘burden-sharing’ negotiations. All these theories consider that there are economic and geopolitical dimensions of both the rise to and decline from preeminence, as well as complex interrelations between them; I shall follow their lead. What is less obvious and so more in need of underlining is the implicit moral claim of the thesis that hegemonic power is necessarily selfliquidating. Jean-Paul Sartre famously declared that ‘hell is other people’. Hegemonic stability theory implies that decline is ‘other people’. The hegemon slaves away in the long-term general interest, but other states, selfishly thinking only of their own short-term goals, slowly undermine what is in fact the very source of their own prosperity. Thank goodness that virtue is its own reward! Sustained scepticism about this claim will return us to alternate theories of decline made forcefully aware of their very different moral flavour. Whilst it is very important indeed to recognize that key elite members as well as large sections of public opinion of hegemonic powers do sometimes see themselves in Durkheimian terms as providers of norms and services for the larger society, it is crucial to remember the view from below. Let us recall the comment of the British general Calgacus which Tacitus records in The Agricola: ‘To plunder, steal, rape, these things they falsely call imperial rule; they make desolation and call it peace.’14 Words quite as strong have been directed against the British and the American desire for open markets. Although openness is, according to liberal theory, designed to benefit all, the worm’s eye view, especially as formulated by Alexander Hamilton and his German disciple Friedrich List, naturally leads to charges of hypocrisy—for some are more equal than others, and economically advanced nations in capitalist society naturally favour openness when they have the capacity to make the most of it.15 This general point gains further salience once we remember the way in which leading powers tend to behave when they consider their power to be diminishing. Thus the United States has chosen to run a budget deficit nearly every year in the last quarter century; that is, it has broken the norms which its agency, the IMF, has so rigorously imposed on other states. This raises the analytic possibility that it is the hegemon which might choose to free ride on the system, that is, to use its undoubted muscle to extract systematic advantage from smaller and weaker powers. Still more important is the fact that those tasks judged to be common by the United States are often seen in an entirely different light by its allies, as evidenced by European members of NATO refusing to accept ‘out of area’ tasks, to the evident anger of the United States. This sceptical note amounts to saying that a hegemon is not necessarily a neutral normgiver akin to the Latin Christian church in the early European middle ages, as portrayed by Michael Mann.16 On the contrary, a hegemon is a great power. Putting things like this makes it possible to contrast theories of decline which blame others with theories which blame the great powers themselves. We have seen that one such theory, present in the works of Polybius, Machiavelli, Gibbon and Montesquieu, concentrates on the way in which arrogance can lead to disastrous geopolitical over-extension. The rise of a great power is held to result from its having occupied a marcher position, a factor often linked


9

to military prowess; decline results from over-extension which brings with it the costs of twofront wars and of the repression of nationalities.17 Decline is the result of the great power mistakenly overextending itself in a way that entails ruin, both geopolitically and economically. The other theory of this type, as noted, stresses the diffusion of practices throughout capitalist society. Adam Smith insisted against his friend David Hume that there was no inevitable reason why a rich country should not continue to stay rich; rather by moving up the technological ladder it would serve the general interest in allowing new states to take over old industries.18 But this would only be possible, Smith asserted, as long as there were no ‘errors of policy’, that is, no governmental policies which allowed the institutional arrangements of one moment to gel and petrify. If Smith’s theory had much to be said in its favour, it proved nonetheless to be a poor guide to practice. The very notion of society involves ‘gelling’ of expectations and institutions, as Polanyi famously stressed,19 and no nation state within capitalist society has as yet been able to perpetuate the moments of its success. It is worth noting how very difficult it is to imagine this happening. The rise to power is often associated with the possession of a particular manufacture. However, it is possible, as Gershenkron,20 following Veblen, noted some time ago, for an industrializing state either to copy the technology without accepting various institutional limitations and social achievements, or to pioneer a new technology that is closed off to the leading power because of that institutional mix; both factors contribute to making the growth path of rapidly industrializing countries faster than that of the leading power. In addition, the leading power tends, in its senescence, to export the capital its prior industrial success had created; this necessitates export-led industrialization on the part of developing societies; however, such exports to repay loans are likely to undermine the domestic industry of the leading power. All this can be summarized by saying that hegemonic stability theory tends to assume that their place in the world would have been assured except for the burdens that the hegemon was forced to bear. I am suggesting that we should not accept this too easily. The leading edge of capitalist society has never remained in the same place for long, and there is no a priori reason to believe that it would have remained in the United States for ever had that country not entered the world scene. What has been said to this point can be summarized and highlighted. A diminution in power of a leading state inside capitalist society might be explicable in traditional terms, as the result of over-extension or internal social rigidities, rather than as the result of the cost of services that it provides for capitalism as a whole. We should in particular be suspicious of ideologists claiming that hegemonic services are somehow neutral; differently put, we should remember de Gaulle’s complaint that the foreign policy of the United States represented a traditional drive for ascendancy cloaked in idealism—a complaint that was raised quite as much against Britain in the nineteenth century. Nonetheless, it would be a mistake not to underline a key difference between the traditional policy of great powers and that sought by Britain and the United States. Where traditional rule often favoured conquest, Britain and the United States preferred when possible to maintain an empire of free trade; a contrast needs to be made, in other words, between formal and nonterritorial empire. The success or failure of ruling powers within capitalist society is likely to depend upon the extent to which they can avoid formal territorial possession.21 If liberalism’s aims here might in principle stave off decline, the


10

same may well not be true of adherence, as noted, to other aspects of that protean doctrine. One final set of theoretical observations is necessary before turning to a comparison of the British and American cases. If real purchase is to be gained in this comparison, something needs to be said about the sources of state economic and geopolitical policy. Bluntly, to speak without question of ‘the state’ in the international arena is to prejudge key matters, to fail to examine whether state leaders act at the behest of societal forces. It is not at all easy to separate key social actors from each other. Was American post-war policy designed to secure a free world or a world open to a free economy? To what extent were such policies those of an elite genuinely autonomous from key capitalist actors? Occasionally such separation is analytically impossible: Cobden and Bright really believed that the progress of trade would serve the politics of peace. Nonetheless, it is often possible to locate moments of autonomy for particular social actors; it is worth while saying something about the particular actors that will have importance in the analysis given in this chapter. The external actions of states tend to be seen in two ways. Realism in all its forms stresses that state leaders seek security for their society in an asocial world; in the modern world, this search has necessarily had to involve economics quite as much as traditional geopolitics. This is a rich theoretical tradition, but it is capable of being brutalized; Morgenthau did precisely this when he argued that the goal of every state was to enhance its power.22 In fact, the goals of states can and do vary, with the desire for peace or glory being quite as real as the desire for power.23 Another way of making this point is to insist that perceptions of security vary; such variations may result from national experience as a whole but they may, of course, reflect the perceptions of particular groups within a nation. The force of realism in the contemporary world seems amply justified by the behaviour of states in Southeast Asia since the withdrawal of the United States. Great powers like to have weaker ones on their borders, and it is not at all surprising to discover China’s hostility to Vietnam, particularly as seen in its adherence, in its support of Pol Pot, to the old geopolitical maxim that ‘the enemy of my enemy is my friend’.24 I mention this because the dominant version of the second way in which foreign policy has been conceptualized, namely that stressing its domestic roots, argues that the control of the state by capitalists causes war. This theory was first adumbrated by Hobson, and it is the way in which mainstream Marxism, which ought to be more troubled than it is by the sheer fact that capitalism is divided by states, has sought to explain state behaviour. As this approach is both powerful and well known, let me pay more attention to another domestic factor in the making of foreign policy, not least because its improperly understood implications are intellectually exciting and morally disturbing. In so far as liberalism divorced itself from capitalist theory, it produced a distinctive theory arguing for popular control over state policy, a theory which reached an idiosyncratic apogee in the American founding fathers’ desire to continue the spirit of their revolt against the state by controlling power completely by splitting responsibility between Congress and the Executive.25 This tradition raises very complex issues, but two basic approaches can be distinguished, the more negative of which has tended to dominate debate. Tocqueville was amongst the first who stressed the negative side of the


11

equation in the American case.26 He argued that the people would be slow to anger, but remorseless in the execution of war. In either case, it would be difficult to conduct foreign policy according to realist principles: state leaders would be constrained from threatening when threats were needed, and incapable, because of popular passion, of calling a halt once war had ceased to fulfil its Clausewitzian role as a ‘continuation of policy by other means’. This negative view, especially in connection with the United States, has been further stressed by George Kennan,27 the author in the late 1930s of a manuscript seeking to curtail democracy, and by Henry Kissinger.28 One reason for caution in accepting this view is the plentiful evidence of consensus between the people and foreign policy makers; Kissinger’s habit of blaming foreign policy failures on the domestic political system occasionally masked his own misjudgements. Of course, this is to voice scepticism about the extent to which popular will constrains the makers of foreign policy. But there is a more positive view of the way in which popular will does constrain foreign policy elites. One element to this viewpoint received its most vigorous statement in Kant’s Perpetual Peace (1795).29 He argued that liberal states should establish a league, based on constitutionalism and economic interdependence, and that this would guarantee peace. In a brilliant article, Michael Doyle, in common with Margaret Thatcher, has pointed out that liberal states have a remarkable record of not going to war with each other—albeit he adds the cautionary note that the ideological nature of liberal states may make them particularly warlike to those they judge to be not just geopolitical but ideological rivals as well.30 A second element to a positive appreciation of liberalism is that an appeal to the people may enable the foreign policy elite to escape from sectoral pressure: the state may become free from particular groups if it can insist on a right to be free to represent a more general will. Nonetheless, the ultimate validation of liberalism remains the capacity of preventing state elites making chaotic mistakes. As it happens, I believe that a liberal political system has opportunities and costs for the conduct of foreign policy. In order to see the ways in which this is so, and to cast light on the other issues raised, it is necessary, given that they cannot be resolved by fiat, to turn to the historical record. Let us first examine, in broad contours, the reasons for the decline and fall of Britain, and then turn to the more complex case of the United States in the modern world political economy.

Britain’s Decline and Fall Most British academics have their own theories about British decline,31 and the version presented here amounts in large part to a gloss upon them. Before considering decline, however, it is as well to remember the nature of initial British pre-eminence and, still more important, for reasons which will become apparent, to recount the way in which this pre-eminence came to be challenged. Britain entered the world stage at the end of the eighteenth century. John Brewer has recently demonstrated the sheer extent of the armed forces which made geopolitical triumph possible.32 A crucial enabling factor in this connection was the financial revolution of the late seventeenth century, that is, the founding of the Bank of England


12

and the related rise to pre-eminence of the City of London. This in turn is explicable only because of the absence of absolutism; as Tocqueville realized, the upper classes cooperated with the state, particularly in matters of taxation, precisely because it felt able to control it—itself a consequence both of the centralized nature of English feudalism and of limited state competition consequent on isolation from the European landmass. It is noticeable that this entry onto the world stage predates the industrial revolution; equally striking is the fact that it was the result of well-developed militaristic skills. However, after the industrial revolution, Britain not unnaturally adopted a rather full-blooded version of liberalism: commercial specialization and free trade were emphasized, even though this meant removing the protected status of home agricultural producers. Trading with the world was initially conducted without the benefit of much formal empire: The middle class did not require diplomats to tout for trade for them, which would have been distasteful, or to go for war for trade, except in situations where such wars could be justified on higher and purer grounds. A young, vigorous, dynamic economy like theirs, without serious competitors in the world, could get along on its own, without help.33 This is to say that Victorian ideology, according to which the rule of the market had replaced the traditional conflict of states, was highly self-serving. Other states were well aware of this. They pointed out in particular that the supposed loss of geopolitical security consequent on the Repeal of the Corn Laws was in reality no such thing as long as Britain maintained a navy powerful enough to secure its sources of supply. Furthermore, Britain was perfectly capable of continuing to play at traditional balance of power politics when it proved necessary, as it did in the case of Egypt, even though middle class non-interventionist liberals might be unhappy with this. It seemed for some considerable time as if the British liberal dream of peace through interdependence, of the spread of trade and eventually of the parliamentary system, might be realizable. Throughout the 1850s and 1860s tariff walls in Europe fell, and it could be claimed that this aided the general economic advance of the time. But there was always a worm in the bud of this liberal dream. States interfered with the market in the most obvious way. Each state desired to have a set of productive industries that would ensure strategic security. These military origins to industrialization led to the creation of surplus capacity in the world economy, something which became obvious in the recession of the 1870s.34 At this moment, two possible strategies, neatly encapsulated in the title of Werner Sombart’s Handler and Helden, came to the fore.35 The trading policy was to adapt one’s national society to the logic of the world market, to allow productive activities to die out as they became unprofitable; the heroic alternative was to maintain such activities, not least because they played a vital part in securing national security.36 The close similarity of the interests of capitalists and those of foreign policy makers in Britain meant that its trading strategy, albeit with the protection of the Royal Navy, was bound to be maintained. In contrast, Imperial Germany moved towards the alternative pole.37 The state elite was responsible for this change. It drew upon its own militarist traditions, themselves based on Germany’s insecure geopolitical position in central Europe, in insisting that to allow the entry of cheap grain would have made Germany


13

dependent on the world market as long as access to that market was controlled by the British navy. Closely related to this was the fact that tariffs were an especially good source for raising revenue.38 In the longer run, the Junkers, as traders of grain historically in favour of free trade, came to support this policy as it protected them against cheap grain from North America; their views were much magnified by the over-representation they gained in the Prussian Estates. Similarly, protectionism became attractive to heavy industrialists, albeit much less so to the traders of Hamburg. Nonetheless, the prime move towards protectionism was made by Bismarck, largely in order to be able to finance the army whose victories had cemented German unity. The initial German move towards protectionism did not by itself lead to a visceral, noholds-barred conflict between Britain and Germany. As long as Germany’s expansionist aims were limited to mainland Europe, Britain would accede to them. In the last analysis Bismarck understood this. Perhaps too he understood the logic of the market. Germany’s first bid for colonies was very clearly an exercise in geopolitics rather than in foreign economic policy, and Bismarck had no trouble at all in withdrawing from this first bid— not least perhaps because his banker Bleichroder had very quickly realized that the rates of return from colonial possessions were miserably low.39 Nonetheless, German policy came to demand a place in the sun, even in the Congo, and a consequence of this was the 1897 decision to build a navy—the real origin of the AngloGerman antagonism. Such a policy was in the interest of particular capitalists. In addition, domestic reformist pressure, most notably that of the Social Democratic Party, supported an expansion of the navy rather than of the army, for fear that the latter could be used for internal repression. But the decision depended most of all on key state actors, most notably on Tirpitz whose brilliant propaganda campaign captured the ear of the Kaiser. In one sense perhaps, the decision needs no real explanation. It is normal for rising powers to exert their strength, and almost inevitable that challenges to Britain would be mounted once she had disrupted the balance of power by taking on a world role. Perhaps the fact that complex interpretations have been offered is the result of our knowledge that World War I was a catastrophe. Nonetheless, it is worth emphasizing that what matters about economics is often less economic reality per se than what is believed to be the facts of the matter: colonies might not have been immediately profitable but the rationale that they were necessary for long-term prosperity was widely accepted. It is noticeable in this context that capitalists tend not to produce their own geopolitical visions. In Imperial Germany many capitalists who did not stand to benefit from empire nonetheless acceded to the geo-economic vision produced by others. It would be unsatisfactory to leave matters at this point. Britain wished to appease Germany, and would have done so had that been possible. The problem was that no specific and negotiable demands were made by Imperial Germany, and this ruled out the sort of colonial settlements that had been achieved with France and Russia. The reason for this was simple. The German state was not truly modern; it was rather a court at which policy resulted from favouritism.40 The absence of a bureaucratic and rational state meant that Germany ended up without a properly worked out grand strategy. For a short period under Caprivi a genuine attempt was made to embrace a trading strategy. However, concentration tended to be on two heroic strategies. Weltpolitik did not mean that the traditional drive to the East was abandoned thus avoiding, through reconciliation


14

with Russia, a war on two fronts. Instead no clear decision was made, with the result that a system of alliances against Imperial Germany was formed. Fears of encirclement, together with increasing middle class nationalist and militarist pressure,41 meant that state leaders felt increasingly trapped; in the end they resorted to war in part from despair. The lack of a modern state apparatus was quite as much to blame for Germany’s drift to war as was authoritarianism per se. Some time has been spent understanding the German challenge because it is a key step in explaining British decline; later on, it will, by process of analogy, help us to understand American foreign policy. We can begin negatively by asking whether British decline resulted from the hegemonic services that it was providing for capitalist society as a whole. There is little truth to this. Britain was never a hegemon in the sense ascribed to that term by modern international relations theory. It had a short-lived primacy in the market, but scarcely ever enjoyed preeminence in state to state relations. Its decline cannot be ascribed to excessive defence expenditures for capitalist society as a whole since its main rivals had comparable costs by the beginning of the twentieth century.42 In a similar vein, it is probably mistaken to see the pre-World War I monetary system as genuinely hegemonic. The Germans and the French had their own monetary blocs, and the latter were able to invest in Russia in francs rather than in sterling. The crucial evidence for a monetary hegemony comparable to that of the United States in the 1960s when one power alone could increase money in circulation, thereby extracting seigniorage—would have been persistently large deficits of Britain with Paris, Berlin and New York, that is, evidence to show that Britain financed its deficit by making others hold sterling. But most scholars believe this not to have been the case: ‘Britain is said to have had sufficient income from trade, investment, and services, plus the Indian milk cow, to remain in balance with the other major centres’, whilst such sterling balances as were held resulted from economic calculation rather than hegemonic coercion.43 An assessment needing examination is that British industry was hurt by allowing continued access to its own markets whilst its own products were banned from much of the continent and from the United States—a claim which led Arthur Balfour to argue for what one theorist has termed ‘specific reciprocity’, that is, to force open protected markets by threatening to close off its own.44 It is hard to know how to weight this factor; against it can be set, in a moment, a different economic model. The more traditional, neutral factors of previous theories of decline seem to explain the British case with greater conviction. The single most important fact is simply that of exhaustion brought on by fighting Germany in two world wars. The financing of those wars led to the liquidation of most claims against the rest of the world. Much of this was, of course, hidden by the fact that Britain emerged on the winning side; nonetheless, fundamental weakness was absolutely apparent within months at the end of World War II.45 Furthermore, Britain had become exhausted as the result of state competition more generally. It was the challenge from other states, and particularly the fear that imperial rule by other powers might close off markets, that led to the acquisition of formal empire towards the end of the nineteenth century. This was a sign of weakness rather than of strength. It was always likely that a territorial empire would cost more than it was worth, and modern economic historians have shown this to be the case;46 the popularity of the empire and the need, given Britain’s dependence on imported food, to protect trade routes


15

nonetheless make the acquisition of territory comprehensible. As it was, resources and commitments were brought fairly closely into line before 1914, and it was this that enabled the first German challenge to be surmounted. In the inter-war years, however, the increasing strength of nationalism made it more difficult to concentrate attention on the challenges posed by Hitler.47 Economic decline is probably best explained in similarly neutral terms. British industry had been based on a very limited set of key technologies, and these had conquered the world market without the benefit of having been sharpened by competitive rivals. It was scarcely surprising that an institutional package gelled on the basis of its historical success. Even when Imperial Germany was moving towards a second industrial revolution, based on technical education and on chemicals, it remained possible for Britain’s older industries to remain in profit. Nonetheless, the British economy did not under-perform particularly badly before 1914, as its development of service and leisure industries clearly demonstrated.48 Britain’s initial economic decline was in largest part relative and inevitable. If one reviews British policy as a whole, what is most striking is how normal is decline; it does not require, as would a miracle of rejuvenation, much commentary. The uneven nature of capitalist development was bound to lead to a loss of economic preeminence. When this was allied to geopolitical challenges and thereby to increasing imperial costs, British power was always going to be unsustainable. But if Britain had very poor cards to play, could it have managed better with them? The answer to this for the most part must be negative. The challenge from Germany was, as noted, real, and there were genuine limits to how far Britain could go in appeasement even though allowing the rising power a major role was the only route to safety. I am no longer even sure that the British decision to continue fighting in 1917 was so irrational, even given the destruction of war: if Germany had then held on to the territory that it had conquered, as it demanded, a further dose of Weltpolitik might have proved even more fatal. More importantly, it is extremely unlikely that the radical modernizing strategy of Chamberlain, much favoured by some later critics, would have reversed decline if it had been instituted. The fundamental weakness of the policy was less working class hatred of expensive food than the reluctance of Canada and Australia to be forced to remain primary producers; these states wanted genuine rather than dependent development.49 Furthermore, imperial preference would certainly have intensified imperialist pressures elsewhere—pressures which were partly mollified historically by Britain allowing others to trade in her empire until the 1930s. All this can be summarized by saying that most British policy was determined by factors beyond national control, that placing foreign policy-making in different hands would have made very little difference. Nonetheless, it would be grossly mistaken to argue that every avenue of change was closed over nearly a century. It became particularly clear after 1945, for example, that British decline was no longer simply the effect of other states catching up as the transfer of people from agriculture to industry enhanced their growth rates: that did help account for the strengthening of the position of both France and Germany, but it failed to account for the continued low rates of British economic growth. Whilst it is certainly true that British decline was exacerbated by the character of market forces created by British history, that is, trade tended to be in low technology goods with the Commonwealth, largely to repay wartime loans, rather in more competitive and dynamic European


16

markets in the years immediately after 1945, the inability of the state to alter such structures and to modernize becomes ever more surprising.50 It thus becomes necessary to establish which actors were responsible either for policies which exacerbated decline or for blocking others which might have led to renewal. Some responsibility can be laid at the door of the people, In foreign policy, domestic pressure played some role in preventing Sir Edward Grey giving Imperial Germany the unambiguous warning that might have changed its perceptions.51 In general, however, foreign policy makers were relatively insulated from popular pressure, and their successes and failure were largely their own. A much more celebrated argument concerning the way in which popular pressure caused economic decline is, of course, seen in the claim that a militant working class prevented economic restructuring. This view has dominated recent British politics and much academic debate. It is largely wrong. The British working class has only rarely been politically militant, and such economic views as it has had—whether original or in favour of keeping the status quo— were largely the result rather than the cause of economic decline. This point can be put differently by saying simply that the British elite had very considerable autonomy from working class pressures; the failure to create an economically dynamic economy is accordingly to be laid at its door. Some shading needs to be given to the boldness of this picture. Union power was from the beginning deeply entrenched in the Labour Party, and sustained attempts at economic renewal from that quarter might have met with resistance. Furthermore, the British working class was always likely to prove a poor corporatist partner because its long history prior to industrialization meant that it—in distinction, say, to its Swedish counterpart—lacked centralized institutions and a vital interest in national economic affairs. But it is doubtful how much of an obstacle this would have proved to a determined modernizing elite. No such elite was present in the crucial period until the 1960s—by which time union militancy had increased in such a way as to block plans for reform. If attention is to be given to internal social blockages, to distributional coalitions standing in the way of societal flexibility, it makes much more sense, as many scholars now realize, to ask whether the policies of the British state were determined by the interests of the financial sector of capital.52 There can be no doubt of the increasingly deleterious impact of the unholy trinity of City-Treasury-Bank of England upon British industry. First, the great sophistication of the equity market has meant that profits have been made through trading in money, rather than through investing in domestic industry. In consequence, British industry has suffered from low levels of capital formation, a factor which goes much further in explaining low rates of worker productivity than de haut en has comments about the laziness of British workers; in addition, industrialists have constantly had to concern themselves with the provision of short-term profits so as to pay out dividends to shareholders—inattention to whose interests can easily lead to takeover bids. Secondly, the City has consistently argued in favour of high exchange rates. This lay behind the catastrophic return to gold in 1925, and the stop-go policies of the period from 1945–71; the growth of the Eurodollar market in London in more recent times has, if anything, enhanced the City’s power. The judgement to be made about this is simple: the single biggest obstacle to British economic recovery has probably been the excessive strength of the pound and of the volatility of interest rates necessary to ensure


17

that strength.53 Thirdly, the triumvirate of City, Bank and Treasury has consistently argued against the adoption of industrial policies, a factor which reflects the idiosyncratic, ‘budgetary’ nature of British state capacity. Indeed, the best way in which the spirit of this whole package of policies can be summarized is that of adherence to economic liberalism in a very full-blooded manner; it is this complete trust in market forces that characterizes Anglo-Saxon liberalism as a whole. Such policies were a source of strength when Britain had a strong industrial lead, but they led to less than optimal industrial performance when other nation states developed strong banking-industrial links and to genuine catastrophe when, between 1925 and 1931 and since 1945, they caused Britain’s exports to be priced out of the world market. One way in which the importance of the financial sector can be appreciated is by reference to the work of Rubinstein.54 His finding—that the very rich in Britain have been landowners and financiers rather than, at any time, industrialists —allows considerable scepticism to be cast on the thesis of British decline proposed by Martin Wiener in his influential English Culture and the Decline of the Industrial Spirit.55 Wiener suggested that the aristocratic embrace undermined bourgeois virtue, sending the sons of businessmen to semi-rural retreat in the spirit of William Morris and Laura Ashley. In fact, of course, the British elite had no aversion to money-making per se; to the contrary, it discovered the best avenue available. Nonetheless, we do need a theory which stresses the sleepiness of the British elite. There remains much to be said for the contention that the financial sector constrained rather than controlled the political elite.56 That elite sought to restore sterling’s international role for its own autonomous prestige reasons; thus we should not accept the view of Harold Wilson in the early sixties as a determined opponent brought to heel by finance capital—it makes no sense to say that he was defeated when he and his party had no real alternative strategy of their own. This conclusion can be generalized, and for the most banal of reasons. The reinvigoration of most advanced nation states within capitalist society since 1945 was the result of their elites being shocked by catastrophe, most notably that of defeat in war. Victory was taken as an imprimatur of the success of British institutions, something symbolized by the success of history and the scorn shown to sociology within that country. There was recognition that times had changed, but the strategy adopted was to continue things as close to normal as possible. This was particularly clearly expressed in Macmillan’s words to Crossman whilst attached to Eisenhower’s headquarters in Algiers in 1942: [We] are the Greeks in an American empire. You will find the Americans much as the Greeks found the Romans—great, big, vulgar, bustling people, more vigorous than we are, and also more idle, with more unspoiled virtues but also more corrupt. We must run this [HQ] as the Greek slaves ran the operations of the Emperor Claudius.57 Of course, this Polybian strategy of playing Greece to America’s Rome was then generalized58—curiously given that the fate of the Greeks inside the Roman Empire was by no means entirely pleasant. This policy has proved to be disastrous. If continual loyalty to finance rather than to industry is one side of the coin, the other is continued adherence to the tradition of national militarism. A full 50 per cent of Britain’s research


18

and development funding now goes towards weapons, an absolute madness given the short production runs involved. In this connection, it is worth saying something about the ‘Thatcherite Revolution’. There is no doubt about Margaret Thatcher’s desire to treat economic decline as the moral equivalent to defeat in war, and there is no doubt but that she has fundamentally changed British political discourse by making generally obvious what has anyway been true for sometime—that Britain is a small state whose future depends on her ability to survive in the international market. But the desire for reform has not been met by any comparable overhaul of institutions or policy. If the key institutions that gave Britain initial success were those of the City and of the military, they have by no means been dismantled: to the contrary, what is most striking is the extraordinary degree of basic continuity. This is more generally true. Adherence to economic liberalism has not been limited to the recognition that Adam Smith still rules externally; rather, disastrous attempts have been made to impose his marketist views internally. This continued total loyalty to laissez-faire has meant that little attempt has been made to create comparative advantage through the creation of a skilled workforce and by means of industrial policy. Britain has survived the Thatcherite years by means of North Sea oil whose revenues have masked the huge increase in expenditures on unemployment that would otherwise probably have caused electoral revolt. But the failure of the ‘Thatcherite Revolution’, that is, the likelihood that Britain will continue to do far less well than it might, is a different story. All-in-all the best summary judgement on British decline in general remains that of A.J.P.Taylor: ‘the English people of the twentieth century were a fine people and deserved better leaders than on the whole they got’.59 The United States: Down, but not Out Two tasks confront us in considering the position of the United States in the contemporary world. On the one hand, the lack of appropriateness of the analogy with British decline can be demonstrated by showing first that American decline has not gone very far, and second that the structure of the world polity is likely to limit the extent of that decline. But, on the other hand, we need to see if the processes explaining such decline as there has been are similar to those that affected Britain. Is American decline, such as it has been, the result of traditional factors, or does it rather result from the provision of services for capitalist society as a whole? One doubt is worth mentioning immediately. Theorists of decline occasionally give the impression that the United States once could do as it wished, but now is more or less impotent. This image of a golden age is much exaggerated. Difficulties with allies have plagued the United States throughout the postwar period, and it is not the case that American views always prevailed or that various allied contributions did not affect important outcomes. The United States sought, for example, to establish genuine multilateral liberal economic norms and it hoped too to avoid a continental commitment to Europe: in fact it ended up with what Ruggie has felicitously called ‘embedded liberalism’ and with NATO, whilst its allies, from the start, refused to share its perceptions of the Soviet Union and thereby to place security above trade.60 Nonetheless, the United States did and still does gain what it wants on crucial occasions, even if much


19

sound and fury hides this basic fact. This is not to deny that there has been some change, most dramatically with the closure of the gold window in 1971. But we are coming to realize that what has changed is the manner in which power is exercised—now far less benign, indeed predatory on the part of the hegemon61—rather than any absolute loss of power. Let us consider in turn the military, economic and monetary bases of continuing American strength. The military power of the United States is scarcely in question. It alone stands in rivalry to the Soviet Union, albeit its lead in nearly every key military technology makes it very much first amongst equals. But only these two states are genuine superpowers possessed of first- and second-strike nuclear capabilities. In itself, however, this does not convince the theorists of decline for the perfectly sensible reason that defence procurement in the end rests upon the capacity of an economy to support it. Can the American role be sustained now that the United States has moved from being $141 billion in credit to the rest of the world in 1980 to something like $500 billion in debt by 1988, with the clear likelihood that that debt will massively increase? It is necessary, however, to be rather sceptical of the various indices used to mark American decline. The figures of indebtedness to the rest of the world, for example, are exaggerated by the fact that the book value of American interests overseas is given at purchase price rather than current worth; such debt is, of course, much more the result of the policies of Ronald Reagan than of any long-term processes of secular decline. It should not be forgotten, in addition, that for much of the 1980s against the federal budget deficit should be set surpluses in key states such as California; once that has been done the total deficit stands at something like 2 per cent of GNP—not a strikingly high figure, and historically normal for this particular nation state. Further, one should not forget that debt is denominated in dollars; it is thus subject to diminution should the United States print money, and thereby inflate the dollar and the world’s economy. Equal care should be taken when dealing with protectionism. Hegemonic stability theory suggests that a decline in economic strength will lead to a demand for protectionism, and the very considerable upsurge in informal quotas of various sorts seems to suggest that this is true of the United States at this time. But protectionism is only half-hearted. It is important to note that the 1988 Trade Bill demanded, as have American negotiators over the last years, increasing openness in services, agriculture and shipping. It is equally important to be suspicious of what trade figures reveal given the huge increase in intra-firm trade—by now perhaps a third of all American trade. A more accurate index of American economic power is the share of world GNP controlled by American companies—and let it be remembered that much American investment has been of the multi-national rather than portfolio variety favoured by the British—and that this shows continued strength; more specifically, the loss of preeminence as debilitated economies recovered seems to have bottomed out in the early 1970s, with the United States holding more or less the same share over the last 15 years.62 This index is itself open to question, and Robert Reich has suggested that American companies’ share of world production means little if the international division of labour in such companies has the United States responsible only for invention and assembly, with profit-creating complex skills moving ever more towards East Asia.63 This picture might seem to be supported by the discovery that an increasing number of patents for manufacturing purposes are taken out in Japan; but this in turn is questioned


20

by those who argue that real industrial strength is best measured by software patents, vital for most advanced contemporary industrial products: these are still the preserve of the United States. It seems possible, moreover, that America is regaining a competitive edge in middle sized companies;64 the most recent study suggests that the United States continues to dominate in the newest areas of high technology goods.65 In general, there is little agreement as to the nature of economic power today; it is scarcely surprising, in consequence, that no index is accepted as its measure. A similarly mixed story can be told about the position of the United States within the world monetary system. At first sight the seigniorial privilege of the dollar standard, that is, the unilateral right to expand the money supply, has been removed with the floating exchange rate ‘system’ inaugurated in 1973. But the dollar remains the world’s key currency, and the United States seems to have retained enormous power within the world monetary system.66 Under the floating system, Germany ceased to be the key supporter of the dollar, but its place was taken first by Saudi Arabia and then by Japan. One of the services that a hegemonic power was supposed to provide, at least in Kindleberger’s eyes, was that of exporting capital to the rest of the world. In fact, world capital has flown to the United States, largely as the result of Reaganomics. There is certainly no denying that it was the policy of the United States which created a strong dollar in the early 1980s, and it is the policy of the United States since the Plaza Accords which have made for a weaker dollar: in both cases, the largest player determined the rules of the game, and it has proved impossible for smaller players to design alternative rules without the cooperation of that player. The impact of these policies on Latin America is such that its export of capital to Western banks is now running at somewhere between 4 and 5 per cent. This is the brutal exercise of great power—so great, indeed, as to make the Third World’s success in controlling its minerals and commodities pale into insignificance.67 Hence this is an appropriate point at which to recall that a leading power can ‘free ride’ on the system, that is, strength can be used to extract advantage. The present power of the United States can be summarized by saying that it stands at the top or close to the top on all ‘power indicators’. But the question of decline is not likely to be resolved by noting this since what some see as a glass half full will seem half empty to others. Let us turn instead to those features of the world polity which make the position of the contemporary United States unlike that of Britain at the end of the nineteenth century. An initial point worth emphasizing is that the United States has not suffered a massive defeat in war. This is not to deny the importance of Vietnam. But the United States was able to pass on part of the costs of the war to its allies by sending its inflation throughout capitalist society. But perhaps this consideration will not much sway the school of decline: might not Vietnam be the Boer War of the United States with worse to come? There are good reasons for believing this to be most unlikely. There is no equivalent to Imperial Germany facing the United States. Its geopolitical rival stands outside capitalist society, whilst its capitalist economic rivals are geopolitically dependent upon it. There is, in other words, none of that superimposition of conflicts which tends to so increase the intensity of conflict.68 Very importantly, both these sets of relations are essentially stable. Recent evidence does give the clear impression that the difficulties that face the Soviet Union—above all, slowing growth at a time when the dynamism of capitalist society is


21

ever more apparent—are far more serious than those which face the United States, a situation exactly opposite to that in which Britain found herself from the 1870s. The behaviour of socialist China since 1957 makes it now impossible to deny that the communist menace is a myth; the Soviet Union faces a war on two fronts, with a correspondingly enormous strengthening of the geopolitical position of the United States. The reform policies of both China and the Soviet Union seem set to make communism infinitely less threatening in any case. Moreover, there are well understood rules concerning spheres of influence between the United States and these major powers; the two superpowers, in particular, are ever more, to use Raymond Aron’s expression, ‘enemy partners’, perhaps most obviously in wishing to prevent the proliferation of nuclear weapons.69 The situation that faces the United States within capitalist society is quite as stable, and it is so in a manner that is historically novel. Japan and Germany were reconstructed as the result of American geopolitical victory: both are secure democracies wedded to trading rather than heroic strategies. It is extremely unlikely that either Germany or Japan will mount a challenge to the American system. In addition to the particular character of these states, there are good general reasons for believing that the world economy will not return completely to a no-holdsbarred conflict between trading blocs—albeit, there may well be some increase of this type.70 The fundamental justification for this belief is simply that the speed of technological change makes it ever more catastrophic to withdraw from the world market, something realized by most state leaders and enshrined by them and their leading industries in the sudden spread of joint ventures of very varied sorts.71 In addition, there is now some awareness that traditional protectionist policies are no longer likely to work: how can one protect one’s industries against, say, Japan, if that country chooses to assemble Thai and Taiwanese parts in South Korea and then to import them into the United States via Mexico? In the case of the United States, the impact of free trade institutions and mentalities, both internally and externally, is likely to make it particularly difficult to adopt any pure protectionist stance.72 In addition, new political groupings demanding a retention of liberal multilateralism are springing up to counter others seeking protectionism.73 Finally, it is very clear that the post-war settlement is popular amongst many states, and that they will go to considerable extents to preserve it: that settlement solved the ‘German problem’ and the Japanese problem’, whilst there is widespread awareness that bipolarity in a nuclear age has a great deal to recommend it. This general situation gives the United States certain clear advantages. The weakness of its geopolitical rival and the absence of challenges from within capitalist society mean that it does not have to acquire formal territorial empires. Of course, Vietnam represented precisely such a formal commitment. But what may prove to be important about that debacle is that it will not be repeated; certainly the United States has an alternative available, as argued in detail below. But its situation as a whole is completely different: after all, the economic challenge of Germany and Japan is that of allied states. And there are additional reasons that make it probable that such states will not replace their trading strategy with a new autarchic and heroic alternative. Britain’s move to territorial empire was made virtually inevitable by the extent to which it traded with the world, and it was scarcely surprising that this led other states to fear they might in the long run be excluded; in contrast, the United States trades far less, and therefore has less need to


22

secure markets through territorial possession—something which may mean that the development of capitalist society is not again so dramatically interrupted by the logic of geopolitics. Crucially, however, the United States has very considerable leverage over its allied economic rivals, both because it provides their defence and because their economic success is partly dependent on the sheer size of the American market—which the United States could, unlike most other states, close given the huge resource base its continental status affords it. At times, seigniorage has been obvious, notably in the passing on of inflation under the Bretton Woods system; but it is as present today, from the necessity of supporting the dollar to Japan’s enforced abandonment of its plans to build fighter planes, a geopolitical form of industrial subsidy. The situation of the United States differs from that of Britain in a further, and absolutely fundamental way. Paul Kennedy is right to argue that Britain’s decline had to be great since it had so massively overreached its natural power ranking, that is, the portfolio of demographic, geopolitical and natural resources.74 Nonetheless, there certainly has been relative decline, and it is time to turn to assessing its causes. Let us begin by asking whether we should accept the claims of hegemonic stability theory in this case. Has the United States become exhausted because of the burden of defence it has provided for capitalist society as a whole and by its obeisance to liberal multilateralism in the face of formal and informal protectionisms elsewhere? There is some truth to both these claims, but each has recently been subject to much exaggeration. Figures that indicate that the United States’ economy is being undermined by high defence expenditure, both absolutely and in comparison to her allies, need to be treated with the utmost care.75 Most obviously, defence expenditure is not now particularly high by historical standards, and it is hard to credit it with causing economic decline given the economic success of South Korea and Sweden, both of which pay as large a share of GNP for defence as does the United States.76 Equally, it should not too easily be believed that defence spending is bad for the American economy. The trade-off between defence spending and economic performance is extremely complex,77 but it is clear that the United States, with long production runs and research at the frontiers of technology, has gained something; certainly the Japanese regard American defence spending—especially when overseen by the Defence Advanced Research Projects Agency—as more or less equivalent to an industrial policy.78 Recent studies suggest that concentration on defence has not had deleterious effects on domestic industrial capital formation;79 the National Research Council has added to this the discovery that there is sufficient engineering talent left after concentration on defence for the health of domestic productive industry.80 Figures for allied defence expenditures tend to be highly distorted:81 they do not include ‘offset payments’, nor do they allow for hidden allied costs such as those of conscription and the provision of physical assets.82 A proper accounting would suggest that the major members of NATO—and France—pay nearly as large a share of GNP for defence as does the United States for its worldwide interests; Japan, of course, pays very significantly less. In the European case, some of the difference is explicable by the fact that Europeans prefer a defensive strategy which happens to be cheaper, in distinction to the expensive offensive strategy not unnaturally favoured by the United States. Importantly, all figures do not include the informal economic privileges that accrue, as argued, to the United States as military rent and as


23

bribes to ensure that its markets remain open. It is this which leads the advanced states to prop up the dollar and, in Japan’s case, to massively increase its aid budget rather than to establish its own aerospace industry; in these ways, the allies informally pay for the cost of defence. Equally sceptical points must be made about the claim that decline results from the protectionism of others. German economic success at the end of the nineteenth century may have been helped by tariff walls, but it was not fundamentally ascribable to that; the same seems true of Japan today. Bergsten and Cline argue that if Japan had no import barriers at all, America’s 1985 near $50 billion trade deficit with Japan would only have been reduced by $5 to $8 billion, and that $5 billion would have been added to the deficit if the United States had removed its own considerable barriers to Japanese imports.83 Although they are hard to prove in any decisive manner, traditional theories of decline seem to offer a more plausible account of what is happening to the United States. Just as Britain insitutionalized its moment of economic success, so too America allowed a set of institutions to gel around those Fordist politics of productivity that came to the fore under Roosevelt.84 Such industrial giantism seems less adapted to the flexible trading system of the contemporary world economy, partly because of the failure of the United States to provide the sort of social infrastructure that underlies it.85 American failure to adapt partly reflects its sheer size; it has often chosen to use its power to change the rules of international norms rather than to make its society flexible enough to compete—an option which now seems to have ever diminishing returns. Equally important, however, may well be the importance of finance capital within American society; if this sector has not yet reached the historic importance it gained in British history, the ways in which finance is currently favoured over industry, as in the fact that ‘junk bonds’ are tax deductible, offer obvious resonances. The whole point at issue can be summarized by saying that one reason for distrusting the ‘decline by service provision’ thesis of hegemonic stability theory is that it fails to pay proper attention to the inventiveness, diligence and adaptability of modern trading states.86 Imitating such virtues would require lowering the extraordinarily high levels of GNP given to consumption by means of increased taxation, less perhaps for direct industrial policy and more for the creation of suitable social infrastructures. If such policies are resisted because they are held to go against ‘the American grain’, this can only be ascribed to the particular character of Anglo-Saxon liberalism. As it happens, however, there are good reasons to believe that increased taxes, if mandated for educational or infrastructural renewal, might well be granted. Certainly individual states in America have been able to convince their citizenry of the need for greater taxation for industrial regeneration.87 There is, in other words, room for responsible national leadership to make a difference. It is equally noticeable that the American economy has suffered from geopolitical over-extension, as seen most clearly in Vietnam. Involvement in Vietnam did not seem to have any immediate economic rationale to it, in terms of markets, investment or raw materials, and this has led Krasner (1978) to stress that this policy was the result of anticommunism—an acte gratuite whose extraordinary craziness will become more apparent in a moment.88 There is certainly some truth to the claim that the ‘best and the brightest’, brought to Washington by Kennedy, suffered markedly from the arrogance of power, and should be held responsible for their actions. Nonetheless, Domhoff (1987) is surely right


24

to argue that these advisers were but lesser men carrying out the implications of the grand strategy that America created between 1941 and 1950;89 the responsibility of the advisers of the 1960s was—as had been the case with those who followed Bismarck—not to think their times for themselves. Still, it behoves us to try and understand the sources of postwar American grand strategy, not least so that the previous claim, that the United States has non-territorial possibilities closed to Britain, may be justified. Discussion of the origins of American postwar strategy has been exceedingly lively for several years, and it is not possible here to give anything like a summary of recent research.90 Nonetheless, it now does seem apparent that no account will be satisfactory unless it recognizes the autonomous impact of four factors. First, the revisionist historians and their followers are surely quite correct to stress that the American state, due to its liberal character and its historic lack of geopolitical involvement, was especially permeable—largely through the Council for Foreign Relations—to the wishes and demands of its domestic capitalists.91 It was at this period, for example, that involvement in Vietnam became likely as Southeast Asia became defined as part of ‘the national interest’. But the revisionist account is incomplete. In particular, secondly, we must note that many state leaders had, from the turn of the century, geopolitical visions of their own, a remarkable number of which were formed in the surprisingly Kiplingesque surroundings of Groton.92 This elite enjoyed the power it discovered during the war, and embraced empire willingly. Of course, there was a considerable overlap between the first and second sets of actors; this was scarcely surprising since the latter saw multilateralism as a means to ensure peace. Nonetheless, if the statements of the political elite, public and private, are to be believed, they were far more worried by questions of security than by the needs of the American economy, whether seen from their own point of view or as interpreted for them by capitalists and their experts. It was traditional balance of power reasons that underlay the Truman administration’s decisions to allow the multilateral norms they preferred to be diluted and to accept continental involvement in NATO. A third set of actors was not American at all. The collapse of Britain joined with the vigorous if defensively inclined security demands of Stalin meant that many Europeans, most notably Bevin, actively sought an American presence.93 This was an ‘empire by invitation’, and some part of its dynamic came from allied actions.94 Finally, the actual character of the grand strategy was markedly influenced by the nature of American institutions and experience. The American people and in particular, voters with ethnic ancestries in Ireland and Germany had long been suspicious of foreign entanglements, and it did not prove easy to gain support for a global policy.95 The Truman administration in 1946 was faced with an ebullient Republican Congress which was at once anticommunist and keen to balance the budget. It proved possible to turn highly anticommunist Republicans such as Vandenberg, who himself faced re-election from mid-Western Polish-American voters, in an internationalist direction, and to split them from that fiscally cautious mainstream headed by Taft which remained suspicious of foreign involvement. There is as yet no agreement amongst historians as to the exact input of public opinion on policy formation at this time, but there can be no doubt but that there was some. It would in principle have been possible to strike a simple Spheres of influence’ deal with the Soviet Union, as Kennan argued; that this did not happen was the result of the American people, and of many of their leaders, seeing foreign affairs in


25

rather moralistic terms—involvement could be by crusade alone.96 This all-or-nothing approach finally became cemented by the Korean War which seemed to justify the charges of the ‘loss of China’ made by McCarthy. The fear of electoral retribution, probably much exaggerated in point of fact, made politicians rather reluctant to see the world in other than bipolar terms. This led to over-extension—not in Europe, it should be noted, but in the Third World where bipolar vision prevents proper appreciation of local forces and where there are few strategic interests of any significance.97 The United States has an alternative grand strategy that it can adopt. One key element of such a strategy is that of learning, in contradistinction to American political experience, that states in the Third World need to be strengthened for developmental purposes, and that this does not represent an unadulterated attack on liberal principles, especially since the spread of nationalism is likely, as Kennan realized in the 1940s, to dilute the cohesion of the communist movement.98 The second key element of an alternative American grand strategy is for the United States to realize the extent of its extraordinary power. It really does not matter if a state withdraws from the world market; this is likely to be temporary since the costs of such withdrawal, given the fact that it is capitalist and not socialist society which has abundant capital and significant markets, exact such an incredibly high price. One of the most interesting questions in world politics today is whether modifications to postwar grand strategy can be made which will recognize regional dynamics and ‘take class out of geopolitics’. It seems likely that capitalists can be convinced, given the obvious increase in market power even with the loss of Vietnam; capitalists will not stand in the way of such a policy. Similarly, a determined political elite is likely to be able to achieve such a policy despite fear of electoral punishment. It is important to have international understanding so that no shocks will create such strong domestic pressure, as was the case at the end of President Carter’s term of office when the Soviet Union invaded Afghanistan. However, domestic pressures can all too easily be exaggerated. Kennedy was able to ‘lose’ Laos, and Carter Zimbabwe, without such punishment, largely because the future of these states was not defined in advance as something that would adversely affect American prestige. The same could have happened in Vietnam, perhaps as late as 1965; that it did not necessitates blaming the political elite of the time rather than the American people. There is room, in other words, for responsible leadership.

Conclusion The argument that has been made deserves summary. Suspicion has been shown to the claim of hegemonic stability that the decline of Britain and the United States results from their bearing the burden of services for capitalist society. In general, the decline of hegemons—of which the United States is in fact the only genuine representative—is best understood in traditional and neutral terms, that is, as the result of geopolitical overextension and the inability to overcome social blockages. In particular, both Britain and the United States adhere to a full-blooded marketist ideology—seen most clearly in the freedom given to financial sectors and in the absence of industrial policies of varied types—that makes it hard to adapt within capitalist society. This more particular factor


26

has been given separate analytic consideration because there is no reason in the abstract to believe that a hegemonic leader has to embrace every idea and institution beloved by Anglo-Saxon liberalism. Two implications of the argument are worth highlighting. First, the theoretical presuppositions at work are largely realist ones, that is, it sees decline as the result of a hegemon’s own mistakes. This argument is perfectly compatible with the other elements of hegemonic stability theory; indeed the most distinguished theorist of that school oscillates between the sorts of arguments I have made and the false self-liquidating theory of hegemonic rule.99 Secondly, a distinctive policy implication follows from the argument. As the external burdens of hegemony are not the fundamental reasons for the relative decline of the United States, policy drives emphasizing burden-sharing are effectively a distraction of time and energy. Renewal is in the hands of the United States alone. Final thoughts can now be offered as to whether the United States will decline further in a manner akin to that of Britain. A central tenet of the paper has been that the United States is in a much more advantageous position than was Britain. The uniqueness of its position as defender of capitalist society makes it likely that for some considerable and unspecifiable time the United States will be able to shore up its position by extracting seigniorage of various kinds from its allies. Nonetheless, some social processes reminiscent of Britain seem to be at work in the United States; this suggests that decline may yet go much further—albeit, given differential size, the United States is most unlikely to cease to be the key player in the modern world political economy. Despite the fact, for example, that American industrialists feature as heroes of a popular culture generally suspicious of Wall Street, the increasing importance of financial capital to the American economy is not in question; importantly, the sophistication of the equity market, not least when financed by junk bonds, places a disastrous premium on shortterm returns only too familiar from a reading of British economic history. A shared Anglo-Saxon economic liberalism also seems present in the lack of industrial policy and of attempts to manufacture comparative advantage—although something like this may be happening at the state level in America today. If all this is similar, there is one place at which the United States seems almost worse off than Britain. American politics are not at present coping creatively with the problems discussed in this paper; it is proving especially difficult to contain or reverse decline by the means indicated—that is, by raising taxes to enhance competitiveness and by adopting the alternate geopolitical strategy. These policies have forceful logics to recommend them in the abstract; at present they do not seem politically feasible. Why is this? Some authors suggest that changes in economic organization are responsible: as banks and industry become internationalized, they increasingly favour free trade at the expense of domestic renewal.100 There may be some truth to this, as there is to the argument that the greater internationalization of business diminishes domestic manufacturing employment in a way that affects the American electoral system (Harrison and Bluestone, 1988).101 But the argument is by no means completely convincing: it falsely suggests a degree of coordination between business interests and foreign geopolitical and geoeconomic policy for which there is, at least as yet, little evidence. Other authors have suggested that electoral pressures, particularly in a political system in which power is


27

divided between the executive and legislative branches, make it hard to remain flexible, especially in foreign affairs. There may be some truth to this point as well, but it should not be accepted too easily. There are opportunity/costs to democracy in all areas: there is a danger of moralistic swings of mood replacing calculation but the possibility remains that democratic review may enhance the policy process. If I were to risk a judgement on the costs and benefits of democracy to the conduct of American foreign affairs, it would be to argue that the most serious mistakes of what has been all-in-all an amazingly successful post-war foreign policy are attributable to autonomous elite actions, some of which, notably those concerning the Reagan administration’s actions towards Iran, would have benefited greatly from more open democratic scrutiny. Democracy is a resource quite as much as a stumbling block, at least for an intelligent and determined elite. This suggests a final factor whose impact is improperly understood but of undoubted importance. During the Second World War something like a political class was created in the United States. If this class was at times constrained by the popular passions it had itself in pan aroused and by the institutional surroundings in which it had to work, its fundamental unity meant that it could, when it acted skilfully, have its way most of the time; a determined political elite could, I believe, still succeed in most of the tasks it set itself. But the unity of that class has been ruptured—by differential responses to the student movement and to Vietnam and as the result of divided loyalties over Israel. Much could be done in the United States with responsible leadership. The fragmentation of the political elite at present suggests, however, that American decline may not by any means as yet have bottomed out.

Notes 1. 2.

P.Kennedy, The Rise and Fall of the Great Powers (New York: Random House, 1987). B.Russett, ‘The Mysterious Case of Vanishing Hegemony, or, Is Mark Twain Really Dead?’, International Organisation, vol. 39 (1985); S.Strange, Casino Capitalism (Oxford: Basil Blackwell, 1986); S.Strange, ‘The Persistent Myth of Lost Hegemony’, International Organisation, vol. 41 (1987). 3. P.Schmeisser, ‘Is America in Decline?’, New York Times, 17 April 1988. 4. J.M.Mann, States, War and Capitalism (Oxford: Basil Blackwell, 1988). 5. Cf. M.Olson, The Rise and Decline of Nations (New Haven, Connecticut: Yale University Press, 1982). 6. R.Gilpin, ‘Economic Interdependence and National Security in Historical Perspective’, in K.Knorr and F.Trager, eds, Economic Issues and National Security (Lawrence, Kansas: Regents Press, 1977); R.Gilpin, War and Change in World Politics (New York: Cambridge University Press, 1981); R.Gilpin, The Political Economy of International Relations (Princeton, New Jersey: Princeton University Press, 1987). 7. C.Kindleberger, The World in Depression (Berkeley, California: University of California Press, 1973). 8. R.Keohane, After Hegemony (Princeton, New Jersey: Princeton University Press, 1984). 9. G.Liska, ‘Empire by Invitation?’, Journal of Peace Research, vol. 23 (1968). 10. D.Calleo, Beyond Hegemony (New York: Basic Books, 1987). 11. S.Krasner, Asymmetries in Japanese-American Trade (Berkeley, California: Institute of


28

International Studies, 1987). 12. M.Olson and R.Zeckhauser, ‘An Economic Theory of Alliances’, Review of Economics and Statistics, vol. 46 (1968). 13. W.Grieder, ‘Why Can’t Our Allies Defend Themselves?’, Rolling Stone, 16 June 1988. 14. My attention was drawn to this reference by my friend David Spiro. 15. W.M.Earle, ‘Adam Smith, Alexander Hamilton, Friedrich List: The Economic Foundations of Military Power’, in W.M.Earle, ed., Masters of Modern Strategic Military Thought from Machiavelli to Hitler (Princeton, New Jersey: Princeton University Press, 1943). 16. J.M.Mann, Sources of Social Power, vol. 1, From the Beginning to 1760 AD (New York: Cambridge University Press, 1986). 17. R.Collins, Weberian Sociological Theory (New York: Cambridge University Press, 1986). 18. I.Hont, ‘The “rich country-poor country” debate in classical Scottish political economy’, in I.Hont and M.Ignatieff, eds, Wealth and Virtue (Cambridge: Cambridge University Press, 1983). 19. K.Polanyi, The Great Transformation (Boston, Massachusetts: Beacon Press, 1944). 20. A.Gershenkron, Economic Backwardness in Historical Perspective (Cambridge, Massachusetts: Harvard University Press, 1962). 21. R.Robinson and J.Gallagher, ‘The Imperialism of Free Trade’, EcHR, vol. 6 (1953). Cf.B.Porter, Britain, Europe and the World, 1850–1982 (London: George Allen and Unwin, 1983); M.Doyle, Empires (Ithaca, New York: Cornell University Press, 1986). 22. H.Morgenthau, Politics Among Nations (New York: Alfred Knopf, 1973). 23. R.Aron, Peace and War (London: Weidenfeld and Nicolson, 1966). 24. D.Smith, ‘Domination and Containment’, Comparative Studies in Society and History, vol. 19 (1977). 25. J.A.Hall, Liberalism (London: Paladin, 1988). 26. A.de Tocqueville, Democracy in America, vol. 2 (New York: Vintage Books, 1945). 27. G.Kennan, American Diplomacy (Chicago, Illinois: Chicago University Press, 1951). 28. H.Kissinger, White House Years (Boston, Massachusetts: Little, Brown and Company, 1979); H.Kissinger, Years of Upheaval (Boston, Massachusetts: Little, Brown and Company, 1982). 29. I.Kant, ‘Perpetual Peace’ (1795), in C.J.Friedrich, ed., The Philosophy of Kant (New York: Modern Library, 1949). 30. M.Doyle, ‘Kant, Liberal Legacies and Foreign Affairs’, Philosophy and Public Affairs, vol. 12, nos 3 and 4 (1983). 31. Cf. G.Ingham, Capitalism Divided (London: Macmillan, 1984), A.Gamble, Britain in Decline, 2nd edn (London: Macmillan, 1985); P.Anderson, ‘The Figures of Descent’, New Left Review, no. 161 (1987); Mann, States, War and Capitalism; K.Middlemas, Politics in Industrial Society (London: André Deutsch 1979). 32. J.Brewer, The Sinews of Power (London: Hutchinson, 1989). 33. Porter, Britain, Europe and the World, p. 16. 34. G.Sen, The Military Origins of Industrialisation and International Trade Rivalry (London: Frances Pinter, 1984). 35. W.Sombart, Handler und Helden (Leipzig, Dunckler & Humblot, 1915). 36. Polanyi, The Great Transformation; P.Gourevitch, Politics in Hard Times (Ithaca, New York: Cornell University Press, 1986). 37. P.Kennedy, The Rise of the Anglo-German Antagonism, 1860–1914 (London: George Allen and Unwin, 1980); H.U.Wehler, The German Empire 1871–1918 (Leamington Spa: Berg, 1986).


29

38. J.Hobson, ‘A Tariff History of Germany, 1818–1914’ (unpublished paper). 39. F.Stern, Gold and Iron (New York: Vintage Books, 1986). Cf. L.Davis and R. Huttenback, Mammon and the Pursuit of Empire (Cambridge: Cambridge University Press, 1987). 40. I.Hull, The Entourage of Kaiser Wilhelm II, 1888–1918 (New York: Cambridge University Press, 1982). 41. G.Eley, Reshaping the German Right (New Haven, Connecticut: Yale University Press, 1980). 42. John Hobson and I are presently collecting comparative information on state revenue and expenditure figures. Preliminary results lead me to disagree with Patrick O’Brien, ‘The Imperial Component in the Decline of the British Economy before 1914,’ in M.Mann, ed., The Rise and Decline of the Nation State (Oxford: Basil Blackwell, 1990). 43. D.Calleo, ‘The Historiography of the Interwar Period: Reconsiderations’, in B. Rowland, ed., Balance of Power or Hegemony: The Inter-War Monetary System (New York: New York University Press, 1975), p. 241. Cf. P.Lindert, Key Currencies and Gold, 1900–1913 (Princeton, New Jersey: Princeton University Press, 1969). 44. A.Friedberg, The Weary Titan (Princeton, New Jersey: Princeton University Press, 1988), chapter 2; R.Keohane, ‘Reciprocity in International Relations’, International Organisation, vol. 40 (1986). 45. R.Gardner, Sterling-Dollar Diplomacy (London: Macmillan, 1969). 46. Davis and Huttenback, Mammon; O’Brien, ‘Costs and Benefits’. 47. Kennedy, The Rise and Fall, chapter 6. 48. S.Pollard, Britain’s Prime and Britain’s Decline (London: Edward Arnold, 1989). 49. B.Semmel, Imperialism and Social Reform (London: Routledge and Kegan Paul, 1960). 50. A.Milward, The Reconstruction of Western Europe, 1945–51 (London: Methuen, 1984); P.Hall, Governing the Economy (New York: Oxford University Press, 1986). 51. C.Nicolson, ‘Edwardian England and the Coming of the First World War’, in A. O’Day, ed., The Edwardian Age: Conflict Stability, 1900–1914 (London: Macmillan, 1979). 52. Ingham, Capitalism Divided; Anderson, ‘Figures of Descent’; S.Newton and D. Porter, Modernization Frustrated (London: Unwin Hyman, 1988); Mann, States, War and Capitalism. 53. S.Strange, Sterling and British Policy (Oxford: Oxford University Press, 1971). 54. W.D.Rubinstein, Men of Wealth (London: Croom Helm, 1981). 55. M.Wiener, English Culture and the Decline of the Industrial Spirit (Cambridge: Cambridge University Press 1981). 56. S.Blank, ‘Britain: The Politics of Foreign Economic Policy, the Domestic Economy, and the Problem of Pluralistic Stagnation’, in P.Katzenstein, ed., Between Power and Plenty (Madison, Wisconsin: University of Wisconsin Press, 1978). 57. A.Home, Harold Macmillan. Volume One: 1894–1956 (New York: Viking, 1988), p. 160. 58. H.B.Ryan, The Vision of Anglo-America (Cambridge: Cambridge University Press, 1987); C.Maier, In Search of Stability (Cambridge: Cambridge University Press, 1988). 59. A.J.P.Taylor, ‘Accident Prone, or, What Happened Next’, Journal of Modern History, vol. 49 (1977), p. 18. 60. J.G.Ruggie, ‘International Regimes, Transactions and Change: Embedded Liberalism in the Postwar Economic Order’, International Organisation, vol. 36 (1982); Maier, In Search of Stability, M.Mastanduno, ‘Trade as a Strategic Weapon: American and Alliance Export Control Policy in the Early Postwar Period’, International Organisation, vol. 42 (1988); Gardner, Sterling-Dollar Diplomacy, A.Van Doermal, Bretton Woods (London: Macmillan, 1978).


30

61. Cf. J.Conybeare, Trade Wars (New York: Basic Books, 1987). 62. Strange, ‘The Persistent Myth’; R.Gilpin, US Power and the Multinational Corporation (New York: Basic Books, 1975). 63. R.Reich, Tales of a New America (New York: Vintage Books, 1988). 64. ‘A Portrait of America’s New Competitiveness’, The Economist, 4 June 1988. 65. R.McCulloch, The Challenge to US Leadership in High Technology Industries (Can the US Maintain Its Lead? Should it Try?) (Cambridge, Massachusetts: National Bureau of Economic Research, Working Paper 2513, 1988). 66. S.Strange, ‘Still an Extraordinary Power: America’s Role in a Global Monetary System’, in E.E.Lombra and W.E.Witte, eds, Political Economy of International and Domestic Monetary Relations (Ames, Iowa: University of Iowa Press, 1982); Strange, Casino Capitalism. 67. S.Krasner, Structural Conflict (Berkeley, California: University of California Press, 1984). 68. R.Dahrendorf, Class and Class Conflict in Industrial Society (Stanford, California: Stanford University Press, 1959). 69. Aron, Peace and War, part 4. 70. Gilpin, Political Economy of International Relations. 71. Cf. R.Rosecrance, The Rise of the Trading State (New York: Basic Books, 1986). 72. J.Goldstein, ‘Ideas, Institutions and Trade Policy’, International Organisation, vol. 42 (1988). 73. I.M.Destler and J.Odell, Anti Protection: Changing Forces in United States Politics (Washington, D.C.: Institute for International Economics, 1987); H.Milner, ‘Resisting the Protectionist Impulse: Industry and the Making of Trade Policy in France and the United States during the 1970s’, International Organisation, vol. 41 (1987). 74. Kennedy, The Rise and Fall, chapter 8. 75. I shall address this question more fully on the completion of my collection of US state expenditure figures. 76. Cf. J.Nye, ‘America’s Decline: A Myth’, New York Times, 10 April 1988. 77. S.Chan, The Impact of Defense Spending on Economic Performance: A Survey of Evidence and Problems’, Orbis, vol. 29 (1985). 78. Reich, Tales of a New America. 79. D.Greenwood, ‘Note on the Impact of Military Expenditure on Economic Growth and Performance’, in C.Schmidt, ed., The Economics of Military Expenditures (New York: St Martin’s Press, 1987); K.Rasler and W.R.Thompson, ‘Defense Burdens, Capital Formation and Economic Growth’, Journal of Conflict Resolution, vol. 32 (1988). 80. National Research Council, The Impact of Defense Spending on Nondefense Engineering Labor Markets (Washington, DC: National Academy Press, 1986). 81. K.Knorr, ‘Burden Sharing in NATO’, Orbis, vol. 29 (1985); D.Wightman, ‘United States Balance of Payments Policies in the 1960s: Financing American Forces in Germany and the Trilateral Negotiations of 1966–7’ (unpublished paper); K.Dunn, ‘NATO’s Enduring Value’, Foreign Policy 71 (1988). 82. G.Treverton, The ‘Dollar Drain’ and US Forces in Germany (Ohio: Ohio University Press, 1978); Knorr, ‘Burden Sharing’. 83. F.Bergsten and W.Cline, The United States-Japan Economic Problem (Washington, DC: Institute for International Economics, 1987). 84. Maier, In Search of Stability. 85. M.Piore and C.Sabel, The Second Industrial Divide (New York: Basic Books, 1984); Hall, Liberalism, chapter 7; L.Weiss, Creating Capitalism (Oxford: Basil Blackwell, 1988).


31

86. R.Dore, Flexible Rigidities (London: Athlone Press, 1986); R.Dore, Taking Japan Seriously (London: Athlone Press, 1987); D.Okimoto, Between MITI and the Market (Stanford, California: Stanford University Press, 1988); Weiss, Creating Capitalism; Piore and Sabel, The Second Industrial Divide; C.C.Deyo, ed., The Political Economy of the New Asian Industrialism (Ithaca, New York: Cornell University Press, 1987). 87. E.Vogel, Comeback (New York: Touchstone, 1985), chapter 10. 88. S.Krasner, Defending the National Interest (Princeton, New Jersey: Princeton University Press, 1978). 89. W.Domhoff, ‘Defining the National Interest, 1940–42: A Critique of Krasner’s Theory of American State Autonomy’ (unpublished paper, 1987). 90. J.L.Gaddis, The Emerging Post-Revisionist Synthesis’, Diplomatic History, vol. 7 (1983); The Long Peace (New York: Oxford University Press, 1987). 91. G.Kolko, The Politics of War (New York: Random House, 1969); J.Frieden, ‘Sectoral Conflict and US Foreign Economic Policy, 1914–40’, International Organisation, vol. 41 (1988); W.Domhoff, The Powers That Be (New York: Vintage Books, 1979); ‘Defining the National Interest’; The Ruling Class Does Rule: State Autonomy Theory and the Origins of the International Monetary Fund’ (unpublished paper, 1988). 92. H.K.Beale, Theodore Roosevelt and the Rise of America to World Power (New York: Collier Books, 1967); J.L.Gaddis, The United States and the Origins of the Cold War, 1941–47 (New York: Columbia University Press, 1972); J.L.Gaddis, Strategies of Containment (New York: Oxford University Press, 1982); W.Widenor, Henry Cabot Lodge and the Search for an American Foreign Policy (Berkeley, California: University of California Press, 1980). 93. V.Mastny, Russia’s Road to the Cold War (New York: Columbia University Press, 1979). 94. G.Lundestad, ‘Empire by Invitation?’, Journal of Peace Research, vol. 23 (1986). 95. G.Gilbert, To the Farewell Address: Ideas of Early American Foreign Policy (Princeton, New Jersey: Princeton University Press, 1961); Gaddis, Origins of the Cold War. 96. F.Klingberg, ‘Cyclical Trends in American Foreign Policy Moods and Their Policy Implications’, in C.W.Kegley and P.McGowan, eds, Challenges to America: United States Foreign Policy in the 1980s (Beverly Hills, California: Sage, 1979); S. Hoffmann, Gulliver’s Troubles, or the Setting of American Foreign Policy (New York: McGraw Hill, 1968). 97. S.Van Evera, American Strategic Interests: Why Europe Matters, Why the Third World Doesn’t’. Testimony prepared for hearings before the Panel on Defense Burdensharing, Committee on Armed Services, US House of Representatives, 2 March 1988. 98. S.Hoffmann, Duties Beyond Borders (Syracuse, New York: Syracuse University Press, 1981); Gaddis, Strategies of Containment; Hall, Liberalism, chapter 7; Van Evera, ‘American Strategic Interests’. 99. Gilpin, War and Change; Gilpin, Political Economy of International Relations. 100. Frieden, ‘Sectoral Conflict’. 101. B.Harrison and B.Bluestone, The Great U-Turn (New York: Basic Books, 1988).


PART FOUR: States and Political Behaviour


Commentary The great hold that liberalism has exercised over social science thought has been especially clearly seen in the fact that studies of the state have tended to concentrate on the ways in which the organised interests of civil society mould and control political power. This part of the collection represents the powerful and beneficial paradigm shift which is allowing us to see the ways in which states structure social movements. The first section approaches this matter in the most general terms by looking in turn at industrial workers, social revolutionaries in the contemporary Third World, and a notably wide range of political parties within the historical record. One common theme that turns through the articles in this section is the extent of the resources of the political élite: its capacity to communicate laterally and to make the most of the opportunities of bureaucratization gives it the leading edge in power terms. In consequence, it takes a very great deal to challenge a state. A second common finding of this section is that challenges are rarely mounted, thereby making state structures once they are in place essentially stable: only the most recalcitrant states doom themselves to extinction by removing the safety valve of incremental reform. The second section comprises three articles which note ways in which states affect ethnicity. There is, again, a common position at the back of the different arguments. Ethnicity may be something with a measure of reality, but that reality becomes vital when states start to take it as the measure of reality—indeed, some ethnic identities are fabricated in order to take advantage of political opportunity. Further, it is very noticeable that differences in patterns of ethnic mobilisation result very much from the way paths provided by state policies. The structure of interests is again determined very much by a political rather than a purely societal factor. In recent years no topic has been more discussed than that of the possibility of liberalising and even democratising authoritarian regimes. The articles in this section address the social preconditions for liberal rule from several directions, historical, analytical and comparative, but most agree that the actions of the political élite amount to being the key factor in determining the chances for success of any political opening. A negative way in which this is so may determine the possibility of democratisation in postcommunist societies. It may be that the communism so pulverised civil society and encouraged such generalised distrust, that no partners from below in that part of the world—other than nationalists, bent on creating their own states—will come forward to allow the game of liberalisation to be played successfully. If it is as yet too early to theorise with much sense about post-communism, the studies assembled here at least point to the difficulties involved in creating softer political regimes. It was at one time believed that there was an inexorable logic of social evolution at work sure to make capitalism embrace welfare. Whilst this view no longer gains assent, it is interesting to note that it has been replaced by an equally deterministic viewpoint suggesting that the age of welfare states is coming to an end. The articles that comprise the section on welfare states go beyond such evolutionary nonsense towards analyses of the political bargains which account for the diversity of the provision of welfare across

capitalist society. It is particularly important to note, as do the authors, that welfare should not simply be seen in terms of programmes to do with unemployment insurance, pensions and child care. As relevant for social science consideration, and sometimes far more important than caring programmes, are benefits given to middle classes in terms of the ability to write off house purchases and pension plans against their taxes. When welfare is considered in this fuller sense, we can be quite sure that its study will not somehow come to an end given the increasing interdependence of the world economy. A final section takes the opportunity to look at the peculiarities of the United States, the most important Great Power in the world, but one rarely understood from the outside. Whilst it is distinctively not true to say that the United States has no state, its polity is decentred in ways that affect both industrial and foreign policy making.

SECTION ONE: States and Social Movements


Commentary M.Mann’s ‘Ruling Class Strategies and Citizenship’ is a brilliant tour de force upending received wisdom about the character of working class movements. Economistic thinking, most obviously of marxist hue, has made us think that the working class is unitary, and that it has a singular interest set against capitalist society. Mann demonstrates that before 1914—and since!—there has been a very great variety in the degree of militancy shown by manual workers. Differently put, there are different working classes, not surprisingly since workers tend to be national rather than transnational actors. Within particular countries, levels of militancy are determined by the nature of the state with which workers interact. Liberal states allow workers to organise, and thereby encourage them to seek advancements at the industrial level. In contrast, authoritarian and autocratic states rule out organisation by workers, and thereby necessitate such workers taking on the state. Liberal regimes, in other words, are made stable by the way in which they diffuse conflict through society; despotic regimes, in contrast, tend to be unstable in the long run since they concentrate conflict upon themselves. Mann does not quite stress this principle, and this led him to exaggerate the stability of the Soviet Union. However, he is particularly insightful as to the way in which geopolitics can structure class relations. One way of describing Mann’s findings would be to say that political exclusion radicalises. By and large, people would prefer to be reformers rather than revolutionaries, given generalised dislike for violent death: it is only when the avenue of reform is closed off that the hazardous revolutionary option gains wide appeal. It is exactly this point that J.Goodwin’s ‘Revolutionary Movements in Central America’ makes so strikingly for a very different empirical terrain. Goodwin finds that there is little to be said in favour of economistic and societal views of revolution always coming from below in Central America, supposedly due to endemic peasant poverty consequent on inequality of landholding. This view fails for the simplest of reasons: only a very few countries give rise to revolutionary movements, whilst others seem very stable—a difference that nonetheless rests upon a similar mode of production. Goodwin shows with marvellous perspicacity, and with reference to changes in attitudes of political actors consequent on political openings, the way in which social movements gain character as the result of the extent to which the political system is more or less open. M.Shefter’s ‘Party and Patronage: Germany, England and Italy’ represented a major breakthrough in the study of political parties. One key achievement here was to demonstrate that party cadres have their own agenda, and that they have the capacity to structure events by creating collective identities such that interests take on one form rather than another. A still greater achievement was to explain differences in party types with reference to the historic patterns of state and nation building. Shefter demonstrates the enduring importance of an autonomous bureaucracy in Germany, of parties set against the state in England, and of patronage politics in Italy. The concepts we are offered here are fertile and sophisticated, and the article itself a major piece of historical sociology.


40

43 Ruling Class Strategies and Citizenship*1 Michael Mann

* Source: Sociology, 1987, vol. 21, pp. 339–354.

Abstract Marshall’s theory of citizenship is criticized for being Anglocentric and evolutionist. Comparative historical analysis of industrial societies reveals not one but at least five viable strategies for the institutionalization of class conflict, here called liberal, reformist, authoritarian monarchist, Fascist and authoritarian socialist. In explaining their origin and development emphasis should be placed upon the strategies and cohesion of ruling classes and anciens régimes rather than upon those of the rising bourgeois and proletarian classes (as has been the case in much previous theory). In explaining their durability emphasis should be placed upon geopolitical events, especially the two world wars, rather than on their internal efficiency. If Marshall’s third stage of citizenship is a reasonably accurate description of contemporary Europe, this is primarily due to the military victories of the ‘Anglo-Saxon’ powers.

Marshall’s Theory Novel, important and true ideas are rare. Such ideas which are then developed into a coherent theory are even scarcer. T.H.Marshall is one of the very few to have had at least one such idea, and to develop it. That is why it is important to understand and to improve upon his theory of citizenship. Marshall believed that citizenship has rendered class struggle innocuous; yet citizenship is also in continuous tension, even war, with the class inequalities that capitalism generates. He identified three stages of the struggle for, and attainment of, citizenship: civil, political and social. Civil citizenship emerged in the 18th century. It comprised ‘rights necessary for individual freedom—liberty of the person, freedom of speech, thought and faith, the right to own property and to conclude valid contracts, and the right to justice.’ Political citizenship emerged in the 19th century: ‘the right to participate in the exercise of political power, as a member of a body invested with political authority or as an elector of the members of such a body.’ The third stage, social citizenship, developed through the 20th century: ‘the whole range from the right to a modicum of economic welfare and security to the right to share to the full in the social heritage and to live the life of a civilized being according to the standards prevailing in

Ruling Class Strategies and Citizenship

41

the society.’ It is what we mean now by the Welfare State and social democracy. Through these stages the major classes of modern capitalism, bourgeoisie and proletariat, institutionalized their struggles with the ancien régime and with each other. Citizenship and capitalism were still at war, Marshall declared, but it was institutionalized, rule-governed warfare. Such was the model developed in his famous 1949 lecture, Citizenship and Social Class (1963 edition). It has continued to seem true and important. Major sociologists like Reinhard Bendix, Ralf Dahrendorf, Ronald Dore, A.H.Halsey, S.M.Lipset, David Lockwood and Peter Townsend have acknowledged his influence (e.g. Halsey, 1984; Lipset, 1973; Lockwood, 1974). It remains strong today (see, for example, the recent admiring work by Turner, 1986). This is for a good reason: Marshall’s view of citizenship is essentially true—at least as a description of what has actually happened in Britain. There is one rather remarkable feature of Citizenship and Social Class. It is entirely about Great Britain. There is not a single mention of any other country .2 Did Marshall regard Britain as typical of the capitalist West as a whole? He does not explicitly say so. Yet the most general level of the argument explores the tension between economic inequalities and demands for popular participation, both generated everywhere by the rise of capitalism. This certainly implies a general evolutionary approach, and indeed he does intermittently use the term ‘evolution’. In his book Social Policy (1975 edition), evidence from other countries is only introduced to illustrate variations on a common, British theme. Finally, others have used his model in explicitly evolutionary theories of the development of modern class relations (e.g. Dahrendorf, 1959:61–4). Flora and Heidenheimer (1981:20–1) have observed that general theories of the modern welfare state have been dominated by British experience, chronicled especially by Marshall and Richard Titmuss.

Six counter-theses I wish to deviate from this Anglophile and evolutionary model in six ways. (1) The British strategy of citizenship described by Marshall has been only one among five pursued by advanced industrial countries. I call these the liberal, reformist, authoritarian monarchist, Fascist, and authoritarian socialist strategies. (2) All five strategies proved themselves reasonably adept at handling modern class struggle. They all converted the head-on collision of massive, antagonistic social classes into conflicts that were less class-defined, more limited and complex, sometimes more orderly, sometimes more erratic. Thus evolutionary tales are wrong. There has been no single best way of institutionalizing class conflict in industrial society, but at least five potentially durable forms of institutionalized conflict and mixes of citizen rights. (3) In explaining how such different strategies arise, I will stress the role of ruling classes. By ‘ruling class’ I mean a combination of the dominant economic class and the political and military rulers. I do not mean to imply that such groups were unchanging or even united—indeed the degree of their cohesion will figure importantly in my narrative. But I do imply the pair of general explanatory precepts expressed in (4) and (5) below. (4) Influence on social structure varies according to power. As a ruling class possesses most power, its strategies matter most. In fact, many anciens regimes could survive the


42

onslaught of emergent classes with a few concessions here and there. Neither the bourgeoisie nor the proletariat has been as powerful as has been argued by the dominant schools of sociology, liberal, reformist (like Marshall) and Marxist. Indeed, ruling class strategies tended to determine the nature of the social movements generated by bourgeoisie and proletariat, especially whether they were liberal, reformist or revolutionary. This argument has also been made by Lipset (1985: Chapter 6). (5) Tradition matters. We generally exaggerate the transformative powers of the Industrial Revolution. That Revolution was preceded by centuries of structural change— the commercialization of agriculture, the globalization of trade, the consolidation of the modern state, the mechanization of war, the secularization of ideology. If anciens régimes had learned to cope with these changes, they could master the problems of an industrial society with traditional strategies, up-dated. If not, they were usually already vulnerable and internally divided before the actual bourgeois or proletarian onslaught. Others have also stressed the survival of tradition through the Industrial Revolution— classically Moore (1969) and Rokkan (1970), more recently Mayer (1981) and Corrigan and Sayer (1985). (6) The durability of régime strategies has been due less to their superior internal efficiency than to geopolitics—and specifically to victory in world wars. The geopolitical and military influences on society have been considerable but neglected in sociological theory. However, they have recently been receiving the attention they deserve (e.g. Giddens, 1985; Hall, 1985; Mann, 1980, 1986a and b; Shaw, 1987; Skocpol, 1979). Let us approach the historical record with these six theses in mind. What were the traditional regime strategies used to cope with the initial rise of the bourgeoisie?

Absolute and Constitutional Regimes3 We can divide the regimes of pre-industrial Europe into approximations to two idealtypes, absolute monarchies and constitutional regimes. By 1800 the principal absolutists were Russia, Prussia and Austria. Their monarch’s formal despotic powers were largely unlimited. Citizenship was unknown. The rule of law supposedly operated, but personal liberties, and freedom of the press and association could be suspended arbitrarily. Indeed, any conception of universal rights was restrained by the proliferation of particularistic statuses, possessed by corporate groups—estates of the realm, corporations of burghers, lawyers, merchants and artisan guilds. Yet the real, infrastructure powers of the monarchs were far from absolute. They required the cooperation of the regionally and locally powerful. Repression was cumbersome and costly, and far more effective if used together with ‘divide-and-rule’ negotiations with corporate groups. The monarch’s crucial power was tactical freedom: the capacity to act arbitrarily both in conducting negotiation and in using force. It is important to realize that these three characteristics—arbitrary divide-and-rule, selective tactical repression, and corporate negotiations—survived intact into the 20th century. Britain and the United States were the main constitutional regimes. There civil citizenship was well-developed. Individual life and property were legally guaranteed, and


43

freedom of the press and of association were partially recognized—they were licensed’ under discernible rules. Political citizenship also existed, though it was confined to the propertied classes who Virtually represented’ the rest. Social citizenship was as absent here as in absolutist regimes. All this was well understood by Marshall. Not all regimes were either predominantly absolutist or constitutional. Some formerly absolutist regimes had experienced revolution or serious disorder, and were now bitterly contested between constitutionalists and reactionaries: France after 1789, Spain and several Italian states. In others absolutism and constitutionalism merged through less violent, more orderly conflict: principally the Scandinavian countries. Capitalist industrialization changed much, but we can nonetheless see the initial imprint of these four types of regime: absolutist, constitutional, contested and merged. Let us follow this in more detail, concentrating in turn on the US, Britain and Germany.

From Constitutionalism to Liberalism—The US and Britain In Britain and the US the rise of liberalism strengthened civil and political citizenship. The rule of law over life, property, freedom of speech, assembly and press was extended, as was the political franchise. But any social citizenship remained equivocal. The regime provided basic subsistence to the poor out of charity and a desire to avoid sedition. But provision came from local worthies and private insurance; and legislation encouraged rather than enforced. Subsistence was not a right of all, but the result of a mixture of market forces, the duty to work and save, and private and public charity. The state was not interventionist or ‘corporatist’: interest group conflict was predominantly left to the economic and political marketplaces, its limits defined by law. However, collectivities could legitimately exploit their market powers, and the regime devised rules of the ensuing game. Under liberalism individuals and interest groups, but not classes, could be accommodated within the regime. Repression, now fully institutionalized, was reserved only for those who went outside the rules of the game. Such was one basic strategy of dealing with the rise of the bourgeoisie. But could it cope with the working class? The two main cases, the United States and Britain, coped differently. In the US labour was eventually absorbed into the liberal regime. A broad coalition, from landowners and merchants down to small farmers and artisans had made the Revolution. White, adult males could not be easily excluded from civil and political citizenship. By the early 1840s all of them, in all states, possessed the vote—50 years earlier than anywhere else, 50 years before the emergence of a powerful labour movement. Thus the political demands of labour could be gradually expressed as an interest group within an existing federal political constitution and competitive party system. As Katznelson (1981) has shown, workers’ political life became organized more by locality, ethnicity and patronage than by work, unions or class. In the sphere of work there was severe and violent conflict between unions and employers aided by government and the law courts. But here too the ruling class eventually came to accept the legitimacy of unions in essentially liberal terms; while the Wagner Act allowed unions to negotiate freely, TaftHartley compelled them to act only as the balloted representatives of their


44

individual members. The US gives us the truest picture of what would have happened to class conflict without the politics of citizenship. If class struggle had only concerned the Marxist agenda, of relations of production, labour processes, and direct conflict between capitalists and workers, then liberal regimes would have dominated industrial society. As the (white) working class was civilly and politically inside the regime, it had little need for the great ideologies of the proletariat excluded from citizenship socialism and anarchism. American trade unions became like other collective interest groups exploiting their market power. If workers did not possess effective market powers, they would be outside this liberal regime and tempted by socialism and anarchism. But they could be repressed—with the consent of labour organizations accepting the rules of the game. Consequently, neither class nor socialism has ever appeared as a fundamental organizing principle of power in the United States. Those groups who in other countries constituted the core of the labour and socialist movement—male artisans, heavy industrial, mining and transport workers—became predominantly interest groups inside the liberal regime, while the unskilled, those in other sectors, females and ethnic minorities were left outside. Liberalism was thus the first viable regime strategy of an advanced industrial society. It still dominates the United States, and is also found in Switzerland. In these countries social citizenship is still marginal. Economic subsistence and participation is provided overwhelmingly out of the economic buoyancy of their national capitalisms, from which the large majority can insure themselves against adversity. Below that, there are welfare provisions against actual starvation, though they vary between states and cantons, are often denied to immigrant workers, and are sometimes provided only if the poor show their ‘worth’. It is closer to the 18th century Poor Law than to what Marshall meant by social citizenship. Its social struggles remained defined by liberalism. If civil and political citizenship could be attained early, before the class struggles of industrialism, then social citizenship need not follow. The most powerful capitalist state has not followed Marshall’s road. It shows no signs of doing so. But Britain strayed from liberalism towards reformism, as Marshall depicted. Britain’s initial struggle for liberal political citizenship was more of a class struggle, waged predominantly by the rising bourgeoisie and independent artisans. However, the British constitution has not excluded classes or status groups as systematically as have most constitutions of continental Europe. The franchise before 1832 was extraordinarily uneven; then, until 1867, it passed through the middle of the artisan group; between 1867 and 1884 it grew to include 65% of the adult male population. In 1918 all adult males and many females were included, and in 1929 all females. Hence at any particular point in time emerging dissidents—petty bourgeois radicals, artisan and skilled factory worker socialists, feminists—have been partially inside, partially outside the state. Thus liberalism and socialism have both remained attractive ideologies. Indeed, perhaps only the splits in the Liberal Party consequent on the First World War may have ensured that a joint liberal/reformist ideology would be carried principally by an independent Labour Party, rather than through Lib-Lab politics. Britain has enshrined the rule of both interest groups and classes, jointly. The labour movement is part sectional interest group, part class movement, irremediably reformist, virtually unsullied by Marxist or anarchist


45

revolutionary tendencies. Britain is thus a mixed liberal/reformist case. The state remains liberal, unwilling to intervene actively in interest-group bargaining—it has incorporated the lower classes into the rules of the game, not into the institutions of ‘corporatism’. Yet social citizenship has advanced somewhat beyond the American level. The state guarantees subsistence through the welfare state, but this meshes into, rather than replaces, private market and insurance schemes. Thus its major social struggles are fought out in terms of an ideological debate, and a real political pendulum, between liberalism and social democracy. In reaction to the Thatcher government’s liberal strategy, the reformist strategy is now becoming more popular again.

Contested and Merged Regimes—France, Spain, Italy, Scandinavia In France, Spain and Italy, reactionaries (usually monarchist and clerical) and secular liberals struggled over political citizenship for most of the 19th and 20th centuries, with many violent change of regime. Citizenship remained bitterly disputed, though there was undoubtedly some secular progress in the Marshallian direction. As radical bourgeoisie, peasantry and labour were erratically but persistently denied political citizenship, these developed competing excluded ideologies. Sometimes they rejected the state, as in anarchism and syndicalism; sometimes they embraced it, as in Marxist socialism. The fierce competition between anarcho-syndicalism, revolutionary socialism and reformist socialism was not solved until after World War II, for reasons I mention later. In several other countries the absolutist/constitutional struggle proceeded to more peaceful victory for a broad alliance between bourgeoisie, labour and small farmers. Over the first four decades of this century they achieved civil and political citizenship, and proceeded furthest along the road to social citizenship. The absolutist inheritance, never violently repudiated (unlike in France), provided a more corporatist tinge to regime negotiations which still endures. The Scandinavian countries are the paradigm cases of this route, less affected by the dislocations of war than any other. This second road, a corporatist style of reformism, corresponds closely to Marshall’s vision (more so than the British case does). Its social struggles are avowedly class ones, but they are managed by joint negotiations, and constrained more by pragmatic that ideological limits. Continuing reform, it is agreed, will be limited primarily by the growth record of each national economy. But to investigate properly the absolutist legacy suggests a methodology of examining the ‘purer’ and longer-lasting cases of absolutism, in Russia, Austria, Japan and especially in Prussia/Germany.

From Absolutism to Authoritarian Monarchy—Germany, Austria, Russia, Japan The absolutist regimes entered the 19th century with two conflicting predispositions. First, monarch, nobility and Church were unwilling to grant universal citizen rights to


46

either bourgeoisie or proletariat, since that would threaten the particularistic, private and arbitrary nature of their power. Second, despite their despotic appearance, they were pessimistic about their infrastructural capacity to overcome determined resistance with systematic repression. When it became obvious that neither the bourgeoisie nor the proletariat would go away, the regimes not only cast around for other solutions to maintain their power—they also realized that to incorporate these rising groups would ‘modernize’ the regime and increase its Great Power status. The most successful regime in Europe was Wilhelmine Germany, on which I will therefore concentrate.4 German absolutists were willing to concede on civil citizenship. Often this did not seem like ‘concession’ at all. Ancien régime members were major propertyholders, gradually using their property more capitalistically. They were not opposed to the spread of universal contract law and guarantees of property rights—including the liberal conception of freedom of labour. Recent Marxists have observed that classical liberalism, combining capitalism with democracy, has not often appeared subsequently: much civil can exist with little political citizenship (e.g. Jessop, 1978). Blackbourne and Eley (1984) have demonstrated this case with respect to 19th century Germany: liberal legal rights (civil citizenship) were achieved through a consensus between the Prussian regime and the bourgeoisie over what was needed to modernize society. Absolutist regimes also favoured a minimal social citizenship. Their ideology and particularistic practices were already paternalist. Particular groups like artisans or miners often had their basic wages, hours and working practices guaranteed by the state. When state infrastructural powers expanded, after about 1860, so could a minimal social citizenship. As is generally recognized today, Bismarck and Kaiser Wilhelm, and not liberals or reformists, were the founders of the Welfare State, though it is true that they did not take it very far (Flora & Alber, 1981). The sticking-point was over political citizenship. Real parliaments could not be conceded; democrats could not be allowed absolute freedoms of the press, speech or assembly. Gradually, however, the more astute monarchists institutionalized a workable political strategy. The regime conceded a parliamentary shell but weighted the franchise, rigged ballots, and only allowed elected representatives limited powers alongside an executive branch responsible to the monarch alone. Thus the bourgeoisie, even the proletariat, could be brought within the state but could not control it. By this sham political citizenship they were ‘negatively incorporated’, to use Roth’s (1963) term. The tactics were divide-and-rule: negotiate with the more moderate sections of excluded groups, then repress the rest; play off incorporated interest groups and classes against each other; and preserve a vital element of arbitrary regime discretion. In the hands of a Bismarck the discretion could be used quite cynically: Catholics, regionalists, National Liberals, classical liberals, even the working class, would be taken up, discarded, and repressed according to current tactical exigencies (see the brilliant biography of Bismarck by Taylor, 1961). Divide-and-rule was corporatist and arbitrary— both qualities inherited from absolutism. Groups and classes were integrated as organizations into the state, rather than into rule-governed marketplaces. The state could alter the rules by dissolving parliament, restricting civil liberties, and selecting new targets for repression. By these means authoritarian monarchism emasculated the German bourgeoisie, dividing it among Conservative, National Liberal, Catholic and regionalist


47

factions, all vying for influence within the regime. By 1914 the German bourgeoisie was finished as an independent political force (as Max Weber so often lamented). Only a small radical rump was prepared to ally with the excluded socialists against the regime. The proletariat was treated more severely. Though the regime became somewhat internally divided, and though different Länder also varied (with liberals arguing that concessions to labour unions would detach them from socialism), in the end the authoritarians proved to be the heart of the regime. Apart from a brief period (1890–94) under the Chancellorship of Caprivi, a liberal Prussian general, the politics of conciliation never carried the court—and the Kaiser dismissed Caprivi rather than make concessions to labour. The regime was essentially united and so could respond with a clear strategy. The German working class could elect representatives to the Reichstag, but these were excluded from office or influence on the regime. Unions were permitted, but—even after the anti-Socialist Laws were repealed in 1889—their legal rights were unclear. The state could exploit legal uncertainties or invoke martial law to repress strikes, meetings, marches, organizations and publications. It did so arbitrarily, according to its traditions. Faced with a strategy largely of civil and political exclusion, labour responded predictably. It followed the Marxist Social Democrats, ostensibly revolutionary but geared up in practice to fight the elections. Most activist workers joined the socialist unions, committed to SPD rhetoric, but able to make reformist gains in some industries and localities. But to be a reformist brought frustration, because of regime intransigence. By 1914 Karl Legien, the crypto-reformist leader of the socialist unions, had carefully built up a measure of autonomy from the SPD. But he was forced to confess that reform was impossible without a fundamental change in the state. The working class was largely outside political citizenship. It responded with a flawed revolutionary Marxism—extreme rhetoric, practical caution, and a leadership, conscious of the isolation of the movement, concentrating on electoral politics. How frightened was the regime of the socialist threat? In the 1912 election the SPD achieved its greatest success, capturing a third of the votes, and becoming the largest single party in the Reichstag. The regime was taken aback but quickly recovered. The Chancellor, Bethmann-Hollweg, used the Red Scare against his major enemy at the time, which was the Right, not the Left. He exploited the fears of the propertied classes to finally push through an income tax, long desired by the regime, long resisted by the agrarian landlords.5 Authoritarian monarchy was still successfully dividing-and-ruling and modernizing at the onset of the first World War. Each of the authoritarian monarchies provided its variation on this German theme. I discuss them briefly in order of their success, beginning with Japan, the most successful. The Japanese monarchy itself had less freedom of action. Instead a tightly-knit Meiji elite, modernizing but drawn from the traditional dominant classes, used the monarchy as its legitimating principle. The Meiji Revolution represented an unusually self-conscious regime strategy of conservative modernization. After a careful search around Western constitutions, the German constitution was adopted and modified according to local need.6 It is worth adding that forms of organization from liberal-reformist countries were also borrowed where they could fit into an authoritarian mould—notably French army and British navy organization.7 Authoritarian monarchy became rather more corporate, less dependent on the personal qualities of the monarch, than in Europe—an apparent


48

strengthening of the strategy. Less successful was Russia, whose regime generally favoured more repression and exclusion, yet vacillated before modern liberal and authoritarian influences from the West. Two periods of regime conciliation (1906–7 and 1912–14) enabled the emergence of bourgeois parties of compromise and labour unions run by reformists. But each time the subsequent return to repression cut the ground from under liberals and reformists. They could promise their followers little. Many became embittered and moved leftward. Socialist revolutionaries took over the labour and peasant movements and even some of the bourgeois factions (see e.g. on the workers’ movement, Bonnell, 1983, and Swain, 1983). Divisions and vacillation at court prevented successful emulation of the German model. The ancien régime still possessed the loyalty of the nobility and propertied classes in general, but its modernization programme began to disintegrate from within (as Haimson, 1964 and 1965, classically argued). The regime lacked a corporate core of either liberal or conservative modernizers. Stolypin, the architect of the agrarian reforms designed to recruit rich and middling peasant support, was the potential conservative saviour of the regime, yet his influence at court was always precarious. The divided regime became buffeted by the personal irresoluteness of Nicholas and the reactionary folly of Alexandra. When monarchy begins to depend on the personal qualities of its monarchs, it is an endangered species. Russia represented the opposite pole to Japan within the spectrum of authoritarian monarchy—no corporate regime strategy, much depending on the monarch himself. On the other hand, economic and military modernization was proving remarkably successful in pre-war Russia. Could the regime find a comparably coherent political strategy? In 1914 the answer was not yet clear. Though regime weaknesses had begun to create what later proved to be its revolutionary grave-diggers, their influence was still negligible in 1914. The least successful case was Austria (become the Dual Monarchy of Austria-Hungary in 1867), uniquely beset by nationality conflicts as well as class struggle across its variegated lands.8 The monarchy attempted divide-and-rule on both fronts at once, but was faced by defections among ancien régime groups (Hungarian and Czech nobilities) as well as the hostility of bourgeois liberal nationalism. As the monarchy faltered, some peculiar alliances developed. After 1867 the most loyal and dominant groups in the two halves of the Dual Monarchy were the German nobility and bourgeoisie and the Hungarian nobility. But the monarchy found their support unwelcome because it alienated all the other nationalities these two exploited. After 1899 the Marxist SPD rejected nationalism as a bourgeois creed, thereby becoming to its surprise the major de facto supporter of the transnational monarchy. The monarchy belatedly converted to parliamentary institutions similar to Germany’s (universal suffrage to parliaments whose rights were subordinate to the monarchy’s), and tried to reach out to exploited nationalities and even classes. But noble and bourgeois nationalists, not the proletariat, made the parliaments unworkable, and they were dissolved. This authoritarian monarchy could not even retain the loyalty of the whole ancien régime, let alone incorporate the bourgeoisie. By 1914 the regime consisted of the monarchy, the army, and the largely tactical support of various national and class groupings. Its corporate solidarity was probably the weakest of the four cases. The four cases reveal considerable variation in regime strategy and success. The


49

crucial criteria of success were to maintain the corporate coherence of the ancien régime, and to modernize by incorporating sections of the bourgeoisie. It is outside the scope of this article to attempt to explain why some régimes did much better than others at these tasks. However, régimes seem not to have prospered or faltered because of the strength in general class and numerical terms of bourgeoisie and proletariat. In these terms the rising classes in Germany were initially the most threatening, those of Japan the least threatening, with Austria and Russia somewhere in between. This is not the same ordering as for regime success. The bulk of the explanation of success would seem to lie among the traditional regimes and classes, not among the rising classes. At its most coherent, authoritarian monarchy provided a distinctive mixture of citizen rights—a fair degree of civil citizenship, minimal social citizenship, limited political citizenship, the whole varying by class and tactically undercut by an arbitrary monarchy and court-centred elite. Its social struggles were part ideological class struggle, part incorporated interest-group jostling, erratically violent yet institutionalized nonetheless. Was this the third viable strategy for advanced industrial societies? Could it have survived the working class pressure indefinitely? But for the fortunes of war, would it still survive today in three of the four greatest industrial powers in the world, a united Germany, a Tsarist Russia, and an Imperial Japan? We cannot be sure because these regimes collapsed in war. But let us consider four supports for this counter-factual possibility. First, in its own time Wilhelmine Germany was not idiosyncratic. Its emerging institutions were better-organized versions of the European mainstream. As Goldstein (1983) has shown, the combination of selective repression and sham parliaments was the late 19th century norm, not well-developed liberalism, still less reformism. For this reason German institutions were much copied, especially by Austria and Japan. Second, by the time of their entry into the decisive war, 1914 (or 1941 in the case of Japan), the authoritarian monarchies were already becoming great industrial powers. Germany had overtaken Britain and France and was matched only by the United States. Japan and Russia were industrializing rapidly and successfully; and Russian economic resources, then as now, made up in quantity what they lacked in quality (quantitative indices of the economic strength of the Great Powers can be found in Bairoch, 1982). Authoritarian monarchy was surviving into advanced industrial societies in Germany and Japan, still had a reasonable chance in Russia, and was obviously failing only in Austria, where nations, not classes, provided the main threat. Third, we must beware a too-homogeneous view of industrial society and its class struggles. The main reason the working class was not so threatening was its limited size. National censuses conducted between 1907 and 1911 show Britain to be exceptional. Only 9% of its working population was still in agriculture, compared to 32% in the US, 37% in Germany, and more than 55% in Russia. Among the major powers only in Britain were more working in manufacturing than in agriculture (Bairoch, 1968: Table A2 has assembled the census data). Outside Britain, labour needed the support of peasants and small farmers to achieve either reform or revolution. It achieved this partially in the ‘Contested’ cases of France, Italy and Spain, and more sustainedly in the ‘mixed’ cases of Scandinavia. But in Germany, Japan and Austria it failed dismally. Socialism was trapped in its urbanindustrial enclaves, outvoted by the bourgeois-agrarian classes, and


50

repressed by peasant soldiers and aristocratic officers. Authoritarian monarchy could continue to divide and rule and selectively repress provided it could manipulate divisions between agrarian and bourgeois classes, and motivate them both with fear of the proletariat. Few 20th century socialists have broken this strategy—Lenin being the obvious exception. Fourth, the numerical weakness of labour has continued, though in changed form. The rise of the ‘new middle class’ and of the ‘service class’, the reemergence of labour market dualism, and the increasing size and variety of service industries soon introduced new differentiations among the employed population, just as agriculture declined. Successful labour movements in the post-war period, like those of Scandinavia, have managed to repeat their earlier populist strategy (Esping-Anderson, 1985). They have recruited whitecollar workers and new economic sectors into the Social Democratic movement, just as they earlier recruited bourgeois radicals and small farmers. But could labour movements which had already failed to attract the bourgeoisie or farmers, as in Germany or Japan, now do better among newer groups? It is surely more plausible to conceive of divide-andrule, selective repression strategies, wielded by arbitrary authoritarian monarchies, surviving successfully today in Germany and Japan, and possibly also in Russia and constituent parts of Austria-Hungary. I conclude that the third strategy, authoritarian monarchy, could probably have survived into advanced, post-industrial society, providing a distinctive, corporately organized, arbitrary combination of partial civil, political and social citizenship. This was not envisaged by Marshall, or indeed by any modern sociologist.

Fascism and Authoritarian Socialism World War I resulted in two further strategies, fascism and authoritarian socialism. Nazi Germany and the Soviet Union are their exemplars. Both used more repression, using the infrastructural capacities of the 20th century state, and proclaiming violent legitimating ideologies. In practice, as in all regimes, repression had to be combined with negotiation. Both regimes delineated out-groups with whom they would not negotiate: for both, anyone providing principled opposition; for the Nazis labour leaders, socialists, Jews and other non-Aryan groups; for the Soviets, major property-owners. But other interest groups—never acknowledged as antagonistic classes—could join the regime, establish cliques within and clients without, and bargain and jostle in time-honoured absolutist style. Now social struggles were not openly acknowledged at all. But within the regime they would continue, flaring into intermittent life with purges, riots and even armed factional struggle. Neither regime provided civil rights; neither provided real political citizenship (though they provided the institutions of sham corporatism and socialism). Yet they moved furthest toward social citizenship. Fascism’s move was hesitant: full employment and public works programmes were not greatly in advance of others of the time (and were partially an outcome of a more important policy goal, rearmament). But had the regime survived the war, its encroachments on capitalism would surely have extended the state’s role in guaranteeing subsistence. The Soviet regime has gone much further, proud of its


51

programme of social citizenship. The state formally provides the subsistence of all (though the reality, with private peasant plots and black markets, is less clear-cut). Of course, German Fascism was deeply unstable. But this was due to the restless militarism of its leaders in geopolitics, not to its class strategy. Indeed, this was remarkably successful in a short space of time. The proletariat was suppressed more completely than any of the regimes discussed so far would have believed possible. Its leaders were killed or exiled; its organizations disbanded or staffed by the regime’s paramilitary forces; its masses silenced, seemingly with the approval of other social classes. The bourgeoisie was emasculated even more effectively than the Wilhelmine regime had managed. The liberals were killed or silenced, the rest kept quiet or loudly voiced their support. Ruthlessness was no longer hidden by scruple. Thus Fascism might have offered a fourth, chilling resolution to class struggle in advanced societies. Its main test would have been the next one: could it take on capital too? It was already beginning to do this by subordinating economic profit to militarism. This proved its downfall—but not at the hands of domestic social classes, who fought loyally for the Nazi regime down to its last days. The stability of the fifth solution, authoritarian socialism, cannot be in doubt. The Bolsheviks and their ruling successors soon cowed the bourgeoisie, and gradually domesticated the labour movement. The trade unions were converted into apolitical welfare state organizations (sometimes headed by ex-KGB men). It took almost fifty years for the institutionalization to be complete. But once in place, it appears no less stable than other enduring types of regime.

The Impact of War and Geopolitics I have described five viable regime strategies and mixtures of citizenship: liberalism, reformism, authoritarian monarchy, Fascism and authoritarian socialism. Yet industrial society today has lost some of this variety. Authoritarian monarchy and Fascism no longer exist. Why? Is it because of their inherent defects or instability? I have already suggested not. There is an alternative explanation. To paraphrase a famous epitaph on the Roman Empire—these regimes did not die of natural causes, they were assassinated. Of course, Fascism and authoritarian socialism were also born out of assassination. But for the fortunes of World War I, authoritarian monarchy might be alive today, while Fascism and authoritarian socialism might never have been born. But for the fortunes of World War II Fascism might dominate the world today. True, it is difficult to see American liberalism being overthrown by the German, Austrian and Japanese alliances. But Europe and Russia might well have had viable futures under very different regimes. Of course, proof of this argument would require disposing of the reverse causality: regime type might have determined the role of war. This could have happened in two stages. Certain regimes—obviously the more authoritarian ones—may have been more militaristic and provoked the world wars; yet they may have been less effective at fighting them. The first stage has validity. The Nazis and Japanese did aggress in World War II; and, in a more confused, stumbling way, the authoritarian monarchies did start


52

World War I. But is the second stage of the argument valid? Were liberalism, reformism and authoritarian socialism better suited to mass mobilization warfare? The ideologies of the victors suggest the answer ‘yes’. I have only time here to give fragmentary evidence, but my answer is ‘no’. In both wars the German army fought better than its enemies, who continuously needed numerical superiority to survive. German civilians also loyally supported their regimes to the end. Both points hold also for the Japanese in the second war. The Eastern Front in the first war offers further shocks to the liberal/reformist perspective. Authoritarian monarchy Russia outfought the by now semi-authoritarian monarchy of Austria-Hungary, whose troops in turn outfought the by now largely liberal regime of Italy. Indeed, when in 1917 the Austro-Hungarian armies against Russia collapsed, they were stiffened by Prussian officers and NCOs and then began to get the upper hand (Stone, 1975). The Central and Axis powers were correct in their view that the fortunes of war turned less upon citizenship than on efficient military organization. Unfortunately for them, military efficiency became over-weighted by numbers. Numbers resulted principally from the alliance system—how may powerful states were on each side? Authoritarian monarchy and Fascism were defeated by superior geopolitical alliances, not by their domestic socio-political structure. After 1945 this result was deliberately rammed home by the victors, careful not to repeat the mistakes of the peace treaties of 1918 (see Maier, 1981). Eastern Europe was made safe for authoritarian socialism by the Red Army. Western Europe and Japan were more subtly made safe for liberal/reformist regimes (though Japan’s regime does not fit happily into this categorization, because of the survival of many authoritarian traditions). In Western Europe the authoritarian Right was eliminated by force, the revolutionary Left had the ground cut from under it by reforms and economic growth offered to governments and industrial relations systems of the Centre and Centre-Left. By 1950 the contest was over. A cross between Marshallian citizenship and American liberalism dominated the West, less through its internal evolution than through the fortunes of war. It still dominates today. Marshall’s general argument was that industrial society institutionalized class struggle through mass citizenship. This seems true. All regimes have guaranteed some citizen rights. But they have done so in very different degrees and combinations. It is a more complex and less optimistic overall picture than he envisaged. But for the logic of geopolitics and war—including the sacrifices of his own generation—it might have been a very different and infinitely more depressing picture in Europe. Sociologists are prone to forget that ‘evolution’ is usually geopolitically assisted. Dominant powers may impose their strategies on lesser powers; or the lesser may freely choose the dominator’s strategy because it is an obviously successful modernization strategy. This means that what ‘evolves’ depends on changing geopolitical configurations. Let me quote Ito Hirobumi, the principal author of the Meiji constitution of 1889: ‘We were just then in an age of transition. The opinions prevailing in the country were extremely heterogeneous, and often diametrically opposed to each other…there was a large and powerful body of the younger generation educated


53

at the time when the Manchester theory was in vogue, and who in consequence were ultra-radical in their ideas of freedom. Members of the bureaucracy were prone to lend willing ears to the German doctrinaires of the reactionary period, while, on the other hand, the educated politicians among the people having not yet tasted the bitter significance of administrative responsibility, were liable to be more influenced by the dazzling words and lucid theories of Montesquieu, Rousseau and similar French writers.’ I have taken this quotation from Bendix (1978:485) who uses it in support of a general evolutionist model of how Western ideals of popular representation supplanted monarchy everywhere. He rightly notes the importance of ‘reference societies’, more advanced societies to which modernizes could point with approval. But the quotation reveals that at the end of the 19th century there were at least three—Britain, France and Germany—and this reflected a real balance of power among several great powers. No single power could impose its will on others (outside its colonial or regional sphere of influence). Modernizers could choose from among several regime strategies. That is far less the case today. The Soviet and Anglo-American strategies were imposed—in the East by force, in the West by assisting certain political factions and subverting others. The two strategies have worked in their different ways for 40 years, and are now backed by the economic, ideological, military and political resources of two hegemonic superpowers. Eastern Europe is still held down by force. In the Western European periphery, deviant regimes in Portugal, Spain and Greece have succumbed to the Anglo-American vision of modernization desired increasingly by their domestic elites. In the Third World there is more variety of choice, because most countries are more insulated from both Western and Eastern blocs, but the choices tend to be around the two models provided by the superpowers. Geopolitics has also provided a second recent change: the emergence of nuclear weapons. Warfare at the highest level would now destroy society. Therefore, the warassisted pattern of change dominant in the first half of the century cannot be repeated. The emergence of the superpowers and of nuclear weapons both indicate that the future of citizenship will be different from its past. Our assessment of its prospects must combine domestic with geopolitical analysis.

Notes 1. An earlier version of this article was given as the 1986 T.H.Marshall Memorial Lecture at the University of Southampton. My thanks go to the University’s Department of Sociology for its invitation and hospitality and to John Hall and David Lockwood for their helpful criticisms of that version. 2. I write ‘Great Britain’ rather than ‘The United Kingdom’ because there is also no reference to Northern Ireland, which, of course, would not fit well into his theory. 3. The historical generalizations contained in the rest of this essay are given more empirical and bibliographic support in Mann, 1989. For the distinction between despotic and infrastructural power, see Mann 1984. 4. The literature on Wilhelmine Germany is enormous and often controversial. Apart from


5. 6. 7. 8.

54

works cited later, good concise general accounts are provided by Calleo, 1978:57–84, and by various essays in Sheehan, 1976. Kaiser, 1983:458–62, makes this argument, against the more traditional view of writers like Berghahn, 1972, that the regime feared the Left and militarized society to counter its threat. Bendix, 1978:476–90, gives a succinct summary of the Meiji strategy. I am grateful to Professor Michio Morishima for this observation. Historical sociologists have tended to ignore Austria, except in relation to nationalism. For a narrative that enables us to piece together most of the complex relations between regime, classes and nations, see Kann 1964.

References Bairoch, P. 1968. The Working Population and its Structure. Brussels: Université Libre de Bruxelles. Bairoch, P. 1982. ‘International industrialization levels from 1750 to 1980’, Journal of European Economic History, 11. Bendix, R. 1978. Kings or People. Power and the Mandate to Rule. Berkeley: University of California Press. Berghahn, V . 1973. Germany and the Approach of War in 1914. London: St.Martins Press. Blackbourne, D. & Eley, G. 1984. The Peculiarities of German History. Oxford: Oxford UP. Bonnell, V.E. 1983. Roots of Rebellion: Workers Politics and Organizations in St.Petersburg and Moscow 1900–1914. Berkeley: University of California Press. Calleo, D. 1978. The German Problem Reconsidered. Germany and the World Order 1870 to the Present . Cambridge: Cambridge University Press. Corrigan, P. & Sayer, D. 1985. The Greate Arch: English State Formation as Cultural Revolution. Oxford: Blackwell. Dahrendorf, R. 1959. Class and Class Conflict in an Industrial Society. London: Routledge and Kegan Paul. Esping-Anderson, G. 1985. Politics against Markets: The Social Democratic Road to Power. Princeton, NJ: Princeton UP. Flora, P. & Alber, J. 1981. ‘Modernization, Democratization, and the Development of Welfare States in Western Europe’ in Flora, P. & Heidenheimer, A.J. (eds.) The Development of Welfare Slates in Europe and America. New Brunswick, N.J.: Transaction Books. Flora, P. & Heidenheimer, A.J. 1981. ‘Introduction’ in their The Development of Welfare States op. cit. Giddens, A. 1985. The Nation State and Violence. Oxford: Polity Press.


55

Goldstein, R.J. 1983. Political Repression in 19th Century Europe. Beckenham: Croom Helm. Haimson, L.H. 1964 and 1965. ‘The problem of social stability in urban Russia, 1905– 1917’, Parts 1 and 2, Slavic Review, 23 and 24. Hall, J. 1985. Powers and Liberties. Oxford: Blackwell. Halsey, A.H. 1984. ‘T.H.Marshall: Past and Present, 1893–1981’, Sociology, 18. Jessop, B. 1978. ‘Capitalism and democracy: the best possible political shell?’ in G. Littlejohn et al. Power and the State. London: Croom Helm. Kaiser, D.E. 1983. ‘Germany and the origins of the First World War’, Journal of Modern History, 55. Kann, R.E. 1964. The Multinational Empire. New York: Octagon Books. 2 vols. Katznelson, I. 1981. City Trenches: Urban Politics and the Patterning of Class in the United States. New York: Pantheon Books. Lipset, S.M. 1973. ‘Tom Marshall—Man of Wisdom’, British Journal of Sociology, 24. Lipset, S.M. 1985. Consensus and Conflict: Essays in Political Sociology. New Brunswick: Transaction Books. Lockwood, D. 1974. ‘For T.H.Marshall’, Sociology, 8. Maier, C.S. 1981. ‘The two postwar eras and the conditions for stability in 20th century Western Europe’, American Historical Review, 86. Mann, M. 1980. State and Society, 1130–1815: an analysis of English state finances, in Zeitlin M. (ed.) Political Power and Social Theory. Vol 1. Connecticut: J.A.I.Press. Mann, M . 1984. ‘The autonomous power of the state’ in Archives Européennes de Sociologie, 25. Mann, M. 1986a. The Sources of Social Power, Vol. I: A History of Power from the Beginning to 1760 A.D.New York & London: Cambridge UP. Mann, M. 1986b. ‘War and Social Theory: into Battle with Classes, Nations and States’, in Shaw, M. & Creighton, C. (eds.) The Sociology of War and Peace. London: MacMillan. Mann, M. 1989. The Sources of Social Power, Vol. II: A History of Power in Industrial Societies New York & London: Cambridge UP. Marshall, T.H. 1963. ‘Citizenship and Social Class’ in his Sociology at the Crossroads. London: Heinemann. Marshall, T.H. 1975. Social Policy. London: Hutchinson, 4th revised edition. Mayer, A.J. 1981. The Persistence of the Old Regime. London: Croom Helm. Moore, B. 1969. The Social Origins of Dictatorship and Democracy. Harmondsworth: Penguin. Rokkan, S. 1970. Citizens, Elections, Parties: Approaches to the Comparative Study of the processes of Development. Oslo: Universitetsforlaget. Roth, G. 1963. The Social Democrats in Imperial Germany. Totowa, NJ: Bedminster


56

Press. Shaw, M. 1987. The Dialectics of Total War. London: Pluto Press. Sheehan, J.J. 1976. Imperial Germany. New York: Franklin Watts. Skocpol, T. 1979. States and Social Revolutions. Cambridge: Cambridge UP. Stone, N. 1975. The Eastern Front, 1914–1917. New York: Charles Scribner’s Sons. Swain, G. 1983. Russian Social Democracy and the Legal Labour Movement, 1906– 1914. London: Macmillan. Taylor, A.J.P. 1961. Bismarck: the Man and the Statesman. London: Arrow Books. Turner, B.S. 1986. Citizenship and Capitalism: the Debate over Reformism. London: Allen & Unwin.


57

44 Revolutionary Movements in Central America: A Comparative Analysis* Jeff Goodwin

* Source: Centre for Research on Politics and Social Organization, 1988, Working Paper 7.

Abstract This paper presents a comparative analysis of revolutionary movements in four Central American countries: Nicaragua, El Salvador, Guatemala, and Honduras. Unlike many analyses, it argues that Central America is not experiencing an undifferentiated regional crisis which is simply playing itself out at different speeds in different countries. Instead, it emphasizes the importance of the highly distinctive political contexts in each country, despite broad similarities at the regional level. Revolutionary movements in Central America have faced distinctive political challenges and have had quite different political fortunes. However, variations in socioeconomic factors—for example, poverty, class structure, land concentration, and international dependency—are not sufficient to explain the variable success of these movements. Nor is there evidence to indicate that prerevolutionary economic or coercive crises can account for this variation. Instead, this paper argues that differences in the political fortunes of revolutionary movements in Central America—here conceptualized as “proto-state” organizations—are best explained in terms of the specific nature of each state or polity and of the revolutionary movements themselves. Of particular importance, in this regard, are the ways in which states and revolutionary movements relate (or do not relate) to domestic social classes, on the one hand, and to international actors, on the other—although neither states nor revolutionary movements are simply the instruments or vehicles of a social class or foreign power. The Sandinista National Liberation Front (FSLN) successfully seized state power in Nicaragua by organizing a broad, multi-class opposition movement against an internationally isolated neo-patrimonial or “Sultanistic” dictatorship. By contrast, revolutionary movements in El Salvador and Guatemala have been less successful thus far since they have had to confront more bureaucratic authoritarian regimes which have undertaken limited political openings to the center within an international context less propitious for revolutionaries. Finally, none of several revolutionary groups in Honduras has become a significant political actor, due largely to the cyclical pattern of civilian and relatively moderate, even reformist, military rule in that country.

Revolutionary Movements in Central America: A Comparative Analysis

58

1. Introduction Social revolutions have been the subject of considerable speculation and debate since the emergence of modern social theory itself in the wake of the French Revolution. Indeed, the very word “revolution” only assumed its modern connotation at this time. The causes of social revolutions, in particular, have been the subject of considerable theorization among social scientists. Among the fundamental questions most often asked by sociologists of revolution are the following: What are the necessary and sufficient causes of social revolutions? Why do they happen exactly when and where they do? And why do revolutionaries win broad popular support and even seize state power in some countries, while they languish or remain completely marginal in others? Unfortunately, most empirical work on revolutions has not been methodologically designed in a manner most conducive to actually answering these questions. Two particular problems stand out. First of all, most such work only examines one or more successful cases of revolution. Very few social scientists, by contrast, have systematically examined “failed” revolutions, that is, major insurrections or civil wars in which revolutionary movements have been decimated or defeated outright; nor, for that matter, has there been much analysis of the failure of revolutionary movements to emerge at all in countries which might seem conducive to them (very poor countries, for example, with long-standing dictatorships or traditions of military rule). “Yet in all comparative logic,” notes Robert Dix, “this should be done in order precisely to understand the ingredients present in revolutionary triumph that were lacking in the (more numerous) failed attempts” (Dix 1984, p. 423). As Charles Tilly has suggested, “[A] lifelong study of successful revolutions alone would probably yield nothing but shaky hypotheses about the causes of revolution” (Tilly 1975, p. 489). Secondly, most comparative studies of revolution have examined revolutions which are of very different types and which have occurred during very different “worldhistorical” periods and in countries with very different social structures. In particular, comparativists have tended to jumble together revolutions based on more or less “spontaneous” rebellions or insurrections with those in which a revolutionary “vanguard” has consciously mobilized the populace for protracted armed struggle. Obviously, such an approach is likely to produce vague and overgeneralized conclusions about a phenomenon which merits more refined analysis. As a result of these two failings, I would argue, much of the sociological literature on revolutions is grossly oversimplified. Contemporary Central America, fortunately, offers fertile ground for addressing the fundamental questions about the causes of revolution—or, at least, one type of revolution—in a more adequate manner. Indeed, as one observer has suggested, Central America is a veritable “laboratory for theories of revolution” (Krumwiede 1984, p. 10). There are several reasons for this. To begin with, one finds in Central America not just one, but several revolutionary movements of essentially the same type to compare and contrast, namely, “national liberation” movements engaged in protracted armed struggles; all of these movements, moreover, are led by revolutionary socialists, as opposed, for example, to “mere” nationalists, religious fundamentalists, or counter-revolutionaries. In


59

addition, it should be noted, all of these movements emerged at the same “worldhistorical” moment, becoming important political forces during the late 1970s and early 1980s. Revolutionary movements, as the term is used here, are armed mass movements organized and led by professional revolutionaries, whatever their ideological orientation. These revolutionaries are “would-be state builders,” as Theda Skocpol has called them (Skocpol 1979, p. 165), who are attempting to monopolize state power—that is, monopolize the legitimate use of violence in a given territory—and to reconstruct the economy and social relations of that territory from that vantage point. Revolutionary movements, according to this view, are not simply social movements, nor even (or necessarily) ideologically radical social movements, but something qualitatively different: “proto-state” organizations which wish to replace or radically reorganize existing, incumbent regimes. The existence of a strong revolutionary movement in a given country implies a situation of “dual power,” that is, the existence of “essentially incompatible governmental organizations” (Trotsky 1961, Vol. I, p. 207). Minimally, “dual power” implies the existence of two armies loyal to competing political leaderships, although it may come to entail two separate and complete governments controlling more or less fixed territories within a given country. How revolutionaries are able or unable to create a situation of “dual power,” and thence proceed to monopolize state power, is the fundamental problem for sociologists of revolutionary movements. Central America is also useful for thinking about theories of revolution because the countries which comprise the region (excepting Costa Rica) share broadly similar populations, cultures, social structures, and histories. The region, in fact, was a single political unit, the Captaincy General of Guatemala, during three centuries of Spanish rule and was briefly a federal union, the United Provinces of Central America, after independence. Politically, Central American countries have generally been ruled over the past century or so by military governments allied to small landowning “oligarchies” and supported by the United States. Democratic rule in these countries, if it has appeared at all, has been short-lived and very unstable. During this same period, capitalist export agriculture has been the most dynamic economic sector in Central America and the primary source of wealth and political influence. Landownership has become increasingly concentrated, and landlessness, unemployment, and poverty are widespread throughout the entire region (see, e.g., Williams 1986). Indeed, “reactionary” capitalist development, to use Barrington Moore’s term (see Moore 1966, Ch. 8) has generated widespread social discontent throughout the region—as it undoubtedly has, in fact, throughout much of the Third World. It is not surprising, then, given the similarities of the four “core” countries in the region—Nicaragua, El Salvador, Guatemala, and Honduras—that most observers view recent events in Central America as part of a “regional crisis,” working itself out inexorably, albeit at different speeds and with particular nuances, in each country. These analyses speak of the regional disintegration of “liberalism” (see, e.g., Woodward 1984 and 1985, esp. Ch. 8–9, and Perez Brignoli 1985, esp. Ch. 6) or of “reactionary despotism” (see, e.g., Baloyra 1983 and Weeks 1986), meaning the symbiosis of agrarian capitalism and political despotism in the region. Most of these analyses, moreover, see social revolution as the likely or even unavoidable outcome of this crisis. Indeed, one


60

popular history of the region is entitled Inevitable Revolutions (LaFeber 1984). Revolution, writes James Dunkerley, certainly one of the more astute observers of the region, has been nurtured in the social conditions of states that comprise a region that is politically balkanised but of a relatively homogenous social fabric. If the peculiarities of Nicaragua under the Somoza dynasty proved to be the weakest link in the chain, and the small size, high population density and remarkable omnipotence of the local oligarchy in El Salvador provided an obvious sequitur, there is nothing so remarkable in the physiognomy of the other countries [of Central America] that would confer upon them some kind of immunity. [Dunkerley 1985, p. 3] Clearly, according to this view, the fact that the region is “politically balkanised” is less important than the fact that the region is “of a relatively homogenous social fabric.” But it is not at all clear, we shall argue, that El Salvador has provided “an obvious sequitur” to Nicaragua, let alone that the other countries in the region are future “links” in a chain of inevitable revolutions. The core countries of Central America, to be sure, are quite similar, and it is more than pure coincidence that strong revolutionary movements emerged in three of these countries at about the same time. However, there are important variations in the political success of revolutionary movements in the region which need to be explained. These variations tend to be underemphasized, in our view, by those who speak in terms of an undifferentiated regional crisis, even though these variations are extremely useful for thinking about the conditions that produce strong revolutionary movements and successful revolutionary movements. How, then, have revolutionary movements actually fared in the region? At one extreme, the Sandinista National Liberation Front (FSLN) successfully overthrew the Somoza family dictatorship (which had ruled Nicaragua since 1936) in July 1979 after a brief but bloody popular insurrection which left between forty and fifty thousand dead (Booth 1985, p. 183). Strong revolutionary movements also emerged in El Salvador and Guatemala at about the same time, but they have been unable to seize state power and seem unlikely to do so in the foreseeable future; instead, protracted civil wars are likely to drag on in those countries indefinitely. (More than sixty thousand people have been killed in the Salvadoran civil war of the past seven years, and as many as 100,000 may have been killed in Guatemala since the late 1970s.) The revolutionary movement in El Salvador, led by an alliance called the Farabundo Marti National Liberation FrontDemocratic Revolutionary Front (FMLN-FDR), is particularly strong, and a situation approximating “dual power” exists in many parts of the country (see, e.g., LeMoyne 1987a and Ramos 1987). By contrast, the revolutionary movement in Guatemala, led by the Guatemalan National Revolutionary Unity (URNG), has been decimated, although hardly destroyed, by a brutal counter-insurgency campaign (see esp. Barry 1986). In both El Salvador and Guatemala, significantly, revolutionary movements are no longer calling for the overthrow of incumbent regimes, but have instead proposed negotiations aimed at some sort of “power-sharing” arrangement. At the other end of the spectrum, finally, and


61

perhaps most interestingly, none of several revolutionary groups in Honduras—the poorest country in Central America—has developed into a significant movement. These differences, which are schematically presented in Diagram 1, are the focus of this paper

Diagram 1

Events, of course, are still unsettled. A strong revolutionary movement may yet emerge in Honduras, and it is hardly inconceivable that revolutionaries will eventually seize power in El Salvador or Guatemala. But this has not happened, and it demands an explanation. And, as Stephen Gorman has argued, “even if the Nicaraguan Revolution is followed in the near future by others in Guatemala or El Salvador, the process of political change in these latter two countries already demonstrates that the insurrectionary process there will likely follow a course quite different from the one that brought the Sandinista Front to power in Nicaragua” (Gorman 1984, p. 35). To put it another way, it may require the


62

“Nicaraguanization” of El Salvador and Guatemala for revolutionaries to succeed there; but the question, of course, is just precisely what this entails. Central America thus poses a fascinating problem for the comparative sociologist of revolution: How can we account for the different levels of success revolutionary movements have achieved in the region? This paper attempts to tackle this problem by addressing three specific questions. First, why did strong revolutionary movements develop in Nicaragua, El Salvador, and Guatemala when they did? Second, why has no significant revolutionary movement, by contrast, emerged in Honduras? And finally, why has the revolutiohary movement in Nicaragua seized power, while those in El Salvador and Guatemala have not? The answers to these questions, hopefully, will also shed light on more general notions about the causes of social revolutions, a subject I briefly take up by way of conclusion.

2. The Development of Revolutionary Movements in Nicaragua, El Salvador, and Guatemala Many commentators have attributed the emergence of revolutionary movements in Central America to the region’s incredible poverty, the current economic crisis, changes in class structure (especially growing landlessness among the rural population), and/or the region’s extreme external dependency, particularly on the United States. The available evidence, however, does not indicate any clear and direct relationship between these socioeconomic variables and recent political developments in the region, including the political fortunes of revolutionary movements. On the one hand, the data do not indicate that Nicaragua is systematically different from other countries in the region in ways which would be particularly conducive to revolution; it is unclear from these data alone, therefore, why Nicaraguan revolutionaries, alone among those in the region, should have triumphed. Nor, significantly, does available data indicate that Honduras is systematically “exceptional”; again, it is unclear from these data alone, therefore, why Honduras should be particularly immune to the emergence of a strong revolutionary movement. There is certainly no doubt that the “core” countries of Central America are among the poorest in the hemisphere. In terms of absolute poverty, however, Honduras has traditionally been the poorest country, per capita, in the region (see Table 1); as we have noted, however, no significant revolutionary movement has developed in that country. Nicaragua, on the other hand, where revolutionaries have seized power, was the best off country in the region, per capita, at least until that country was swept up into civil conflict in 1978. Absolute poverty, then, is a very poor guide indeed to recent political events. Income distribution, moreover, which is very unequal throughout the region, does not differ significantly among Nicaragua, El Salvador, and Guatemala (see Table 2). Income distribution is not so extremely unequal in Honduras, to be sure, which may partly account for the absence of a strong revolutionary movement in that country; however, there is cause to believe—for reasons we discuss below—that this is a spurious causal inference.


63

Table 1. GNP Per Capita, 1972–82 (Constant 1981 Dollars) El Salvador Guatemala Nicaragua 1972 1129 808 980 1973 1106 825 1032 1974 1269 857 1064 1975 1260 882 1044 1976 1255 881 1098 1977 1264 909 1169 1978 1110 942 1196 1979 818 895 1217 1980 910 794 1218 1981 946 759 1159 1982 878 717 1100

Honduras 605 606 599 565 584 614 635 651 650 624 595

Source: U.S. Arms Control and Disarmament Agency 1984, pp. 25, 28–29, 39 (from World Bank data).

Table 2. Income Distribution [1] and Per Capita Income [2], Circa 1980 (Constant 1970 U.S. Dollars) Nicaragua El Salvador Guatemala Honduras [1] [2] [1] [2] [1] [2] [1] [2] Poorest 20% 3.0 61.9 2.0 46.5 5.3 111.0 4.3 80.7 30% below the mean 13.0 178.2 10.0 155.1 14.5 202.7 12.7 140.0 30% above the mean 26.0 350.2 22.0 341.2 26.1 364.3 23.7 254.6 Richest 20% 58.0 1199.8 66.0 1535.0 54.1 1133.6 59.3 796.3 Source: Barry and Preusch 1986, p. 129. The annual data on per capita income presented in Table 1 also make it difficult to argue that the recent civil conflict in the region is the result of a sudden economic downturn or crisis after a period of growth—the classic cause of “relative deprivation,” or an indication, in Marxist terms, that productive forces are becoming “fettered” by class relations. An economic crisis does grip the region, but economic downturns only seem to have occurred with the onset of serious civil strife—in 1978 in Nicaragua, 1979 in El Salvador, and 1981 in Guatemala—and not before. Carlos Vilas, in fact, has argued that in Nicaragua “the fall of the Somocista dictatorship was the product of a revolutionary political crisis, which at a certain point in its evolution activated an economic crisis” (Vilas 1986, p. 92; emphasis in original). Significantly, moreover, an economic crisis has also beset Honduras since 1980; yet, as we have emphasized, no strong revolutionary movement has emerged in that country. Of course, to quote Trotsky, This circumstance can seem puzzling only to one who looks upon the


64

insurrection of the masses as Spontaneous’—that is, a herd-mutiny artificially made use of by leaders. In reality the mere existence of privations is not enough to cause an insurrection; if it were, the masses would always be in revolt. It is necessary that the bankruptcy of the social regime, being conclusively revealed, should make these privations intolerable, and that new conditions and new ideas should open the prospect of a revolutionary way out. [Trotsky 1961, Vol. II, p. vii] Some have argued that the “new conditions” which have made possible “a revolutionary way out” in Central America stem from changes in the class structure of Central American societies which have been brought about by the increasing concentration of land ownership by the agroexporting bourgeoisie or “oligarchy.” According to this view, the breakdown of the paternalistic ties characteristic of the traditional hacienda and the concomitant formation of a large landless and land poor rural population has class of people available for revolutionary mobilization (see, e.g., Paige 1983 and 1985). There is much to say for this argument, and the rural proletariat and “semiproletariat” have undoubtedly played an important role—although hardly the only role—in the revolutionary movements in the region. Once again, however, it does not appear that class structure and/or patterns of land tenure alone reveal the entire story about revolutionary movements in Central America. As the data in Table 3 reveal, landlessness and land poverty (defined as a holding of less than four hectares) are much greater problems in El Salvador and Guatemala than in Nicaragua and Honduras.1 And yet Salvadoran and Guatemalan revolutionaries, while more successful than their Honduran counterparts, have been less successful than revolutionaries in Nicaragua. Yet another view of Central America posits that revolutionaries are attempting, above all, to end the region’s extreme dependency on the capitalist “world-system,” particularly domination by U.S. “imperialism.” “Even more so than in Vietnam,” suggests Roger Burbach, “Central Americans are rebelling against decades of U.S. domination over all aspects of their society (military, economic, political, and cultural)” (Burbach 1984, p. 17). The self-described “national liberation” movements in the region would undoubtedly accept this view. However, the data on external dependency presented in Tables 4 through 8, like the other data presented thus far, do not explain why Nicaragua should be particularly susceptible, and Honduras particularly immune, to revolutionary upheaval. In recent years, to be sure, Nicaragua has been more dependent on export earnings than other countries in the region, but Honduras is more dependent, in this regard, than either El Salvador or Guatemala (see Table 4).

Table 3. Percentage Distribution of Rural Families by Size of Holding, 1970 Nicaragua El Salvador Guatemala Honduras (hectares) landless 33.8 26.1 26.6 31.4 under.7 1.5 24.4 15.0 10.3 0.7–4 24.2 36.2 42.3 24.1 4–7 7.9 6.2 6.9 11.9


7–35 35–350 over 350 total landless or insufficient land

18.1 13.5 1.0 100.0 59.5

65

4.9 2.0 0.2 100.0 86.7

7.4 1.4 0.4 1000 83.9

18.1 3.9 0.3 1000 65.8

Source: Weeks 1985, p. 112.

Table 4. Export Earnings as a Percentage of Gross Domestic Product (Constant Prices) El Salvador Guatemala Honduras Nicaragua 1960 24.0 20.4 14.6 20.3 1970 30.1 21.3 19.3 29.6 1975 33.7 24.4 21.2 26.4 Source : Weeks 1985, p. 52.

Table 5. Net Direct Foreign Investment (Millions of Dollars) Nicaragua El Salvador Guatemala 1960–64 27.1 31.4 54.7 1965–69 63.3 42.5 110.4 1970–74 65.3 42.6 132.8 1975–78 40.9 112.2 340.6 total 196.6 228.7 638.5

Honduras –6.1 46.6 24.1 35.3 99.9

Source: Weeks 1985, pp. 93–94.

Table 6. Net Direct Foreign Investment Minus Net Profit Transfers (Millions of Dollars) El Salvador Guatemala Honduras Nicaragua 1960–64 10.8 6.3 –11.1 –16.0 1965–69 –36.3 2.3 –40.6 –42.6 1970–74 –135.0 –21.5 –99.4 –75.1 1975–78 –240.9 –24.7 172.5 –203.8 total –401.4 –37.6 21.4 –413.3 Source: Weeks 1985, pp. 93–94. Since 1960, moreover, direct foreign investment has been much greater in Guatemala, and slightly greater in El Salvador, than in Nicaragua (see Table 5). Moreover, when profit transfers are subtracted from foreign investments, Honduras has experienced


66

greater decapitalization than any other country in the region (see Table 6). In addition, while external debt was a greater burden on Nicaragua than on the other countries in the region for most of the 1970s, the debt burden of Honduras overtook that of Nicaragua by the end of the decade (see Table 7). Finally, the data in Table 8 indicate that from the early 1950s until the late 1970s, more aid from the United States and U.S.-dominated “multilateral” banking institutions flowed to Guatemala and Honduras than to Nicaragua. It also seems clear that no “coercive crisis” involving the region’s armed forces suddenly and unexpectedly provided an opportunity for revolutionary mobilization in Central America. Revolutionary movements were not a response to, nor unleashed by, foreign wars or invasions, nor have there been significant defections of military personnel to revolutionary movements, even though these have been key factors in other revolutions (see, e.g., Skocpol (1979)). On the contrary, all of the revolutionary movements in the region have faced a rising tide of government repression, including the popular insurrection in Nicaragua right up until the overthrow of Somoza. As Table 9 indicates, military spending per capita rose significantly after 1973 in Nicaragua, after 1979 in El Salvador, and after 1981 in Guatemala.

Table 7. External Debt Service as a Percentage of Gross National Product El Salvador Guatemala Honduras Nicaragua 1972 3.8 .9 2.0 1.0 1974 3.5 1.5 .9 1.2 1976 4.3 1.6 .3 2.3 1978 5.2 1.0 .5 6.8 1980 3.7 1.2 .8 7.1 Source: Woodward 1985, p. 368.

Table 8. U.S. and “Multilateral” Assistance to Central America, 1953–1979 (Millions of Dollars) Nicaragua El Salvador Guatemala Honduras U.S. economic assistance 345.8 218.4 526.0 305.1 32.6 16.8 41.9 28.4 U.S. military assistance 469.5 479.2 593.0 688.0 “multilateral” banking aid total 847.9 714.4 1060.9 1021.5 Source: Petras and Morley 1983, p. 225. If poverty, economic crisis, landlessness, and dependency—which, it should be noted, are currently rather widespread throughout the Third World—cannot explain the development of revolutionary movements in Central America, what can? How was it that revolutionary groups became politically powerful and militarily effective during the late 1970s and early 1980s? What exactly happened in the 1970s, in particular, which transformed small revolutionary sects into strong revolutionary movements in Nicaragua, El Salvador, and Guatemala?


67

Table 9. Military Expenditures Per Capita (Constant 1981 Dollars) [1] and Armed Forces Per 1000 People [2], 1972–82 El Salvador Guatemala Honduras Nicaragua [1] [2] [1] [2] [1] [2] [1] [2] 1972 21 3.0 10 2.1 10 2.5 11 4.1 1973 16 2.9 9 2.1 8 2.3 10 4.0 1974 19 2.9 10 2.0 9 2.2 9 3.2 1975 22 2.3 10 2.0 12 2.1 11 3.8 1976 26 2.2 10 1.9 11 2.2 11 3.6 1977 32 2.6 10 1.8 14 2.2 12 3.5 1978 38 NA 13 2.3 14 2.1 14 3.9 1979 27 NA 13 2.3 14 2.1 15 3.9 1980 52 6.0 18 2.5 15 2.1 NA 3.7 1981 65 15.4 25 2.8 15 2.2 NA 3.8 1982 NA 27.8 29 5.4 18 2.3 NA 3.9 Source: U.S. Arms Control and Disarmament Agency 1984, pp. 25, 28–29, 39. As late as 1977, it should be noted, revolutionaries did not appear to represent a significant challenge to the status quo anywhere in Central America. The Sandinista Front had split into three warring “tendencies” with decidedly different ideas about how to bring about a revolution in Nicaragua, and the organization as a whole had fewer than one thousand guerrillas in 1977 (see Booth 1985, pp. 137–53, and Nolan 1984, esp. pp. 32–84). In May of that year, the dictator Anastasio Somoza Debayle could declare that “the Sandinistas are finished—divided and conquered by me” (cited in Goodsell 1982, p. 53). The various revolutionary groups in El Salvador and Guatemala, moreover, were politically and organizationally divided—bitterly so in El Salvador, where at least one internal dispute had resulted in bloodshed. None of these groups, furthermore, seems to have fielded more than a handful of armed guerrillas before the late 1970s. Revolutionary movements became powerful political forces in Nicaragua, El Salvador, and Guatemala during the late 1970s and early ’80s, we would argue, because attempts to redress social grievances through elections and peaceful political mobilization were greeted with electoral fraud and indiscriminate repression by military governments closely aligned with the landowning “oligarchies.” As a result, increasing numbers of people joined or collaborated with guerrilla groups, and even moderate political parties and civic organizations joined broad alliances which supported armed struggle. Of course, the Somoza dynasty in Nicaragua and military regimes in El Salvador and Guatemala had perpetuated their rule for decades through carefully managed elections; however, as political opposition grew in the 1970s, they increasingly relied on blatant electoral fraud and indiscriminate violence against political opponents. The Somoza family ruled Nicaragua for nearly a half century through its close control over the National Guard.2 The Guard itself was the principal legacy of the long U.S. marine occupation of Nicaragua—virtually continuous from 1912 to 1933—which was


68

originally occasioned by fears that a possible inter-oceanic canal through Nicaragua might end up in the hands of a European power or even Japan. Anastasio Somoza Debayle, who took over the Guard following the assassination of his father in 1956, was first “elected” president of Nicaragua in 1967 as the candidate of the Liberal National Party, the family’s political patronage machine. More than 600 people are believed to have been killed by the Guard at an opposition demonstration prior to the election, which Somoza won “by the traditional huge majority: this time 70% of the popular vote” (Black 1981, p. 44). In 1971, the U.S. ambassador helped strike a deal which “raised Conservative participation in the legislature to 40% and installed a Constituent Assembly which was to reform the Constitution…[to] pave the way for Somoza’s re-election in 1974” (Black 1981, p. 58). Executive power was formally ceded to a three-man junta, which included Conservative leader Fernando Aguero. “The dictator, needless to say, kept control of the National Guard and continued to exercise real power, representing Nicaragua as before as head of state in international forums” (Black 1981, p. 58). “After the Aguero-Somoza pact of 1971,” notes Booth, “revolutionary student groups began to attract more youths from upper-class backgrounds, both Liberal and Conservative” (Booth 1985, p. 111). The 1974 election, which Somoza won with the traditional overwhelming majority over the traditional hand-picked Conservative opponent, was boycotted by a number of dissident bourgeois politicians including Pedro Joaquin Chamorro, who by now had organised an opposition coalition, UDEL [the Democratic Union of Liberation]. For their pains, 27 leaders of the boycott were arrested and deprived of their political rights until March the following year. [Black 1981, p. 61] Although military rule in El Salvador dates back to 1932, the armed forces governed in an “institutional” manner after 1948, running their own candidates against loyal oppositionists and even creating their own political party.3 Significantly, no single family has controlled the armed forces in El Salvador in the manner of the Somozas in Nicaragua. In 1972, however, institutionalized military rule was seriously challenged when Christian Democrats, Social Democrats, and Communists joined to form the National Opposition Union (UNO). As it became apparent that UNO’s presidential candidate, the Christian Democrat Jose Napoleon Duarte, might win, the army prohibited the announcement of election returns and declared the military candidate the winner. A group of young officers, supported by Duarte, then tried to stage a coup d’état, but they were quickly defeated. Duarte was kidnapped by the army, tortured, and forced into exile. The mayoral and legislative elections in 1974 “were marked by even more blatant manipulation than had been evident two years earlier” (Montgomery 1982, p. 89). And in the 1977 presidential election, also contested by UNO, the military candidate “won handily in an election marked once again by fraud. Ballot boxes were already stuffed when the polls opened” (Montgomery 1982, p. 94). The National Police opened fire on a crowd of 50,000 UNO supporters who had gathered in San Salvador following the voting, killing at least four dozen people. Electoral fraud and violence have been no less severe in Guatemala, where the military


69

has also dominated politics in an institutionalized manner since a U.S.-backed coup in 1954, which ended a decade of democratic rule.4 In 1974, however, military rule was threatened, as in El Salvador, when Christian and Social Democrats joined to form the National Opposition Front (FNO), nominating General Efrain Rios Montt—who was known as one of the few “honest” officers in the country—for president. This moderate opposition group appeared to be headed for victory—again, just like the UNO coalition in El Salvador two years earlier—when the military began to delay voting results and ultimately declared the military candidate to be the victor. After speculation that Rios Montt might lead a coup to overturn the results, he was “quietly shipped off to a diplomatic post in Spain” (Handy 1984, p. 171). In 1978, the military’s candidate, General Romeo Lucas Garcia, won the presidency yet again in an election in which only 15 percent of eligible voters went to the polls. Violence against opposition political parties increased dramatically after General Lucas’s election. The leader of the Social Democratic Party (PSD), Alberto Fuentes Mohr, and of the social democratic United Revolutionary Front (FUR), Manuel Colom Argueta, were murdered in 1979. Vinicio Cerezo, secretary general of the Christian Democratic Party (DCG), himself the object of several assassination attempts, announced that 76 party members were killed between September 1980 and May 1981 alone (Handy 1984, p. 179). When the military candidate won yet again in the 1982 presidential election “in another patent example of electoral fraud, all opposition political parties once again joined in condemnation of the Lucas government,” which had overseen the voting (Handy 1984, p. 182). In addition to electoral fraud and violence, virtually any sort of peaceful political organizing and protest—particularly labor and peasant protest activities such as strikes or land occupations—was extremely dangerous in Nicaragua, El Salvador, and Guatemala during the 1970s, regardless of whether it was undertaken by revolutionaries or reformists, including sectors of the newly activist Catholic Church. Indeed, incumbent regimes made little distinction between reformists and revolutionaries, or between guerrillas and people merely suspected of collaborating with them. Rural and urban unions, student groups, Christian “base communities,” priests and catechists, and moderate political parties and opposition figures—as well as revolutionaries—were singled out for attack by the National Guard in Nicaragua and by the armed forces, paramilitary organizations, and allied “death squads” in El Salvador and Guatemala. (One Salvadoran death squad circulated a handbill which advised: “Be a Patriot—Kill a Priest!”) At times, in fact, virtually anyone of student age seems to have become regarded by the authorities as politically “suspect.” In Nicaragua, notes Booth, “Repression rose and fell according to the level of opposition activity—with major peaks during the late 1950s, the late 1960s, and from 1974 to the fall of the regime in 1979” (Booth 1985, p. 93). “In 1973 and 1974, when cotton workers, hospital workers, and banana workers struck, and when slum dweller organizations, market traders, and other groups staged protest actions, the Somoza government tried to crush these organizations by disrupting protest actions and arresting the leaders” (Williams 1986, p. 167). An attempt to form a union of rural workers in Chinandega resulted in the deaths of 300 people at the hands of the National Guard (Paige 1985, p. 107). Somoza declared a state of siege in December 1974, after a group of Sandinistas took a number of the regime’s luminaries hostage at a private party, winning


70

the release of a number of political prisoners and the broadcast and publication of two lengthy communiques. During the state of siege, The Guard sought to disrupt the FSLN, to capture and kill as many guerrillas as possible, and to take reprisals against any campesinos who, willingly or not, supported the insurgents. The army often appropriated the resources of peasants without compensation. Whole regions underwent “agrarian reform”— population relocation to break up foci of guerrilla support. Suspected guerrilla collaborators suffered horrifying tortures. The Guard often murdered not only FSLN collaborators, but their families as well. Estimates of the total number of peasants killed by the Guard begin at three thousand. [Booth 1985, pp. 94–95] Booth adds that when “in the late 1970s the government began to suspect the FSLN of recruiting mainly teenage males, the Guard seized many—often barely more than children—from Managua’s streets and executed them on the shore of [Lake Managua]” (Booth 1985, p. 95). In El Salvador, the paramilitary force known as ORDEN (“order” in Spanish), which was established in the mid-1960s with U.S. assistance, stepped up attacks on union and political activists among peasants and rural workers during the 1970s. Father Rutilio Grande, a well-known priest who had set up “base communities” around the city of Aguilares, where the Christian Federation of Salvadoran Peasants (FECCAS) was strong, was murdered in March 1977. FECCAS and the Union of Rural Workers (UTC) organized a series of peaceful land invasions the following month, demanding that landowners rent unused lands at affordable prices: Two thousand troops came with helicopters and armoured troop carriers and took over the entire region, surrounding Aguilares in the early hours of 19 May. A peasant sleeping in the church tried to ring the church bell to alert the town but was shot and killed…. Soldiers moved through the town, ransacking houses, and later announced that one soldier and six civilians had been killed in an armed encounter. Eyewitnesses corrected the numbers: at least 50 townspeople had been shot dead and hundreds taken away by the army. [Pearce 1986, p. 168] Another activist priest, Alfonso Navarro, was killed shortly after the events in Aguilares. All in all, ten priests and a seminarian were assassinated between March 1977 and June 1981, and at least sixty priests were expelled or forced into exile. In May 1979, one of the broad political fronts or “popular organizations” recently formed in the country occupied the metropolitan cathedral in San Salvador, demanding the release of imprisoned members of the group. “[President] Romero’s reply was unambiguous: troops fired on the demonstrators outside the cathedral, and some 25 people died in the mad scramble to reach the precarious safety of the church” (Dunkerley 1985, p. 127). In all, nearly 800 people were killed in 1979. And on January 22, 1980, when the “popular organizations” held the largest mass demonstration in Salvadoran history, with more than 200,000 people gathered in San Salvador from all over the country, security forces fired on the crowd, leaving 49 dead and hundreds wounded (Montgomery 1982, p. 129).


71

State violence in Guatemala, although less reported in the United States than that in Nicaragua or El Salvador, has probably been the worst in the region. In the predominantly Indian department of El Quiche alone, 163 co-operative or village leaders were killed between 1976 and 1978 (Handy 1984, p. 244). Over 100 men, women, and children were massacred in the town of Panzos on May 29, 1978, after gathering to protest evictions from nearby lands (see Aguilera 1979). Amnesty International documented some 615 cases of “disappearance” between mid-1978 and November 1980; and the same organization documented the cases of nine Catholic priests who had either “disappeared” or been killed by security forces between April 1980 and July 1981 alone (Amnesty International 1987 (hereafter, AI), pp. 3, 24). In January 1980, a group of peasants from El Quiche peacefully occupied the Spanish embassy in Guatemala City, demanding that military abuses in that department be investigated: Despite the demands by the Spanish ambassador that the peasants inside the building not be attacked, the police stormed the building and 39 people were killed including a former vice-president and a foreign minister of Guatemala, who were both in the building to talk with the peasants. Only the ambassador and one peasant survived. The peasant was later kidnapped from the hospital where he was recuperating from his injuries. After the attack the Spanish government broke off all diplomatic relations with Guatemala and withdrew its ambassador. [Handy 1984, p. 245] A few months later, 25 leaders of the National Confederation of Labor (CNT) were abducted from a meeting at their headquarters in Guatemala City; they were never seen again (see AI 1987, pp. 118–19, and Black 1984, pp. 90–91). Clearly, Jeane Kirkpatrick’s suggestion that “traditional autocrats” such as those who rule Central America “do not disturb the habitual rhythms of work and leisure, habitual places of residence, habitual patterns of family and personal relations” is not a particularly apt characterization of Central America in the 1970s (Kirkpatrick 1986, p. 33). Indeed, given the sort of violence that even moderates and reformists have encountered, it is hardly surprising that many people have come to view armed struggle as the only viable “way out” in those societies. In fact, revolutionary groups in Nicaragua, El Salvador, and Guatemala—or political-military organizations, as they are called (orqanizaciones politicas-militares or OP-Ms)—were formed by dissidents from electorally oriented political parties, particularly Communist and Christian Democratic parties, who were frustrated by continual electoral fraud and political repression and viewed armed struggle as the only way to topple dictatorial regimes. Communist parties, interestingly, have not been the principal vehicle of revolutionary movements in Central America, as they were in Asia, for example, following World War II; however, the Salvadoran CP and a faction from the Guatemalan CP did come to adopt a strategy of armed struggle after the worst repression had begun in those countries. In all three countries, radical student groups and peasant and labor unions became the main source of cadre for these organizations. Like members of revolutionary “vanguards” elsewhere, these cadres were disproportionately middle class in social origin. These revolutionary groups, which originally included no more than a handful of


72

members, profited from the repression described above by providing a protective cover for at least some of the victims—or fearful potential victims—of this repression. Indeed, active collaboration with revolutionaries, which might once have invited persecution, became for many the only alternative to it. “The arbitrariness of the exercise of politicalmilitary power by the dictatorship,” as Vilas has written of Nicaragua, “the indiscriminate—and finally, genocidal—character of the repression…converted active rebellion and participation in the revolutionary struggle into a defensive question—life or death” (Vilas 1986, pp. 112–13). In this way, state violence swelled the ranks of the guerrillas, particularly in relatively isolated regions—north-central Nicaragua, northern El Salvador, and the western highlands of Guatemala—where guerrillas could operate more freely than in cities. The Sandinista National Liberation Front (FSLN) was founded in 1961 by political activists who were inspired by the example of the Cuban Revolution. The Front’s prime mover, Carlos Fonseca Amador, was a former member of the so-called Nicaraguan Socialist Party (PSN), the country’s Moscow-aligned Communist party. After a string of defeats in the 1960s, the Sandinistas built an organizational network of collaborators (or “chain,” as it was called) among peasants in the north-central region of the country. Many of these peasants were adversely affected by land enclosures by large cattle ranchers in the area (see Williams 1986, pp. 129f). The Sandinistas later recruited rural workers on the Pacific coast; they formed the Association of Rural Workers (ATC) in March 1978, which “was able to convert itself into a powerful force of the FSLN, not only in building the armed struggle but in organizing political action by workers and peasants in rural areas” (cited in Paige 1985, pp. 108–9). Finally, during the insurrection of 1978–79 the FSLN was also able to recruit a variety of petty producers and unsalaried workers in the cities—artisans, food vendors, carpenters, shoemakers, and the like. According to Vilas’s study of those killed in the insurrection, these “tradespeople” were the principal social force of the insurgency (Vilas 1986, p. 119). “During 1978–1979,” concludes Booth, “the FSLN’s growth rate skyrocketed”: Training camps in Costa Rica, Honduras, and Nicaragua swelled into substantial operations. The overall number of FSLN troops thus ballooned from between five hundred and a thousand in early 1978 to nearly three thousand by late 1978 to around five thousand by July 1979. To incorporate the many irregular volunteers, the FSLN also ran short combat-training sessions for citizens when Sandinista columns occupied a neighborhood. These trained volunteers were the core of the popular militia. [Booth 1985, pp. 149–50] The Farabundo Marti National Liberation Front (FMLN) in El Salvador, an umbrellagroup formed in 1980, “has two main sources: radicalized religious activists, and the Salvadoran Communist Party [PCS]” (Leiken 1984, p. 115). The largest guerrilla group in the FMLN, the Popular Forces of Liberation (FPL), was formed in 1970 by dissidents from the PCS, led by long-time secretary general of the party, Salvador Cayetano Carpio; the “final straw” leading to this group’s departure was the PCS’s support for the brief 1969 war with Honduras. The second most important OP-M, the People’s Revolutionary Army (ERP), was founded in 1972 by activists from the Communist and Christian


73

Democratic youth movements. A third group, the Armed Forces of National Resistance (FARN), broke from the ERP in 1975 following the execution of Roque Dalton, a ERP member and El Salvador’s leading contemporary poet. (Dalton, interestingly, was charged with being a “Soviet-Cuban and CIA double agent” by the ERP’s military staff.) According to Montgomery, many within the ERP faction which formed the FARN were Protestants, and at least two were Baptist ministers (Montgomery 1982, p. 121). Later, the PCS and the small Revolutionary Party of Central American Workers (PRTC) also formed guerrilla groups. The principal social base of these OP-Ms has been the “semiproletariat” which inhabits the sparsely populated northern part of the country where the guerrillas have been most active. These poor peasants own or rent a tiny parcel of land, but must supplement their incomes by working on cotton, coffee, or sugar estates for part of the year. However, landless laborers and better-off “middle peasants” have also joined or collaborated with the guerrillas (see Cabarrus 1983, pp. 365–67, and Pearce 1986, pp. 150–51). It is estimated that the FMLN fielded between eight and eleven thousand armed guerrillas by the early 1980s (see Dunkerley 1985, p. 221, and Halloran 1987). The Guatemalan National Revolutionary Unity (URNG) consists of four main revolutionary groups, all descending one way or another from the guerrilla groups active in the 1960s which were led by former military officers and the Guatemalan Labor Party (PGT), Guatemala’s Communist party. The largest guerrilla group in Guatemala, the Guerrilla Army of the Poor (EGP), was formed in 1972 by radical Christians and survivors of the “Edgar Ibarra” Guerrilla Front, “one of the strongest guerrilla forces in the continent” that was decimated in the late 1960s (Black 1984, pp. 68, 106). The second most important group, the Organization of People in Arms (ORPA), was launched publicly in 1979; its leaders split off from the Rebel Armed Forces (FAR), another group founded in the 1960s and close to the PGT. The FAR is still active, having been rejuvenated in the mid-1970s by many disaffected Christian Democrats. Finally, the Guatemalan Labor Party-National Leadership Nucleus (PGT-NDN) also fields a small guerrilla force; it split from the PGT’s Central Committee in 1978 over the question of armed struggle (see Black 1984, Ch. 4–5, and Handy 1984, Ch. 11). Unlike their predecessors of the 1960s, the guerrilla groups of the 1970s, especially the EGP and ORPA, have focused their political work on Guatemala’s indigenous population in the central and western highlands. By the early 1980s, there were about six to eight thousand armed guerrillas in the highlands, with an estimated 260,000 collaborators (Carmack 1985, p. 2). As in El Salvador, the guerrillas were particularly strong among the “semiproletariat,” although they also drew support from the rural proletariat and middle peasants (see Paige 1983, p. 728, and Carmack 1985, pp. 21–24). As in Nicaragua and El Salvador, Military repression of reformist efforts and community organization led…to widespread defensive mobilization: to protect themselves from the landowners and the military, the Indian communities began to organize for the purposes of self-defense. Victimized by the military and lacking alternatives for economic self-improvement, the Indians became increasingly willing to turn to armed opposition, swelling the ranks of the guerrilla movement. By and large, their


74

resistance was a result of state repression, not its cause. [Trudeau and Schoultz 1986, pp. 37–38] In addition to swelling the ranks of the guerrillas, electoral fraud and state violence have also led moderate groups, including many middle-class organizations, into open political alliance with revolutionaries. These groups have been most instrumental in helping to provide material aid, political legitimacy, and diplomatic support for the armed opposition in Central America. Perhaps even more importantly, by refusing to participate in “normal” politics, they contribute to the perception that armed struggle is the only viable alternative. In the early summer of 1978, the Sandinista Front organized the United People’s Movement (MPU), which eventually included 22 labor and political organizations, including the country’s main labor and student federations, the teachers’ union, the principal women’s organization, and two left-wing parties. In February 1979, the Independent Liberal Party and the Popular Social Christian Party, among other organizations, joined with the MPU to form the National Patriotic Front (FPN) (Booth 1985, pp. 154–55, 316). “Given its broader political base,” notes Booth, “the FPN’s formal program contained somewhat less farreaching reform proposals than that of the MPU…. [H]owever, the FSLN justified the watered-down FPN program as necessary to unite disparate forces seeking a common anti-Somoza objective” (Booth 1985, p. 155). In El Salvador, each of the OP-Ms was affiliated with a “popular organization” which brought together a variety of labor and political groups. “Government repression solidified the links between the popular organizations and the guerrilla groups,” notes Leiken. “The former came to constitute recruiting ground for the latter” (Leiken 1984, p. 117). The FPL was closely tied to the Popular Revolutionary Bloc (BPR), the largest of the popular organizations with some 60,000 members, which included two large unions of peasants and rural workers (FECCAS and UTC), the country’s main teacher’s union, a shanty town dwellers’ organization, and student groups. The ERP and RN were closely tied to the Popular Leagues of February 28th (LP-28) and the United Popular Action Front (FAPU), respectively. These organizations also included a variety of labor unions and student groups. And the Communist Party and PRTC were aligned with the National Democratic Union (UDN) and the Movement of Popular Liberation (MLP), respectively. In April 1980, these five popular organizations (the BLP, LP-28, FAPU, UDN, and MLP) would join with the social democratic National Revolutionary Movement (MNR), the Popular Social Christian Movement (MPSC), and the Independent Movement of Professionals and Technicians of El Salvador (MIPTES) to form the Democratic Revolutionary Front (FDR), the civilian wing of the armed opposition. Finally, in Guatemala, the guerrilla groups were closely tied to two broad political alliances, the Democratic Front Against Repression (FDRC) and the Popular Front of January 31st (FP-31). The FDRC brought together over 170 organizations, including the National Committee of Labor Unity (CNUS), the Social Democratic Party (PSD), and the social democratic United Revolutionary Front (FUR). The FP-31 included, among other groups, the Peasant Unity Committee (CUC), which had developed very close ties with the EGP, and an important radical Christian organization (Black 1984, pp. 107–9). In sum, small revolutionary groups in Nicaragua, El Salvador, and Guatemala became


75

powerful revolutionary movements not because of social grievances, class struggle, or imperialism per se, but because the brutal and indiscriminate violence with which authoritarian regimes met the political mobilization of the 1970s “backfired,” driving more and more individuals and groups into revolutionary organizations or into political alliances with them. In other words, the exclusionary nature of the Nicaraguan, Salvadoran, and Guatemalan regimes and their indiscriminately violent reaction to political mobilization channeled that mobilization into a revolutionary direction. State violence against popular organizations allowed revolutionary groups to recruit growing numbers of members and collaborators, especially in relatively inaccessible rural areas, and to strike alliances with labor, peasant, and middle-class groups and reformist political parties. This explanation, paradoxically, is confirmed by the case of Honduras, where a strong revolutionary movement has not developed.

3. Why Is There No Revolution in Honduras? While powerful revolutionary movements were developing in Nicaragua, El Salvador, and Guatemala, neighboring Honduras was remarkably quiescent.5 “In a region where the fires of revolution burn white hot,” notes Steven Volk, “Honduras has been a model of stability, a bizarre anomaly” (Volk 1983, p. 225). Indeed, the absence of a significant revolutionary movement in Honduras poses something of a puzzle. According to U.S. military sources, four guerrilla groups in Honduras have a combined membership of only about 600 people (Halloran 1987). Yet, as we have seen, Honduras is the poorest Central American country; like its neighbors, it is experiencing a grave economic crisis; landlessness and land poverty, while not as severe as in El Salvador or Guatemala, rival that of pre-revolutionary Nicaragua; and the country is every bit as externally dependent as the other countries in the region. Honduras, moreover, the classic “banana republic,” also has a particularly sordid history of domination by U.S. corporations as well as a long tradition of military rule. How, then, has Honduras avoided the civil conflict which has befallen its neighbors? To paraphrase Sombart, why is there no revolution in Honduras? The crucial difference between Honduras and its neighbors is what might be called the “semi-openness” of the Honduran polity. Since the Second World War, Honduras has alternated between civilian and relatively moderate military rule; the country even witnessed a brief period of “military reformism” in the early 1970s, during which a significant land reform was enacted (see esp. Salomon 1982, and Posas and del Cid 1983, Ch. 3). As a result of this “semi-openness,” reformist political currents (particularly within the Liberal Party) as well as labor and peasant unions have been unusually free, by Central American standards, to operate and organize within the country, even under military rule. Indeed, “in the 1960s, Honduras gave birth to the most militant and, before long, the bestorganized peasant movement in Central America” (Volk 1983, p. 215). James Morris has argued that this alternation between reformist and conservative rule has produced a “cycle of frustration” in Honduras (Morris 1984a, Ch. 9). Paradoxically, however, this cycle has not produced, but may very well account for the absence of, a strong revolutionary movement. Intermittent reform and the impermanence and relative mildness of military rule seem to have led most leftists in Honduras to eschew a strategy


76

of armed struggle. Even more importantly, they have discouraged ordinary people from joining or collaborating with the guerrilla groups which have sprung up; by all accounts, these groups have not acquired any significant popular support. The “semi-openness” of Honduran politics essentially dates from the general strike in 1954 against the huge U.S. banana companies in Honduras, “the most extensive industrial and political action in Honduran history” (LAB 1985, p. 35). Although the strike resulted in very small wage increases and prompted the banana companies to lay off thousands of workers, the strikers did win their principal demand: the legal recognition of trade unions (on the strike and its aftermath, see esp. MacCameron 1983). After a brief period of military rule, furthermore, the reformist Ramon Villeda Morales of the Liberal Party was elected president in 1957: Villeda set up a public-works program to provide the nation with some basic infrastructure (as of 1950 the capital city had no paved roads) and to employ some of the thousands of dismissed banana workers. He also ordered a number of land-colonization plans, distributing lands in regions of low population density to peasant families…and creating new areas of colonization…. More important, in September 1962 Villeda pushed through the country’s first agrarian reform law. [Volk 1983, p. 210] In 1963, Villeda was overthrown by the army, led by Oswaldo Lopez Arellano, to the delight of landowners and right-wing forces, but Lopez “turned out to be a political chameleon”; in 1967, with land occupations by peasant groups on the rise, “Lopez revitalised the agrarian reform, as part of his own search for a new political base” (LAB 1985, p. 54, 59). The Honduran government began to harass and evict Salvadoran squatters from public lands in order to proceed with Villeda’s agrarian reform without seizing the property of large landowners. El Salvador’s subsequent invasion of Honduras, however, was nothing short of disastrous for the Honduran military. Only the intervention of the Organization of American States (OAS) saved the army from complete defeat. “El Salvador’s more professional and better-trained troops advanced deep into Honduran territory before the OAS threatened an economic boycott” (LAB, p. 61). “For Honduras,” notes Volk, “the war was nothing short of monumental”: Defeat revealed a command structure that was inefficient, backward, and corrupt. And defeat paved the way for reform in Honduras. The Honduran officer corps became increasingly influenced by younger officers who drew their inspiration from the military reformers in Peru and Panama. Many of the older, corruption-tainted officers were purged. [Volk 1983, pp. 216–17] Following a brief civilian government, reformists in the military seized power in 1972, led by none other than General Lopez Arellano, and enacted another agrarian reform law, more thorough-going than the one enacted a decade earlier by Villeda, the very man whom Lopez had ousted. Peasants were granted the right to occupy public lands, and landowners were obliged to either rent or cultivate unused land. In addition, the size of farms liable to expropriation was increased to 500 hectares (see LAB, pp. 63–64). Ruhl


77

estimates that only about 22% of landless and land poor families in Honduras in the mid1970s directly benefited from the reform. But agrarian reform was also “very important symbolically” since “the program demonstrated the continued flexibility and reform potential of the Honduran government and fostered an ‘incrementalist’ policy orientation among the peasant organizations”: The fact that campesinos could win disputes over land titles…and force “stolen” land to be returned by landlords was extremely significant. Such outcomes would have been almost unimaginable in Guatemala, El Salvador, or Nicaragua. The association of the military with land reform during the late 1960s and 1970s also created a much more progressive image for Honduran soldiers than for their counterparts in neighboring countries and demonstrated clearly that the armed forces were not under the control of the Honduran rural oligarchy. [Ruhl 1984, p. 55] The military reformists were pushed out by more conservative officers in the late 1970s, but Honduras has returned to civilian rule, once again, in the 1980s. Ironically, the return to civilian rule coincided with a rise in human rights abuses against labor and peasant activists, generally attributed to conservative sectors of the army associated with General Gustavo Alvarez Martinez, head of the armed forces. In March 1984, however, Alvarez was deposed by younger officers and forced into exile. “The fall of Alvarez Martinez,” notes Perez Brignoli, “is an event of great significance. It implies a setback, although perhaps only temporary, of the most repressive military sectors” (Perez Brignoli 1985, p. 143). What accounts for this “semi-openness” of the Honduran polity? Significantly, Honduras managed to avoid a Nicaragua-style U.S. occupation, in part, no doubt because of the absence of any possible route for an inter-oceanic canal. Perhaps even more important, however, has been the absence of a strong landowning oligarchy in Honduras along the lines of El Salvador or Guatemala. As a result of these two factors, moreover, the Honduran military has displayed greater political autonomy and pluralism than its counterparts elsewhere in the region. The absence of a strong oligarchy in Honduras dates back to the nineteenth century. As opposed to El Salvador and Guatemala, Coffee did not assume a major role in the Honduran economy and the ejidos [common lands] remained intact. Coffee failed to reorient Honduran agriculture for several reasons including the lack of a high proportion of the rich, volcanic soils conducive to coffee production and the preoccupation of Honduran governments both with intra-elite political conflicts and with encouraging foreign silver mining and other ventures…. Because of these various factors, no strong, cohesive coffee oligarchy developed in Honduras to dominate the government and to absorb the ejidal lands. [Ruhl 1984, p. 36] Among the foreign ventures encouraged by Honduran governments were the huge U.S.owned banana plantations of the north coast. Indeed, no Central American country was so


78

thoroughly dominated by U.S. capital as was Honduras during the first half of the twentieth century. However, “The concessions made by the government to foreign investors,” notes Morris, “though perhaps compromising latitude of action and control, did result in ever-increasing revenues for Honduran administrations”: Capturing control over the government then became an attractive goal for the various political factions. To gain control over the state apparatus provided access to wealth, patronage, and personal advancement. No unified social class emerged in Honduras that had a firm grasp on any other solid economic base. [Morris 1984b, p. 197] Partly because of this “absent oligarchy,” those who control the Honduran state control a relatively large domain of public land; much of this land, furthermore, was reclaimed by the state through the eviction of Salvadorans in the late 1960s. Significantly, state control of this land has made it possible for land redistribution to take place without completely relying on the expropriation of the large landowners who do exist. It is also likely that the absence of an entrenched oligarchy—as well as agrarian reform and the relative freedom of labor and peasant unions—accounts for the fact, noted earlier, that income distribution is less unequal in Honduras than elsewhere in the region. “With little local oligarchy to protect,” moreover, “Honduras did not develop a professional military tied to concentrated economic interests, and the military did not view its function as making more or less permanent war on its own people” (Shepherd 1986, p. 127). The army’s sponsorship of agrarian reform in the late 1960s and again in the 1970s, as well as the recent ouster of General Alvarez Martinez, demonstrates the military’s autonomy from the most conservative sectors of Honduran society. Ironically, the conservative National Party, which was ousted in the 1972 military coup, was one of the more insistent voices during the 1970s calling for a return to civilian rule. In summary, we would agree with Jorge Arturo Reina that the principle reasons why there is no violence for now [in Honduras] are: 1. The social reforms enacted by the government of Villeda Morales. 2. The exodus of Salvadorans due to the 1969 war. 3. The reforms enacted in the first stage of the military government led by General Lopez Arellano [i.e., 1972–75]. 4. The hope that the electoral road can bring about changes peacefully. 5. The relative tolerance of military rule compared to the reactionary military regimes elsewhere in the region. [Reina 1981, p. 46] The cycle of civilian and military rule in Honduras, however frustrating for those committed to the principle of civilian government, seems to have prevented the emergence of a significant revolutionary movement in that country.

4. Patterns and Prospects of Insurgency Although, as we have seen, mass-based revolutionary movements did develop in


79

Nicaragua, El Salvador, and Guatemala in the late 1970s and early ‘80s, these movements have had very different political fortunes. To date, only the movement in Nicaragua has actually seized power, successfully toppling the Somoza dictatorship in July 1979. In El Salvador, the rebels’ “final offensive” of January 1981 failed to oust the incumbent regime, although something of a military and political stalemate has persisted in that country ever since. And in Guatemala, a brutal counter-insurgency campaign from 1982 to 1984 decimated the guerrillas and has largely cut them off from their mass base, although it has not eliminated them. These patterns underscore the fact—as do post-war insurgencies in Greece, Malaya, and the Philippines—that the factors which create strong revolutionary movements are not necessarily the factors that create movements which actually succeed in seizing power. Why is it, then, that the Sandinistas have been able to seize state power, while the rebels in El Salvador and Guatemala have not? The triumph of the Sandinistas in the 1978–79 popular insurrection against Somoza was the result of a number of factors which were mutually reinforcing: the failure of any political “opening” in the country; growing elite opposition to Somoza; the inability of the National Guard to contain, let alone halt, the rising tide of armed insurrection; and the growing international recognition of the Sandinistas and the simultaneous international isolation of Somoza. After the National Guard had contained a number of urban uprisings in September 1978, Somoza released 350 political prisoners and agreed to negotiate with the moderate opposition—which was represented by a group called the Broad Opposition Front (FAO)—in talks mediated by an OAS team composed of representatives from the United States, Guatemala, and the Dominican Republic. The FAO, which had been formed in March 1978, included the UDEL, formed by Pedro Joaquin Chamorro, and the Nicaraguan Democratic Movement (MDN), a group of businessmen and professionals led by the industrialist Alfonso Robelo Callejas. Chamorro’s assassination in January 1978, which was widely attributed to the dictatorship and which sparked riots and a business strike, led to the FAO’s formation. Somoza, however, resolutely refused to step down before his term expired in 1981 in order to transfer power to a moderate interim government, and the FAO broke off negotiations in late November 1978. Shortly thereafter, the FAO accepted direct talks with Somoza, but these talks also collapsed in late December. “The dictator misread the situation,” notes Chavarria, “and tried to resolve a societal crisis by adopting exclusively military means” (Chavarria 1982, p. 37). Significantly, no faction within the National Guard, which had been closely controlled by the Somoza family for more than forty years, was able to depose Somoza and initiate a political opening of its own. Eighty-five members of the Guard were arrested in late August 1978 for plotting against Somoza (see Booth 1985, p. 165), but the Guard’s officer corps proved remarkably loyal to the dictator during the course of the insurrection. The breakdown of negotiations not only accelerated the development of elite opposition to Somoza, but also led much of the elite to join ranks with the Sandinistas. Nicaraguan businessmen had long complained of Somoza’s “disloyal competition” (competencia desleal), that is, the use of political power and connections to control business opportunities in the country. Torres Rivas, among others, has noted the “patrimonial” or “Sultanistic” nature of Somoza’s rule: “a structure of loyalties was created which recalls the patrimonial game of feudal domination. The administration of


80

public matters, in this particular context, is the government of private affairs” (Torres Rivas 1983, p. 137). One consequence of the “Sultanism” of the regime was the remarkable personal loyalty of the National Guard to the Somoza family. Another consequence was bourgeois opposition to Somoza, which grew rapidly after the 1972 earthquake which destroyed much of Managua (on divisions between the Somoza and the “non-Somocista bourgeoisie,” see especially Wheelock Roman 1980, Ch. 6). Somoza was not only suspected of embezzling huge sums of relief assistance, but also “cornered the reconstruction of Managua. His company ESPESA took charge of demolition work; Inmuebles SA of real estate speculation; a host of other companies, generally with a monopoly, took on contracts for concrete, building materials, metal structures, roofing, asbestos, and plastic” (Black 1981, pp. 59–60). When the Sandinistas announced their “final offensive” in late May 1979, the FAO joined the FPN in calling for a general strike. And in June 1979, the FAO and the Superior Council of Private Enterprise (COSEP) endorsed the Governing Junta of National Reconstruction formed by the FSLN, which included Alfonso Robelo of the MDN and the widow of Pedro Joaquin Chamorro. Needless to say, the breakdown of negotiations between Somoza and the moderate opposition only contributed to the perception, both in Nicaragua and abroad, that supporting the armed struggle of the Sandinistas was the only way to depose Somoza. Significantly, the three “tendencies” which had developed within the FSLN during the 1970s managed to overcome their differences by early 1979, enhancing the Front’s ability to wage a co-ordinated national struggle. In fact, the National Guard, which was doubled in strength from a mere seven thousand to fourteen thousand troops in 1978, proved unable to contain the growing popular insurrection. Although the FSLN never had more than five thousand guerrillas, its “numerical disadvantage was greatly offset by its massive popular support in organization, logistics, and combat. In almost every battle in a populated center, the Front found its members multiplied severalfold by volunteers” (Booth 1985, p. 176). The FSLN’s strategy of spreading the Guard thin by organizing insurrections in a number of cities and towns throughout the country eventually paid off. The National Guard, although it never seems to have lacked weapons or supplies, became unable to relieve or resupply besieged Guard headquarters around the country (Chavarria 1982, p. 36). In retrospect, the Guard probably could have contained the insurgency only if some foreign power had been willing to “save” Somoza by organizing and financing a massive expansion of the Guard and/or by directly intervening militarily. But this was not forthcoming. Indeed, the growing military strength of the FSLN was very much related to the growing international recognition of the Sandinistas and the concomitant international isolation of Somoza. The ability of the FSLN to operate from base camps in Costa Rica and, to a lesser extent, in Honduras, made it much easier for the Sandinistas to prosecute the war against Somoza (see Seligson and Carroll 1982). As the Front grew, its camps in these countries had no difficulty acquiring new arms. “This was especially true of Costa Rica after 1978, when the government of President Rodrigo Carazo not only permitted but encouraged arms shipment to the FSLN” (Booth 1985, p. 152). Weapons from a number of Latin American governments went first to Panama and were then shipped to the Sandinistas by air or ground through Costa Rica. It has also been revealed that


81

in May-July 1979 twenty-one flights by Costa Rican civilian transport planes brought a total of 320 tons of arms from a Havana military base to the FSLN via the airfield at Costa Rica’s northern town of Liberia. The cost to the FSLN of this assistance was Costa Rica’s retention of half the arms. Virtually every known model of light combat arm manufactured in the United States and Western Europe found its way into the FSLN armory. Light civilian planes flown from Costa Rica even dropped a few bombs on National Guard positions late in the war. [Booth 1985, p. 153] The Somoza regime became increasingly isolated internationally during the course of the insurrection. Costa Rica had broken diplomatic relations with Somoza as early as November 1978 over a border incident. Mexico broke relations in May, and on June 22, the OAS rejected a U.S. proposal to send an inter-American “peace-keeping” force to Nicaragua; the following day the OAS, in a 17-to-2 vote, demanded Somoza’s unconditional resignation. Somoza later blamed his downfall on the OAS resolution (see Booth 1985, p. 181). The Carter administration, for all its equivocations, was unwilling to intervene in Nicaragua unilaterally in order to rescue Somoza or to prevent a Sandinista victory (see LeoGrande 1982). By mid-July, the Sandinistas controlled all the major roads to Managua. During the night of July 16–17, Somoza submitted his resignation and fled the country with his family, the remnants of the National Guard general staff, and the leadership and congressmen of the Liberal National Party. At this point, not surprisingly, the National Guard began to disintegrate. Dozens of officers fled the country, most to Honduras, with or without their troops (for one account of this, see Dickey 1987, pp. 13–68). On the night of July 19–20, the FSLN moved uncontested into Managua. Not one of the factors discussed above which facilitated the triumph of the FSLN in Nicaragua has been present in El Salvador or Guatemala: the armed forces in these countries have been able to contain, although not eliminate, the growth of revolutionary movements; revolutionaries have been more isolated internationally, and governments less so, than was the case in Nicaragua; there has been a political “semi-opening” of sorts in both countries; and although this semi-opening has not necessarily pleased elites, they remain “semi-loyal” to the new regimes and have certainly not sided with revolutionaries. As in Nicaragua, moreover, all of these factors—political, military, diplomatic—have been mutually reinforcing. And as a result of these factors, protracted civil conflicts seem likely to continue in El Salvador and Guatemala; successful revolutionary insurrections, by contrast, are not likely to occur in the foreseeable future. The ability of the Salvadoran and Guatemalan regimes to contain powerful revolutionary movements—in stark contrast to the Somoza regime’s failure to do so in Nicaragua—has rested first and foremost on brutal counter-insurgency campaigns directed against guerrilla strongholds by armies that have been tremendously expanded, armed, and trained with the help of foreign governments. These wars of counterinsurgency were initiated by military factions which overthrew “elected” military governments which were viewed as corrupt and incapable of dealing with the strong guerrilla movements which had developed during the late 1970s. Not surprisingly, these coups were viewed in a particularly favorable light in Washington.


82

In October 1979, shortly after the Sandinista triumph in Nicaragua, and in a context of deepening domestic polarization, elements within the Salvadoran military overthrew the regime of General Lucas Garcia and established a military-civilian junta. Although the most progressive officers and civilians on the junta, including the social democrat Guillermo Ungo, eventually resigned because of continuing state violence against reformists, the junta was reorganized in 1980 with the participation of Christian Democrats, notably Jose Napoleon Duarte, who had been living in exile in Venezuela. Under this junta, which ruled until 1982, the armed forces unleashed a bloody war against the guerrilla groups and the popular organizations affiliated with them; simultaneously, however, preparations were also begun for the election of a civilian government. In Guatemala, General Efrain Rios Montt—the former Christian Democratic candidate for the presidency—took power in a coup in March 1982 in order to prevent the winner of the recent presidential “election,” General Anibal Guevara, from assuming office. This coup also led, as in El Salvador, to the initiation of a savage counter-insurgency campaign in Guatemala’s highlands and, simultaneously, to preparations for the election of a civilian government. Rios Montt was overthrown by his own defense minister, General Oscar Humberto Mejia Victores, in August 1983. but the Mejia Victores government pursued the same general strategy as Rios Montt. The most obvious result of the 1979 coup in El Salvador and the 1982 coup in Guatemala, despite their “reformist” image, was the incredible escalation of violence by the armed forces and allied “death squads.” In order to carry out effective counterinsurgency wars, however, the armed forces in each country needed to be vastly expanded. In El Salvador, the armed forces have grown from some twelve thousand troops in 1980 to more than fifty thousand in 1986; and from 1980 to 1984. the size of the Guatemalan armed forces was expanded from fifteen to forty thousand (see Woodward 1985, p. 367, Barry and Preusch 1986, p. 116, and Karl 1985b, p. 21). The human carnage which counter-insurgency has produced in El Salvador and Guatemala has been truly remarkable. In El Salvador, political murders climbed to more than 1,000 per month in 1980, claiming the Archbishop of San Salvador, six top leaders of the Democratic Revolutionary Front, and four U.S. churchwomen. In the countryside, the army has relied on frequent “sweeps,” aerial bombardment (believed to be the most extensive in the history of the Western hemisphere), and the cutting off of food and supply lines in order to “drain the sea” of guerrilla supporters. In all, more than sixtythree thousand Salvadorans have died since 1980, the vast majority victims of the armed forces and death squads. It is estimated that more than a quarter of the population has been displaced (see Karl 1985a, p. 306), and approximately ten percent of all Salvadorans now live in the United States. Estimates of the number of victims in Guatemala run as high as 100,000 (see Barry 1986, p. 89). In 1984, the Juvenile Division of the Guatemalan Supreme Court asked mayors to compile lists of children who had lost parents since 1980 as a result of political violence. “In September 1984 the Guatemalan press reported that some 100,000 children (and perhaps as many as 200,000) had lost at least one parent and that some 20 per cent of them were orphans” (AI 1987, p. 7). For the years 1981 to 1983, Amnesty International received reports of the mass murder of catechists. “Eye-witness accounts describe incidents in which up to 50 catechists were murdered during army attacks on


83

their villages. Some were garrotted, others were hanged from trees, chopped to pieces with machetes or locked into churches in groups and burned to death” (AI 1987, pp. 24– 25). According to the army 440 Indian villages were razed in army sweeps under the Rios Montt and Mejia Victores governments (Painter 1986, p. 825). As many as 150,000 Guatemalans have fled to refugee camps in Mexico, and internal refugees number around 500,000. Some 50,000 Indians, moreover, have been relocated into so-called “model villages,” which are similar to the “strategic hamlets” established during the Vietnam War (Barry 1986, p. 31). Since 1982, moreover, the military has forced the rural population in strategic areas to participate in so-called Civil Defense Patrols (PACs). By 1986, over one million men and boys, mostly Indian, were serving without compensation on these Patrols (Farnsworth 1987, p. 528). Years of state-directed violence and counter-insurgency have taken their toll on the revolutionary movements, particularly in Guatemala. Revolutionaries have found it difficult to mobilize people, especially in cities, and to protect their members and collaborators. Not surprisingly, the number of armed guerrillas has declined. The Salvadoran guerrillas’ “final offensive” of January 1981 is partly illustrative. This offensive—which the rebels hoped would present the incoming Reagan administration with a fait accompli—scored a number of military victories, and guerrilla forces reached Ilopango, 14 miles from the capital. However, the city of San Salvador, battered by months of violence, was not well organized, and the armed forces were able to militarize public transportation and factories (Montgomery 1982, p. 139). A week after it had begun, the offensive was over. Joaquin Villalobos, one of the guerrillas’ top military strategists, later regretted that the guerrillas had not struck earlier. “What happened,” he suggests, “is that we lost the propitious moment” (quoted in Harnecker 1984, p. 175). Despite the failure of the “final offensive,” however, the guerrilla groups in El Salvador were extremely powerful in the early 1980s. The guerrillas effectively ruled “zones of control” in much of northern El Salvador, where locally elected bodies known as “Local Popular Power” (PPLs) governed independently of the central state (see esp. Pearce 1986). Since 1984, however, the zones of control have been rolled back considerably, and the guerrillas have been forced to break up into small, mobile units. According to recent estimates by U S. military officers in Panama, moreover, the number of guerrillas has declined from a peak of as many as eleven thousand in 1982 to from five to eight thousand in early 1987 (Halloran 1987). A more recent report indicates that there may be no more than four thousand armed rebels in El Salvador (LeMoyne 1987b). By all accounts, the guerrillas in Guatemala are in even worse straits than those in El Salvador. According to one observer, Massacres and village burnings struck fear into Indian communities, causing them to think twice about supporting the guerrillas. The wave of violence that swept through the highlands increased Indian hatred of the army but also resulted in widespread disillusionment with the rebel opposition. While the guerrillas demonstrated their ability to strike out at army targets, they did not prove capable of protecting their popular base of support. [Barry 1986, p. 36] U.S. officers believe that the number of guerrillas in Guatemala has declined to about two


84

thousand (Halloran 1987). The tremendous expansion of the armed forces in El Salvador and Guatemala and the counter-insurgency wars they have fought would not have been possible without international backing, mainly from the U.S., for the new regimes in those countries— support which simply was not forthcoming, as we have seen, in the case of Nicaragua. Indeed, while the Carter administration was unwilling to intervene unilaterally in Nicaragua in order to prevent a Sandinista victory, first the Carter and then the Reagan administration made the prevention of another revolutionary victory in Central America a major foreign policy objective. Consequently, U.S. military assistance to El Salvador has grown to massive sums, from $6 million in 1980 to nearly $200 million in 1984; another $142 million was requested for 1986 (Karl 1985b, p. 13, and Rivera and Sojo 1986, p. 743. In all, U.S. economic and military aid to El Salvador amounted to more than $1.2 billion from 1980 to 1984 (Barry and Preusch 1986, p. 77). Because of a ban on military aid to Guatemala enacted in 1977 by the U.S. Congress, the Guatemalan military has turned to other countries for military assistance, principally Israel, Taiwan, and South Africa. In 1983, Israel helped Guatemala set up two defense factories for manufacturing munitions, Galil rifles, and armored vehicles (Barry 1986, p. 86, and Jamail and Gutierrez 1987, p. 35). According to one recent report, the Israelis have outfitted the entire Guatemalan army, “from helmets to standard-issue automatic rifles” (Jamail and Gutierrez 1987, p. 36). Despite the Congressional ban, moreover, more $30 million of trucks, jeeps, and helicopters were sold by U.S. companies to Guatemala between 1977 and 1984 (Painter 1986, p. 824). U.S. economic aid to Guatemala, moreover, jumped from $11 million in 1980 to $97 million in 1985; another $144 million (including $10 million in military aid) was requested for 1987 (Barry 1986, p. 40). In 1982, furthermore, “the [Reagan] Administration formally erased Guatemala from the list of human rights offenders. The policy immediately affected six World Bank and IDB [Inter-American Development Bank] loans, worth a total of $170 million” (Black 1984, p. 160). Clearly, the incumbent regimes in El Salvador and Guatemala have not faced the same sort of international isolation as the Somoza regime in Nicaragua. Revolutionaries in these countries, by contrast, have been very isolated indeed. No contiguous country, for example, allows Salvadoran or Guatemalan rebels sanctuary, and Honduras, pressured by the United States, has proven especially hostile to the Salvadoran rebels; on a number of occasions, it has co-operated with the Salvadoran military in joint operations against guerrillas and their supporters along the Salvador-Honduras border. There is no evidence, furthermore, of significant arms shipments to rebels from Cuba or Nicaragua since 1981 (see North 1985, pp. 115–16). Salvadoran rebel leaders have often complained that they do not lack combatants, but arms (see Montgomery 1982, p. 144). While U.S.-backed counter-insurgency has been the hallmark of the current Salvadoran and Guatemalan regimes, these regimes have not relied exclusively—like Somoza in Nicaragua—on a military strategy to defeat guerrilla movements. Indeed, the military factions which have directed the counter-insurgency wars have also ended “institutionalized” military rule and organized the election of civilian governments. In other words, they have permitted, even planned, political “semi-openings,” despite the fact that these have been accepted only grudgingly by some of their brethren and by the


85

local oligarchies. The main purpose of these elections, from the military’s point of view, has been to obtain more international aid and to defuse the perception that armed struggle is the only means of effecting social change—in effect, to prevent the “Nicaraguanization” of their countries. Elections, in other words, have been part and parcel of larger counter-insurgency projects. Political openings in El Salvador and Guatemala have essentially taken the shape of nations with the participation of centrist political parties, especially Christian Democrats. Indeed, a whole series of elections have been held in El Salvador and Guatemala over the past five years. In El Salvador, Constituent Assembly elections were held in 1982, a presidential election in 1984, and Legislative Assembly elections in 1985; in Guatemala, Constituent Assembly elections were held in 1984, and two presidential elections (an open contest and a run-off) and Congressional elections in 1985. With the exception of the 1982 elections in El Salvador, which were highly irregular and their results doubtful, Christian Democrats have won a plurality of the vote in all of these elections (see esp. Karl 1985b on El Salvador, and Painter 1986 on Guatemala). In the 1985 elections in Guatemala, significantly, two new centrist parties as well as the center-left Social Democratic Party (PSD) participated along with the Christian Democrats. Currently, the presidents of both El Salvador (Jose Napoleon Duarte) and Guatemala (Vinicio Cerezo) are Christian Democrats. While these elections have obviously demonstrated that political moderates can actually assume office, the real power of Christian Democratic governments has been quite limited. On the one hand, the armed forces have let it be known that civilians should not look too deeply into the conduct of counter-insurgency wars. On the other hand, the local oligarchies, through their friends in the military and right-wing parties such as the National Republican Alliance (ARENA) in El Salvador and the National Liberation Movement (MLN) in Guatemala, have been able to effectively veto reformist legislation. As a result, the political openings in El Salvador and Guatemala have not produced major socioeconomic reforms, nor has a single military officer been brought to trial for human rights abuses. The centerpiece of the political opening in El Salvador has undoubtedly been the land reform enacted by the military-civilian junta in March 1980. This land reform, however, does not address to the needs of El Salvador’s huge rural proletariat; its central provision—which would affect most of El Salvador’s coffee estates—was suspended by the Constituent Assembly in May 1982; and its other provisions have been obstructed by the oligarchy or “benignly neglected” by the government (see Bonner 1984, pp. 191–203, 317–18, and Simon and Stephens 1982). Prior to the 1985 presidential election in Guatemala, moreover, Vinicio Cerezo promised that he would not enact an agrarian reform or nationalize any property if elected (Painter 1986, p. 834). Before turning over power to Cerezo, moreover, General Mejia Victores proclaimed an amnesty for “all people implicated in political crimes” between March 1982 and January 1986. Cerezo, for his part, has said, “We are not going to be able to investigate the past. We would have to put the entire army in jail” (AI 1987, pp. 5–6). And despite pre-election promises, participation in the Civil Defense Patrols has not been made voluntary (Farnsworth 1987, p. 528). The most important aspect of recent elections, however, may very well be symbolic.


86

The very fact that moderate civilians have been allowed to assume office at all creates the perception, or at least the hope, that social change may yet come about through peaceful means. After decades of stasis, even small changes may be encouraging. In El Salvador, it should be noted, Duarte won the support of a coalition of labor and peasant groups known as the Popular Democratic Unity (UPD) for his 1984 presidential bid, signing a “social pact” in which he agreed to enforce economic reforms, accelerate land reform, and negotiate with the guerrillas. Duarte has also used his international backing as leverage to curb the extreme right within the armed forces: Duarte’ s success in Washington brought home the message that to receive substantial increases in U.S. aid El Salvador had to maintain a government acceptable to the U.S. Congress. The Salvadoran Armed Forces responded positively to this message by restructuring the military command. Four leading rightist officers—including Treasury Police head Nicolas Caranza—known to be…linked to death squad activity, were transferred to posts outside El Salvador. In addition, the military agreed to dismantle the intelligence unit of the Treasury Police, the reputed center of death squad activity…. [Karl 1985a, p. 318] In Guatemala, Cerezo disbanded the notorious Department of Technical Investigations (DIT), the intelligence branch of the National Police, shortly after taking office (although most of its members were reassigned and only one, apparently, has been charged with any crime) (AI 1987, pp. 12–13). He has also replaced or transferred the entire membership of the army’s Council of Commanders, which oversees the various regions of the country, and has retired a number of generals (see Farnsworth 1987, p. 532). And while Cerezo has not sought agrarian reform legislation, as promised, “the new government is buying a small number of farms held by state banks for distribution to some of the estimated 419,000 rural workers who have no land of their own” (Farnsworth 1987, p. 528). It is also important to note that guerrillas have abandoned calls for the overthrow of the currently existing regimes in El Salvador and Guatemala; instead, they have sought out direct negotiations with the aim of reaching some sort of “power-sharing” arrangement. It has been said that governments which deal with revolutionaries thereby legitimate those revolutionaries. But the opposite is also true. Thus, whether rebel proposals for negotiations are purely tactical or a result of desperation in the face of a powerful opponent, they effectively help to legitimate incumbent regimes. Since political openings still represent only a potential threat to Salvadoran and Guatemalan elites, however, it is not surprising these elites have remained “semi-loyal” to the new regimes, supportive of elections and civilian rule precisely to the extent that they offer an alternative to the guerrillas which does not threaten their own economic interests. Unlike their Nicaraguan counterparts, certainly, Salvadoran and Guatemalan elites have decidedly not joined ranks with revolutionaries. In fact, elites have skillfully used the political openings, especially in El Salvador, to veto reformist legislation. However, precisely because significant socioeconomic reforms have been blocked and human rights abuses continue, counter-insurgency wars and political openings in El


87

Salvador and Guatemala have not been able to eliminate revolutionary movements. In El Salvador, guerrillas still remain strong in the northern part of the country and have recently scored a number of impressive military victories, including the destruction of an important army base in March 1987, which resulted in the deaths of more than 70 soldiers and a U.S. military adviser. The attack was reportedly co-ordinated with the help of rebels who had infiltrated the base. “Military officials,” reports James LeMoyne, “say they believe the army is now thoroughly infiltrated by the rebels” (LeMoyne 1987b). Since 1984, moreover, there has been a resurgence of street demonstrations and guerrilla activities in San Salvador. A new and powerful leftist labor federation has been founded, the National Unity of Salvadoran Workers (UNTS), which has attracted support from a number of the UPD unions which have become disgruntled with Duarte. The new federation is believed to be close to the rebels and has demanded that the government negotiate with them (see Barry and Preusch 1987, pp. 38–39). In Guatemala, moreover, the number of reported guerrilla actions in 1985 surpassed the levels of 1983 and ‘84. “The EGP guerrillas,” according to Barry, “recently have regained strength in highlands communities. And the two other major guerrilla armies, ORPA and FAR, have built strong revolutionary fronts in the southeast provinces and in the vast Peten region” (Barry 1986, p. 90). In early May 1987, U.S. helicopters based in Honduras ferried “a reported 300 Guatemalan troops into a remote area where Marxist guerrillas have been active” (Kinzer 1987). This operation, which lasted two days and involved 18 American crewmen, was the most visible act of cooperation between the two armies since 1977, when American military aid to Guatemala ceased after a dispute over Guatemala’s human rights record. It was also believed to be the first time since the 1960’s that American helicopters had been used to transport any Central American troops to areas of civil conflict. [Kinzer 1987] What, then, are the prospects that revolutionary movements may yet seize power in El Salvador and Guatemala? Does the recent resurgence of guerrilla activities in El Salvador and Guatemala portend a future seizure of power? Not necessarily. As Eqbal Ahmad reminds us, “A revolutionary guerrilla movement …seeks not simply to inflict military losses on the enemy but to destroy the legitimacy of its government and to establish a rival regime through the creation of ‘parallel hierarchies’” (Ahmad 1971, p. 142). The most favorable context for the latter would undoubtedly result from the closing up of existing political space, presumably as a result of a military coup, and the concomitant international isolation of the regime. However, this situation—tantamount to the “Nicaraguanization” of the current regimes—is precisely what dominant military elites have sought to avoid. In any event, until or unless such political closure occurs, the prospects would seem to be for continued civil conflict. Another alternative, of course, is that negotiations with revolutionaries will open up even more political space, but this has been vetoed thus far by the military (and its domestic and foreign allies) out of a fear that such liberalization would facilitate a guerrilla takeover. As a result, as James Painter has suggested, “the key question [remains] whether a Christian Democrat project [can] do anything to halt the acute social conflict inherent in highly polarised Central American


88

societies when Christian Democrat parties are unable, or unwilling, to confront the traditional holders of economic and political power and cling to the conviction that change can be achieved within existing political structures” (Painter 1986, p. 844). “Defeating the insurgency in Malaysia took 14 years,” one Western diplomat in El Salvador recently noted. “I see it going on like this in El Salvador for another 10 to 12 years” (quoted in LeMoyne 1987a, p. 7). In summary, just as strong revolutionary movements developed in Nicaragua, El Salvador, and Guatemala, but not in Honduras, due to the different nature of the incumbent regimes in these countries, so the Sandinistas were able to seize power in Nicaragua, while their counterparts in El Salvador and Guatemala have not, due to the different nature of the regimes confronted by revolutionaries. The Sandinistas seized power because the “Sultanistic” nature of the Somoza regime precluded a political opening in that country, thereby guaranteeing that the regime would become increasingly isolated from both domestic elites and international actors and unable to stem a growing popular insurrection led by a revolutionary movement which had successfully forged broad alliances with both domestic and international actors. Revolutionaries in El Salvador and Guatemala, by contrast, have been contained by brutal counter-insurgency campaigns and political “semi-openings” that have been initiated by military elites from the previous “institutionalized” military regimes and facilitated, and indeed encouraged, by international actors.

5. Summary and Conclusions During the 1970s, Robert Williams has noted, economic conditions prompted similar patterns of political action throughout the Central American region. “Peasants moved onto idle lands, wage earners demanded cost-of-living adjustments, and large landowners called on the services of the local police and the national security forces. What differed from country to country,” he adds, “was the way national governments responded to the pressures from the different camps” (Williams 1986, p. 166). This paper has argued that revolutionary movements became powerful actors in Nicaragua, El Salvador, and Guatemala due to the political closure and indiscriminate repression of narrowly based authoritarian regimes. (For a schematic summary of the analysis developed here, see Diagram 2.) Revolutionary organizations were formed by dissidents from electorally oriented political parties who became frustrated by repeated electoral fraud and political repression. State violence, against a backdrop of widespread social grievances and political mobilization, created a context which allowed revolutionaries to attract support from, and form alliances with, a broad range of social and political groupings. Diagram 2


89

In Honduras, meanwhile, a very different “cycle of political frustration”—one based on the alternation of reformist and conservative rule—seems to have inhibited the development of a significant armed revolutionary movement. Indeed, the case of Honduras seems to suggest that opportunities for quite modest reforms may sometimes preclude the development of revolutionary insurgencies, both because leftist organizations are likely to take advantage of such opportunities—and eschew armed struggle—in order to win support among workers and peasants, and because large numbers of ordinary people are extremely unlikely to risk their lives in an armed struggle when opportunities for peaceful change are seen to exist. This is not to imply, however, that armed insurgencies will always quickly dissipate if and when a dictatorial regime “opens up.” As the Salvadoran and Guatemalan cases demonstrate (along with present-day Peru, Colombia, and the Philippines), this depends


90

on a number of more specific factors, including the ability of the government to enact reforms which directly benefit real or potential rebel supporters and/or to ensure that revolutionaries and their followers can safely and effectively participate in the electoral arena. Not surprisingly, revolutionaries often decide that it is simply too dangerous to lay down their arms. This paper has also emphasized, however, that the conditions which foster strong revolutionary movements do not guarantee that such movements will actually seize state power. While revolutionaries have become powerful actors in Nicaragua, El Salvador, and Guatemala, they have only come to power (thus far) in Nicaragua, due largely, I have argued, to the differences between the Somoza regime, on the one hand, and the Salvadoran and Guatemalan regimes, on the other. Because of the “Sultanistic” nature of the Somoza regime and the absence of a political opening in Nicaragua, the Sandinistas were able to build a broader coalition of both domestic and international actors than their Salvadoran and Guatemalan counterparts. By contrast, Salvadoran and Guatemalan revolutionaries have confronted regimes which, in the face of revolutionary challenges, have incorporated significant sectors of the moderate opposition; partly as a result of these “semi-openings,” moreover, the Salvadoran and Guatemalan regimes have been bolstered by massive international aid, military training, and direct logistical assistance, mainly, although not exclusively, from the United States. Indeed, whereas the Carter administration ultimately proved unwilling to support the Somoza regime in Nicaragua or otherwise to prevent a Sandinista victory, both the Carter and Reagan administrations subsequently made a major, and thus far apparently decisive, effort to prevent a leftist victory in El Salvador or Guatemala. Because civilian regimes in El Salvador and Guatemala have been unable to enact significant socioeconomic reforms, however, protracted civil conflicts continue to drag on in those countries. What are the implications of these findings for theories of revolution? More specifically, what do they say about the causes of successful revolutionary movements? What factors “predict,” so to speak, successful revolutions? Recent events in Central America would seem to bear out Robert Dix’s argument that revolutionaries need to assemble a broad “negative” (i.e., anti-status quo) coalition of diverse social groups, including elites, in order to seize power. As Dix notes, this is precisely what distinguishes the Nicaraguan from the Salvadoran (and, I would add, Guatemalan) case. Events also bear out Dix’s assertion that the ability of revolutionaries to construct such a coalition depends less on socioeconomic factors than on political factors, specifically, “the catalyzing effect afforded by a narrowly-based dictatorship” (Dix 1984, p. 438). We have seen that socioeconomic factors and trends alone are unable to account for the political fortunes of revolutionary movements in the region. Instead, we have emphasized how the inflexible and “Sultanistic” Somoza regime in Nicaragua, as against the more institutionalized and now “semi-open” regimes in El Salvador and Guatemala, was more vulnerable to a revolutionary challenge. This finding also bears out Jack Goldstone’s contention that “precisely those states that are structurally prone to internal conflicts…[with] elites are most vulnerable to revolutions” (Goldstone 1982, p. 197). The more narrowly based the regime, the more likely such state-elite conflicts. Thus, recent events in Central America also largely confirm the assertion of Charles Tilly that:


91

Despite the many recent attempts to psychologize the study of revolution by introducing ideas of anxiety, alienation, rising expectations, and the like, and to sociologize it by employing notions of disequilibrium, role conflict, structural strain, and so on, the factors which hold up under close scrutiny are, on the whole, political ones. The structure of power, alternative conceptions of justice, the organization of coercion, the conduct of war, the formation of coalitions, the legitimacy of the state—these traditional concerns of political thought provide the main guides to the explanation of revolution…. [L]arge-scale structural changes do, to be sure, affect the probabilities of revolution. But they do so indirectly, by shaping the potential contenders for power, transforming the techniques of governmental control, and shifting the resources available to contenders and governments. [Tilly 1973, p. 447] “Neo-patrimonial” or “Sultanistic” regimes like that of the Somoza family in Nicaragua are particularly vulnerable to revolutions, given their very narrow social base, tenuous legitimacy, and extreme dependency on foreign powers (see Linz 1975, Eisenstadt 1978, and Goldstone 1985). However, there have been very many long-standing personalist dictatorships which have not been overthrown by organized revolutionary movements— the Trujillo regime in the Dominican Republic, the Stroessner regime in Paraguay, the government of Mobutu Sese Seko in Zaire, and the Suharto regime in Indonesia, to name just a few. Indeed, it should not be forgotten that the Somoza dynasty itself survived for more than forty years before it faced a strong revolutionary challenge. The nature of the incumbent regime, therefore, while extremely important in accounting for both the emergence and likely success of revolutionary movements, is by no means the only variable. Our findings also suggest that revolutionary triumph requires a unified vanguard organization able to operate with some modicum of physical security within the country and willing and able to form alliances with broad social sectors and international actors. For revolutionary movements to expand, not only must reformists be unable to operate openly, but revolutionaries must be able to operate clandestinely. Above all, revolutionaries must come to appear to offer a credible alternative to the incumbent regime—“the prospect of a revolutionary way out,” as Trotsky put it. However exclusive and despotic the existing regime, and however weak and ineffectual the unarmed opposition, popular mobilization will be impossible if revolutionaries do not come to be viewed as capable of seriously challenging the status quo. This, however, is typically a long and slow process fraught with difficulties. Revolutionaries may not be able to establish themselves or operate effectively in the country; they may be eliminated before they are viewed as a serious challenge; or they may be unable to resolve their own political differences. The incumbent regime itself, moreover, may very well change—as it has in El Salvador and Guatemala—in ways which undermine the prospects for revolution. Not surprisingly, many radical groups have decided that it is too risky to mount an armed insurgency in the first place. Our findings, not surprisingly, also underline the influence of international factors in revolutionary upheavals—even though the “world system” has not always acted in Central America in the ways that many have suggested. The manner in which Central


92

American countries are incorporated into the capitalist world-economy, however one chooses measure this, does not seem to explain either the incidence or success of revolutionary movements. As we have seen, the case of Honduras, in particular, suggests that extreme external dependency does not in itself explain why revolutions do or do not occur, nor even whether a significant revolutionary movement will emerge at all. And we have also seen that revolutionary movements in Central America were not unleashed by “coercive crises” deriving from international wars, even though these have been important precipitants of other revolutions (see esp. Skocpol 1979). Rather, the “world system” has figured into the Central American conflicts mainly in terms of whether and how foreign governments have aided either incumbent regimes or revolutionary movements in terms of material aid (including weapons), territorial sanctuary, and diplomatic support, Although such assistance cannot in itself create a powerful revolutionary movement, it may prove decisive in determining whether such a movement will or will not seize power. When the “international correlation of forces” favored revolutionaries, as it did in Nicaragua in the late 1970s, revolution became possible, although hardly inevitable; when international forces came to favor incumbent regimes, as they have during the 1980s in El Salvador and Guatemala, revolution became much more difficult, if not impossible. Finally, our findings also challenge simple class analyses of revolutionary movements. However necessary it may be to determine just which classes join or support revolutionary movements (and why), this focus should not lead us away from the crucial questions of political leadership, organization, and opportunity emphasized in this paper. The tendency to view all revolutionary movements as movements of a particular social class—a view evident in Jeffery Paige’s recent work on Central America (see Paige 1983 and 1985)—obscures the independent role played by revolutionary “vanguard” organizations and the ability of such organizations to mobilize a variety of social classes and strata, including elites. Unfortunately, as Skocpol has noted, revolutionary leaderships are typically treated “as representatives of classes or social groups, struggling to realize economic or status interests, and/or as actors attempting to implement a certain ideological vision of the ideal social order.” “What tends to be missed in all of this,” she argues, “is that which political leaderships in revolutionary crises are above all doing— claiming and struggling to maintain state power” (Skocpol 1979, pp. 164–65). We have suggested that revolutionary movements are best thought of as “proto-state” organizations; revolutionary struggles, for their part, are above all competitions in statebuilding. As should now be clear, the crucial political issue in Central America has been whether and how revolutionary organizations, as opposed to other types of political leaderships, have been able to mobilize masses of people and create an “alternative government” capable of seizing power.

Notes The author would like to thank the following individuals for their comments and suggestions on earlier versions of this work: Stephen Cornell, Robert Fishman, Murray Milgate, Orlando Patterson, Debra Satz, Joseph Schwartz, J.Samuel Valenzuela, and the


93

members of the Center for Research on Politics and Social Organization (CROPSO) in the Department of Sociology at Harvard University, particularly Mustafa Emirbayer, Carlos Forment, Marta Gil, Theda Skocpol, and Eliza Willis. An earlier version of this paper was presented at the Section on the Sociology of Revolutions and Political Violence at the 82nd Annual Meeting of the American Sociological Association, August 17–21, 1987, in Chicago, Illinois. I would like to thank Kenneth Bollen for his comments and queries in that forum. 1. The data in Table 3, which are for 1970, undoubtedly underestimate the current extent of landlessness and land poverty, which grew rapidly in the 1970s, especially in El Salvador (see, e.g., North 1985, p. 42). 2. Good accounts of the Nicaraguan Revolution may be found in Booth (1985) and Black (1981). Vilas (1986) and Weber (1981) are more interpretive. 3. Good accounts of recent Salvadoran history may be found in Dunkerley (1985), North (1985), Montgomery (1982), and Baloyra (1982). 4. Good accounts of the recent Guatemalan history may be found in Black (1984), Handy (1984), and Frank and Wheaton (1984). 5. Good accounts of recent Honduran history may be found in Latin America Bureau (1985) (hereafter, LAB), Peckenham and Street, eds. (1985), Morris (1984a), and Posas and del Cid (1983).

References Aguilera, Gabriel. 1979. “The Massacre at Panzos and Capitalist Development in Guatemala.” Monthly Review 31:13–23. Ahmad, Eqbal. 1971. “Revolutionary Warfare and Counterinsurgency.” Pp. 137–213 in National Liberation: Revolution in the Third World, edited by Norman Miller and Roderick Aya. New York: The Free Press. Amnesty International. 1987. Guatemala: The Human Rights Record. London: Amnesty International Publications. Baloyra, Enrique. 1982. El Salvador in Transition. Chapel Hill, N.C.: University of North Carolina Press. Baloyra, Enrique. 1983. “Reactionary Despotism in Central America.” Journal of Latin American Studies 15:295–319. Barry, Tom. 1986. Guatemala: The Politics of Counterinsurgency. Albuquerque, N.M.: Inter-Hemispheric Education Resource Center. Barry, Tom, and Deb Preusch. 1986. The Central America Fact Book. New York: Grove Press. Barry, Tom, and Deb Preusch. 1987.“The War in El Salvador: A Reassessment.” Monthly Review 38:29–44. Black, George. 1981. Triumph of the People: The Sandinista Revolution in Nicaragua. London: Zed Press. Black, George. 1984. Garrison Guatemala. With Milton Jamail and Norma Stoltz Chinchilla. New York: Monthly Review Press.


94

Bonner, Raymond. 1984. Weakness and Deceit: U.S. Policy and El Salvador. New York: Times Books. Booth, John A. 1985. The End and the Beginning: The Nicaraguan Revolution. 2nd ed. Boulder, Colo.: Westview Press. Burbach, Roger. 1984. “Introduction: Revolution and Reaction.” Pp. 9–28 in The Politics of Intervention: The United States in Central America, edited by Roger Burbach and Patricia Flynn. New York: Monthly Review Press, Center for the Study of the Americas. Cabarrus, Carlos Rafael. 1983. Genesis de una revolucion. Mexico City: Ediciones de la Casa Chata. Carmack, Robert M. 1985. “La cultura quiche-maya como factor en la revolucion de Guatemala.” State University of New York at Albany, unpublished manuscript. Chavarria, Ricardo E. 1982. “The Nicaraguan Insurrection: An Appraisal of its Originality.” Pp. 25–40 in Nicaragua in Revolution, edited by Thomas W.Walker. New York: Praeger. Dickey, Christopher. 1987. With the Contras: A Reporter in the Wilds of Nicaragua. New York: Simon & Schuster. Dix, Robert H. 1984. “Why Revolutions Succeed and Fail.” Polity 16:423–46. Dunkerley, James. 1985. The Long War: Dictatorship and Revolution in El Salvador. 2nd ed. London: Verso. Eisenstadt, S.N. 1978. Revolution and the Transformation of Societies. New York: The Free Press. Farnsworth, Elizabeth. 1987. “Voices From Guatemala” (interviews). World Policy Journal 4:527–552. Frank, Luisa, and Philip Wheaton. 1984. Indian Guatemala: Path to Liberation. Washington, D.C.: EPICA Task Force. Goldstone, Jack A. 1982. “The Comparative and Historical Study of Revolutions.” Annual Review of Sociology 8:187–207. Goldstone, Jack A. 1985. “Superpowers and Revolutions.” Department of Sociology, Northwestern University, Evanston, I11., unpublished manuscript. Goodsell, James Nelson. 1982. “Nicaragua.” Pp. 52–60 in Communism in Central America and the Caribbean, edited by Robert Wesson. Stanford, Cal.: Stanford University, Hoover Institution Press. Gorman, Stephen M. 1984. “Social Change and Political Revolution: The Case of Nicaragua.” Pp. 33–66 in Central America: Crisis and Adaptation, edited by Steve C.Ropp and James A.Morris. Albuquerque, N.M.: University of New Mexico Press. Halloran, Richard. 1987. “Latin Guerrillas Joining Forces, U.S. Officers Say.” The New York Times (March 3). Handy, Jim. 1984. Gift of the Devil: A History of Guatemala. Boston: South End Press.


95

Harnecker, Marta. 1984. Pueblos en armas (interviews). Mexico City: Ediciones Era. Jamail, Milton, and Margo Gutierrez. 1987. “The Israeli Connection: Guns and Money in Central America.” NACLA Report on the Americas 21:13–39. Karl, Terry Lynn. 1985a. “After La Palma: The Prospects for Democratization in El Salvador.” World Policy Journal 2:305–30. Karl, Terry Lynn. 1985b. “Imposing Consent? Electoralism Versus Democratization in El Salvador.” University of California, Berkeley, unpublished manuscript. Kinzer, Stephen. 1987. “Guatemalan Army Yields Little Power to Leader.” The New York Times (May 13). Kirkpatrick, Jeane. 1986 (1979). “Dictatorships and Double Standards.” Pp. 14–35 in El Salvador: Central America in the New Cold War, edited by Marvin E.Gettleman, Patrick Lacefield, Louis Menashe, and David Mermelstein. Revised ed. New York: Grove Press. (Originally published in Commentary 68:34–45.) Krumwiede, Heinrich-W. 1984. “Regimes and Revolution in Central America.” Pp. 9–36 in Political Change in Central America: Internal and External Dimensions, edited by Wolf Grabendorff, Heinrich-W.Krumwiede, and Jorg Todt. Boulder, Colo.: Westview Press. LaFeber, Walter. 1984. Inevitable Revolutions: The United States in Central America. New York: Norton. Latin America Bureau. 1985. Honduras: State for Sale. London: Latin America Bureau. Leiken, Robert S. 1984.“The Salvadoran Left.” Pp. 111–30 in Central America: Anatomy of Conflict , edited by Robert S.Leiken. New York: Pergamon Press. LeMoyne, James. 1987a. “After Parades and Promises, Duarte Flounders in Salvador.” The New York Times (February 16). LeMoyne, James. 1987b. “Rebels Gain While Duarte Flounders.” The New York Times (July 26). LeoGrande, William M. 1982. “The United States and the Nicaraguan Revolution.” Pp. 63–77 in Nicaragua in Revolution, edited by Thomas W.Walker. New York: Praeger. Linz, Juan J. 1975. “Totalitarian and Authoritarian Regimes.” Pp. 175–411 in Handbook of Political Science (Vol. 3): Macropolitical Theory, edited by Fred I. Greenstein and Nelson W.Polsby. Reading, Mass.: Addison-Wesley. MacCameron, Robert. 1983. Bananas, Labor, and Politics in Honduras: 1954–1963. Syracuse, N.Y.: Maxwell School of Citizenship and Public Affairs, Syracuse University. Montgomery, Tommie Sue. 1982. Revolution in El Salvador: Origins and Evolution. Boulder, Colo.: Westview Press. Moore, Barrington, Jr. 1966. Social Origins of Dictatorship and Democracy. Boston: Beacon Press. Morris, James A. 1984a. Honduras: Caudillo Politics and Military Rulers. Boulder, Colo.: Westview Press. Morris, James A. 1984b. “Honduras: The Burden of Survival in Central America.” Pp.


96

189–225 in Central America: Crisis and Adaptation, edited by Steve C.Ropp and James A. Morris. Albuquerque, N.M.: University of New Mexico Press. Nolan, David. 1984. The Ideology of the Sandinistas and the Nicaraguan Revolution. Coral Gables, Fla.: Institute of Interamerican Studies, University of Miami. North, Liisa. 1985. Bitter Grounds: Roots of Revolt in El Salvador. 2nd ed. Westport, Conn.: Lawrence Hill. Paige, Jeffery M. 1975. Agrarian Revolution: Social Movements and Export Agriculture in the Underdeveloped World. New York: The Free Press. Paige, Jeffery M. 1983. “Social Theory and Peasant Revolution in Vietnam and Guatemala.” Theory and Society 12:699–737. Paige, Jeffery M. 1985. “Cotton and Revolution in Nicaragua.” Pp. 91–114 in States Versus Markets in the World-System, edited by Peter Evans, Dietrich Rueschemeyer, and Evelyne Huber Stephens. Beverly Hills, Cal.: Sage Publications. Painter, James. 1986. “Guatemala in Civilian Garb.” Third World Quarterly 8:818–44. Pearce, Jenny. 1986. Promised Land: Peasant Rebellion in Chalatenanqo El Salvador. London: Latin America Bureau. Peckenham, Nancy, and Annie Street, eds. 1985. Honduras: Portrait of a Captive Nation. New York: Praeger. Perez Brignoli, Hector. 1985. Breve historia de Centroamerica. Madrid: Alianza Editorial. Petras, James F., and Morris H.Morley. 1983. “Imperialism and Intervention in the Third World: U.S. Foreign Policy and Central America.” Pp. 219–38 in The Socialist Register 1983, edited by Ralph Miliband and John Saville. London: The Merlin Press. Posas, Mario, and Rafael del Cid. 1983. La construction del sector publico y del estado national en Honduras, 1876–1979. 2nd ed. San Jose, Costa Rica: EDUCA. Ramos, Carlos. 1987. “El Salvador, un pais dividido entre dos poderes enfrentados.” El Pais (Madrid) (February 16). Reina, Jorge Arturo. 1981. “Honduras: revolucion pacifica o violenta?” Nueva Sociedad 52:35–56. Rivera Urrutia, Eugenio, and Ana Sojo. 1986. “El perfil de la politica economica en Centroamerica: la decada de los ochenta.” Pp. 13–113 in Centroamerica: politica economia y crisis, by Eugenio Rivera Urrutia, Ana Sojo, and Jose Roberto Lopez. San Jose, Costa Rica: Editorial DEI. Ruhl, J.Mark. 1984. “Agrarian Structure and Political Stability in Honduras.” Journal of Interamerican Studies and World Affairs 26:33–68. Salomon, Leticia. 1982. Militarismo y reformismo en Honduras. Tegucigalpa: Editorial Guaymuras. Seligson, Mitchell A., and William J.Carroll III. 1982. “The Costa Rican Role in the Sandinista Victory.” Pp. 331–43 in Nicaragua in Revolution, edited by Thomas W. Walker. New York: Praeger. Shepherd, Philip L. 1986. “Honduras.” Pp. 125–55 in Confronting Revolution: Security


97

Through Diplomacy in Central America, edited by Morris J.Blachman, William M. LeoGrande, and Kenneth E.Sharpe. New York: Pantheon. Simon, Laurence R., and James C.Stephens. 1982. El Salvador Land Reform, 1980– 1981: Impact Audit. 2nd ed. With a 1982 Supplement by Martin Diskin. Boston: Oxfam America. Skocpol, Theda. 1979. States and Social Revolutions: A Comparative Analysis of France, Russia, and China. Cambridge: Cambridge University Press. Skocpol, Theda. 1982. “What Makes Peasants Revolutionary?” Comparative Politics 14: 351–75. Tilly, Charles. 1973. “Does Modernization Breed Revolution?” Comparative Politics 5: 425–47. Tilly, Charles. 1975. “Revolutions and Collective Violence.” Pp. 483–555 in Handbook of Political Science (Vol. 3): Macropolitical Theory, edited by Fred I.Greenstein and Nelson W.Polsby. Reading, Mass.: Addison-Wesley. Torres Rivas, Edelberto. 1983. Crisis delpoder en Centroamerica. 2nd ed. San Jose, Costa Rica: EDUCA. Trotsky, Leon. 1961 (1932). The History of the Russian Revolution. Translated by Max Eastman. New York: Monad Press. Trudeau, Robert, and Lars Schoultz. 1986. “Guatemala.” Pp. 23–49 in Confronting Revolution: Security Through Diplomacy in Central America, edited by Morris J. Blachman, William M.LeoGrande, and Kenneth E.Sharpe. New York: Pantheon. U.S. Arms Control and Disarmament Agency. 1984. World Military Expenditures and Arms Transfers, 1972–1982. Washington, D.C.: U.S. Arms Control and Disarmament Agency. Vilas, Carlos M. 1986. The Sandinista Revolution: National Liberation and Social Transformation in Central America. Translated by Judy Butler. New York: Monthly Review Press, Center for the Study of the Americas. Volk, Steven. 1983. “Honduras: On the Border of War.” Pp. 203–43 in Trouble in Our Backyard: Central America and the United States in the Eighties, edited by Martin Diskin. New York: Pantheon. Weber, Henri. 1981. Nicaragua: The Sandinist Revolution. Translated by Patrick Camiller. London: Verso. Weeks, John. 1985. The Economies of Central America. New York: Holmes & Meier. Weeks, John. 1986. “An Interpretation of the Central American Crisis.” Latin American Research Review 21:31–53. Wheelock Roman, Jaime. 1980 (1975). Imperialismo y dictadura: crisis de una formacion social. 5th ed. Mexico City: Siglo Veintiuno. Williams, Robert G. 1986. Export Agriculture and the Crisis in Central America. Chapel Hill, N.C.: University of North Carolina Press. Woodward, Ralph Lee, Jr. 1984. “The Rise and Decline of Liberalism in Central America: Historical Perspectives on the Contemporary Crisis.” Journal of


98

Interamerican Studies and World Affairs 26:291–312. Woodward, Ralph Lee, Jr. 1985. Central America: A Nation Divided. 2nd ed. New York: Oxford University Press.


99

45 Party and Patronage: Germany, England, and Italy* Martin Shefter

* Source: Politics and Society, 1977, vol. 7, pp. 403–51.

Political Parties and Patronage: A Theory A political party may employ one of two basic strategies in its efforts to induce voters to support its candidates, contributors to give money to its cause, and activists to work for the party organization. It may distribute divisible benefits—patronage of various sorts— to the individuals who support the party. Alternatively, it may distribute collective benefits or appeal to a collective interest in an effort to elicit contributions of money, labor, or votes from its supporters.1 Under what conditions are parties likely to pursue one or the other of these strategies? Over the past ten or fifteen years a number of social scientists have addressed themselves to this question, and though they have proposed many different answers to it, something of a scholarly consensus has emerged with respect to where such an answer is to be found. The theories of political patronage enjoying the greatest currency today are fundamentally sociological in approach. They all seek to account for variations among nations in the role patronage plays in political parties by searching for variations in social structure or political culture that can be correlated with these differences in party behavior. And they all tend to explain any correlations they find in a similar way—by arguing that voters who have the social characteristic in question are especially likely to demand patronage in exchange for their votes. Thus scholars in recent years have variously argued that machine politics will prevail where the bulk of the electorate is in a transitional stage of social modernization, in societies that have cognatic kinship structures, among groups that have a private-regarding or individualist political ethos, in cities whose electorate is ethnically heterogeneous, among social classes whose orientations are more parochial than cosmopolitan, and so forth.2 These theories of patronage are deficient in a number of respects. In the first place, they are overly narrow in focus: they fail to recognize that the issue of patronage has a bearing upon the interests of groups besides ordinary voters and party politicians, and that these groups may be in a position to influence, or place constraints upon, the strategies parties adopt. In particular, the question whether political parties will be able to raid the bureaucracy in search of patronage raises several of the same issues that were involved in the struggles that occurred in many nations during the predemocratic era over the creation

Party and Patronage: Germany, England, and Italy

100

of a modern bureaucracy and a “reformed” civil service—conflicts that often took place before mass electorates and party politicians appeared on the scene. Furthermore, sociological theories of patronage are deficient to the extent that they are ahistorical. They generally fail to recognize that the interests various groups acquire and the alliances they form during conflicts over patronage in the predemocratic era can persist into the era of mass suffrage, and that the outcome of these earlier struggles can have enduring consequences for the strength of the contending forces in later struggles over party patronage. These consequences can endure, moreover, in the face of social changes of the sort—modernization, industrialization, assimilation, acculturation—that according to sociological theories of patronage lead to the rise or decline of patronage politics. In this article I seek to construct a theory of patronage that avoids these shortcomings. I begin this task in the sections below by presenting some evidence that casts doubt upon the explanatory power of sociological theories of patronage. I then discuss some important conditions that influence whether political parties are likely to rely upon a patronage strategy—conditions that do not receive sufficient attention among current theories—and propose an alternative theory that takes these forces into account. Sections II, III, and IV provide evidence for this theory by showing how it is able to account for the role patronage plays in the party politics of Germany, Great Britain, and Italy— nations that illustrate three of the major historical alternatives specified by the theory I propose. The concluding section indicates how the theory outlined here can be reconciled with two cases, France and the United States, that appear on their face to contradict it. The Neoclassical Theory of Patronage It is possible to discern some common themes within the recent scholarly literature on patronage, themes that distinguish these writings from those of an earlier generation of scholars and that warrant one’s speaking of a new wisdom with regard to how patronage and machine politics are to be understood. In the first place, these writings rely upon a common model of party behavior, one rather similar to the model of the firm in neoclassical economic theory. Political parties in open electoral systems, it is argued, are unable to use coercion to remain in power; rather they must respond to the demands of voters in order to win popular support, just as—or so the argument goes—business firms operating in free markets are compelled to respond to the tastes and preferences of consumers if they are to sell their products and reap a profit. In the second place, as noted above, these writings all assume that the demands and preferences of voters are determined by their social background and cultural heritage. And though their explanations as to why this is so differ widely, they generally agree that immigrants, displaced peasants, and the poor are especially likely to demand patronage in exchange for their votes, and that voters who belong to the middle class and to the industrial working class are most likely to respond favorably to parties that offer collective or programmatic benefits. From these common premises a common conclusion follows: political parties are most likely to be patronage oriented when they rely upon the support of voters of the former type, and they are most likely to stage collective or ideological appeals when they rely upon the support of voters of the latter type. This argument, implicit in the writings of James C.Scott, Edward Banfield and James Q.Wilson, Leon


101

Epstein, and numerous other political scientists, can be termed the neoclassical theory of political patronage.3 The accompanying table enables us to subject this theory to a rough test by classifying parties according to the two attributes suggested by the neoclassical theory: whether they distribute individual or collective benefits to their supporters and whether they rely upon the votes of urban migrants and uprooted peasants, on one hand, or the middle class and industrial working class, on the other. The political parties listed in this table are cited merely as examples, but they do suggest that with little difficulty one can locate cases that are not consistent with the neoclassical theory, namely those grouped in the lower left and those grouped in the upper right.

Classification of Political Parties (According to Social Base of supporters and Inducements Offered to Them)

Social Base Migrants, Uprooted Peasants Middle Class, Industrial Working Class Philippines: Nacionalistas France: Radical Party Individual Ghana: Convention People’s Party New York, Pennsylvania: (patronage) Chicago, Philadelphia: Democratic Republican Party Nassau County: Republican Machine Machine Indiana, Ohio: Italy: Christian Democrats Democratic Party Bologna: Communist Party United States: Progressives, Reformers Collective Republican Spain: Socialist Party, Britain: Labour, (public, policies Anarchosyndicalists Michigan, Conservatives Germany: ideology) Wisonsin: Democrat Party Social Democrats, Christian Democrats Party Inducements

Consider first the parties listed in the lower left: each of these is an ideological or issueoriented party or relies heavily upon solidaristic appeals, yet each of these parties has a social base rather similar to that of the patronage-oriented parties listed in the upper left. In contemporary Bologna, for example, the Communist Party (PCI) relies heavily upon votes of recent migrants from the surrounding countryside, and since it controls the local government, it is in a position to raid the public treasury in a quest for patronage. Nonetheless the PCI has not degenerated into a patronage machine, and the city of Bologna is administered with less corruption and petty favoritism than are comparable Italian cities that are governed by the Christian Democrats (DC).4 More generally, as both Michael Paul Rogin and Sidney Tarrow have noted, in the industrializing cities of Europe in the late nineteenth century, socialist parties had a popular base quite similar to that of the political machines governing the industrializing cities of the United States at the same time: uprooted peasants who migrated to the cities of Europe—to Milan and Paris—were mobilized by parties that preached revolution and class solidarity, while their cousins who migrated to the cities of the United States—to


102

Chicago and Philadelphia—were mobilized by conservative patronage machines.5 Similarly, in rural Europe earlier in this century one could find political parties that mobilized peasants against traditional patron-client structures, as well as parties that dealt with peasants through such structures: Republican Spain provides examples of each.6 Finally, to cite some examples from contemporary American politics, the Democratic parties in the states of the upper Middle West (e.g., Michigan, Wisconsin) are, to use John Fenton’s terminology, “issue-oriented,” while their counterparts in the lower Middle West (e.g., Indiana, Ohio) are “job-oriented,” although both rely upon the support of working-class voters of immigrant stock7 In a similar vein, it is possible to cite many examples of patronage-oriented parties that rely heavily upon the support of the middle classes, parties that would be located at the upper right of the table. A clear-cut case is the French Radical party of the Third and Fourth Republics—the classic middle-class party of small town notables and Parisian ministrables.8 Political parties in the United States from their beginnings have fallen into this same mold. The spoils system of the Jacksonian Democrats was perfected before poverty-stricken peasants from Ireland and from southern and eastern Europe became a major element in the American population, when the United States was still predominantly a country of relatively prosperous, Protestant gentry and yeomanry. Patronage continues to this day to play an important role in the party politics of most states of the Northeast and lower Middle West, including in those parties (e.g., the Republican parties in Pennsylvania, Ohio, and New York) that draw much of their support from the middle classes. (Indeed, throughout most of this century one of the strongest county organizations in the Republican party of New York State has been the machine of G.Wilbur Doughty, J.Russel Sprague, and now Joseph Margiotta in heavily middle-class Nassau County.)9 Finally, one might cite the case of the Christian Democratic party in contemporary Italy, whose orientations toward patronage are attributed by Luigi Graziano and Alessandro Pizzorno to its efforts to win support within the middle classes.10 Taken together these examples suggest that the behavior of political parties is not strictly determined by the composition of their popular base: parties have managed to win the support of migrants and peasants in transitional societies both by working through “vertical” patron-client networks and by organizing them into “horizontal” associations; parties have been able to mobilize the middle and working classes in more modern social settings both by distributing patronage to individual members of those classes and by appealing to broader class and occupational loyalties among them. Constructing an Alternative Explanation If the neoclassical theory is not fully satisfactory, how can a superior theory be constructed? In order to begin this task it is useful to return to the model of the political party with which we began, a model that conceives of political parties seeking votes in open electoral systems as analogous to business firms seeking customers in a free market. This analogy is a helpful one, but in arguing that a political party will be compelled to respond to the preferences of potential supporters if it is to receive their votes, the political scientists cited above introduce a questionable assumption. They assume that the


103

political preferences of voters exist prior to, and apart from, the alternatives that are presented to them in the electoral arena. This assumption is the political equivalent of the notion of the autonomy of preferences (or exogenous preferences) in neoclassical economic theory—a notion that has been subject to telling criticism by economists, and that, as the examples cited above suggest, is equally dubious in the political realm.11 Evidently the political behavior of voters in both “transitional” and “modern” settings is far more variable and malleable than the neoclassical theorists assume. Political parties have found it possible to elicit support from voters in each of these settings by staging both individualistic and collective appeals. If the strategies parties adopt are not totally constrained by the characteristics of the voters whose support they seek, one must turn elsewhere to account for their behavior. Again, the analogy to the business firm is helpful. To the extent that the behavior of a firm is not determined by the (autonomous) preferences of its customers, the chief executive of that firm, in deciding which products to produce and which marketing strategies to pursue, can respond to other considerations.12 The classic example, of course, is the commitment of the Big Three automotive companies during the two decades following World War II to the set of policies—the manufacture of large cars, the proliferation of models, annual restyling—that so outraged their critics and that eventually made them vulnerable to foreign competition. In order to explain the behavior of General Motors, Ford, and Chrysler, one must take into account three factors. First, there are the tastes and preferences of consumers: the American public indeed was prepared to buy such automobiles. But the production and marketing decisions of the auto manufacturers cannot be attributed solely to consumer demand. The automobile companies, after all, advertised extensively to cultivate and to reinforce this demand, and once foreign manufacturers began selling smaller and less stylish cars in the United States it was evident that a market existed for such automobiles, one that the Big Three were rather slow to cultivate. Second, at any moment it was simply more feasible for the American auto manufacturers to supply the market with large cars than small ones. The resources available to them—their physical plant, their technologies, the skills of their employees, and their very organizational structure—all were geared to the production and marketing of large, stylish automobiles. Third, and most important of all, the production of large, stylish automobiles best enabled the top automotive executives to deal with their several “constituencies”—constituencies other than their customers, namely, their managers, creditors, shareholders, labor unions, subsidiaries, suppliers, and potential (American) competitors—in that the strategy they pursued enabled gross revenues, profits, wages, and entry costs to be maintained at a high level.13 By the same token, at any moment there are three considerations that enter into the decisions party leaders make concerning whether to adopt or eschew a patronage strategy. First, there are the orientations and preferences of the voters to whom the party is appealing for support. Second, there are the resources available to the party. Third, there are the interests of the activists who staff the party apparatus and of the elites (if any) with whom the party is allied or who are capable of levying sanctions against it. If a party is to rely upon a patronage appeal, for example, the voters whose support it is seeking must value the particularistic benefits it distributes; the party must enjoy access to a pool of resources from which such benefits can be generated; and the party


104

leadership must calculate that, considering the reactions of elites as well as voters, the gains to be realized exceed the losses the party might incur, including the opportunities it foregoes, if it uses these resources in this way. Voter Demands Now, as I already have suggested, the first of these factors—the orientations of the voters to whom the party is appealing for support—cannot be considered apart from, and prior to, the techniques parties actually have employed to win support among the voters in question. Except under one unusual set of circumstances, which need not be discussed at any length here,14 the social composition of the electorate does not uniquely determine party behavior. Rather, as I shall argue below, the way the members of a social group will react to electoral appeals of various sorts is a function of how that group initially was mobilized into the electorate. Thus the analysis here must begin by considering the factors, apart from voter preferences, that influence the strategic choices that parties make. The Supply of Patronage The next of the above-mentioned factors cannot be laid aside quite so quickly. Clearly, a party must enjoy access to a pool or resources out of which patronage can be generated if it is to distribute patronage to its supporters. Now at first glance it would appear that there are two conditions under which a party will not enjoy such access: (1) if the leaders of the party themselves neither occupy public office nor are allied with elites who control some source of patronage; (2) if governmental agencies are protected by civil service statutes and other general laws that specify how public benefits and burdens are to be distributed and that thereby prevent politicians from intervening in the administrative process on a case-by-case basis. The first of these conditions is, indeed, sufficient to explain why parties founded by outsiders—by leaders who do not occupy positions within the preexisting regime—are compelled to rely upon ideological and solidary incentives before coming to power; it can explain why, for example, the major working-class parties of Europe relied upon inducements of this character in their early years. This condition itself, however, cannot explain why parties such as Labour in Great Britain, the Social Democrats (SPD) in Germany, or the Communists in Italy continue to eschew patronage appeals after they come to power, after they obtain access to the resources of the state. I will argue shortly that the circumstances of a party’s origins—whether or not it enjoyed access to patronage at the time it first undertook to mobilize a popular base—can crucially influence the party’s subsequent behavior, but in order to construct such an argument it is necessary to proceed further down our list of explanations. The second condition under which it would appear that parties will not enjoy access to patronage—if governmental agencies are protected by civil service laws—is suggested by Leon Epstein in Political Parties in Western Democracies.15 Epstein argues that patronage plays a lesser role in the party politics of European nations than it does in the United States because political parties emerged in Europe after the adoption of civilservice laws, while they emerged in the United States prior to the enactment of such laws.


105

For this reason Epstein suggests, politicians in Europe did not enjoy access to patronage during the era of party building, whereas their counterparts in the United States did enjoy such access and acquired a stake in the patronage system, a stake they were willing to defend by fighting rear-guard actions after civil-service laws were adopted. I shall argue below that the relative timing of these two events may be of crucial importance, but in order to see that Epstein’s formulation is not entirely satisfactory, one need only note that many of the most prominent patronage-oriented parties in the world today were founded or acquired mass constituencies after the adoption of civil-service laws. One thinks, for example, of the Christian Democrats in post-war Italy, the Cermak-Kelly machine in Chicago, and, if one’s universe of cases is broadened, the machine parties that emerged in newly independent Third World nations in the 1950s and early 1960s. The mere enactment of a civil-service statute is not sufficient to stanch the flow of patronage because such laws are not self-enforcing. Rather, if a civil-service system (or any administrative arrangement providing for the allocation of public benefits and burdens according to general rules) is to resist the depredations of patronage-seeking politicians, the administrators or public officials who would defend it must be backed by a constituency that has a stake in the system and that is sufficiently powerful to prevail over competing forces. Only if it fears arousing the opposition of such a constituency will a governing party be constrained to forego the immediate gains it would realize by directing bureaucrats to reward the party’s friends and to punish its enemies. The groups that oppose the patronage system, that insist that public benefits and burdens be allocated according to a set of general, universalistic, rules and procedures, and that seek to defend the autonomy of the bureaucracy from politicians who seek to intervene before it on a case-by-case basis, might be termed a “constituency for universalism” or a “constituency for bureaucratic autonomy.” Who comprises such a constituency? Its members wish of course to overcome what they regard as the evils of patronage politics, and patronage politics, as Carl Landé notes, undermines the regulatory and extractive capacities of the state.16 Constituencies for bureaucratic autonomy characteristically are organized by leaders who wish to strengthen the state in these respects. But these leaders generally must enter into coalitions with other forces if they are to have any chance of overcoming those who benefit from the patronage system. At the same time, general rules allocating public benefits and burdens are by definition neutral between individuals, but they need not be neutral between social groups. Leaders of a constituency for bureaucratic autonomy thus can broaden their basis of support by entering into a coalition with groups that seek through general rules to obtain privileged access to public offices and benefits. Elites and Cadres The concept of a constituency for bureaucratic autonomy leads us to the third consideration I mentioned above. A party, I noted, will be compelled to eschew a patronage strategy if the losses it would incur by using the resources of the state to generate patronage exceed the benefits it would realize by so doing, and it will have an incentive to adopt a patronage strategy if the benefits of so doing exceed the costs. A party will be driven in one direction or the other depending upon which group it can least


106

afford to alienate—the opponents of the patronage system (the constituency for bureaucratic autonomy) or its defenders. The latter grouping, to maintain terminological balance, can be labeled a “constituency for patronage,” and its leading members, as Epstein notes, are the politicians—ward bosses, notables, caciques, and askari—who are dependent upon the continued flow of patronage to keep a hold on to their clients and followers. There are many resources that these two groupings may be able to mobilize to back up their claims—wealth, notability, expertise, the capacity to disrupt or overthrow the government—but in an electoral setting the balance can be tipped decisively to one side or the other if one or the other of them is backed by a mass base, and the more extensive and the more highly organized that popular base the more decisive will be the advantage enjoyed by that side. If the patronage-oriented politicians within the party enjoy such support—as they do, for example, in Chicago’s Democratic party today—then “reform” will be the cause of only narrowly based elites, and the party would ignore the claims of its constituency for patronage at the peril of losing its base of popular support. If, to the contrary, the claimants for patronage within a party do not enjoy such backing then the party leadership can heed the demands of a constituency for bureaucratic autonomy without fear that it thereby is courting disaster at the polls. Whether a party will or will not be crucially dependent upon the distribution of patronage to maintain its hold upon its supporters is a function, in turn, of how the leadership of that party initially established a linkage with a popular base. If, for reasons I will discuss shortly, the party was not in a position to distribute particularistic benefits when it first undertook to mobilize mass support, its leaders will have been compelled to rely upon or to establish a network of mass organizations—labor unions, peasant leagues, churches, party sections—that did not need to be fueled by patronage. A party linked to a mass constituency through such an organizational structure will not, once it comes to power, be compelled to raid the public treasury in order to maintain its hold upon its supporters. On the other hand, a party that undertook to win popular support by distributing particularistic benefits through local notables or politicians will not have established such an organizational structure to bind voters directly to the party, and consequently such a party will only be able to maintain itself in office by heeding the demands of the patronage-seeking politicians who are affiliated with it. A “Critical Experience” Theory My argument, then, is this. Once a party does come to power, its tendency to adopt or eschew a patronage strategy is a function of the third of the above-mentioned considerations: whether it will gain more than it will lose if it intervenes within the administrative process on a partisan, case-by-case basis. Whether gains will exceed losses, or losses will exceed gains, depends upon the relative strengths of the elites and party cadre who have a stake in the patronage system, on one hand, and the elites and cadre whose interests are served by more universalistic systems of public administration, on the other hand. And the influence that these two constituencies will have in the inner councils of the party is a function, in turn, of how the party first undertook to mobilize popular support. In other words, the way in which a party initially acquires a popular base is a character-forming or “critical” experience, in the sense in which that term is used by


107

Philip Selznick and by Ira Katznelson.17 Political parties, as Samuel Huntington notes, are “formed by the organized linking of political faction to social force,”18 and the way that linkage initially is established influences the character of the organization the party builds, what it subsequently must do to hold on to its social base, and consequently the bargaining strength within the party of practitioners of patronage politics and of their opponents. Under what conditions will the founders of a political party use or not use patronage to establish a linkage with a mass base, that is, to acquire popular support? In order to answer this question, it is useful to recall a distinction that can be found in the literature on the origins of political parties, a distinction to which I have already alluded. Political parties, as Duvarger and Huntington have both observed, emerge in one of two circumstances.19 Parties have been founded by elites who occupy positions within the prevailing regime and who undertake to mobilize a popular following behind themselves in an effort either to gain control of the government or to secure their hold over it. Such parties might be termed “internally mobilized” parties. The Democratic-Republicans and the Federalists in the United States and the Liberals and Conservatives in Great Britain fall under this heading. Political parties also have been established by outsiders who did not hold positions within the prevailing regime and who organize a mass following either in an effort to gain entry into the political system for themselves and their supporters or in an effort to overthrow that system. Such parties might be termed “externally mobilized” parties. Socialist parties in Europe and nationalist parties in the Third World fall into this category. Because they are established by outsiders, political parties in this second category— externally mobilized parties—do not enjoy access to state patronage at the time of their founding and perforce are compelled to rely upon other means to acquire a following. The situation with regard to internally mobilized parties is more complex. Parties founded by elites that occupy positions within the prevailing regime will be in a position to use the resources of the state to acquire a mass base and will have every incentive to make use of that advantage unless the party undertakes to mobilize a popular following after a constituency for universalism has coalesced and become entrenched. The sequence in which these two events occur—the formation of a constituency for bureaucratic autonomy and the mobilization of a mass electorate—thus has enduring consequences for internally mobilized parties. One can identify two different constituencies for bureaucratic autonomy that historically have emerged in nations governed by internally mobilized parties—two different coalitions of elite groups that historically have sought to abolish patronage in the public service to establish civil-service systems purportedly based upon merit-related standards of entry and promotion, and to defend thereafter the integrity of such systems.20 The first of these, which might be termed the “absolutist coalition,” was led by monarchs who during the seventeenth and eighteenth centuries sought to increase the extractive and regulatory capacities of their regimes and to create an army so that they would be better able to meet foreign military and economic competition and to bring their opponents at home to heel. This entailed creating a modern, centralized, bureaucratic state to replace the decentralized, patrimonial standestaat of the earlier era. The way modernizing monarchs proceeded toward this end varied from place to place, but generally speaking


108

they won some crucial political support for this program by filling positions in the reformed administrative system with members of the haute bourgeoisie and the aristocracy, that is, by creating a new administrative elite that was drawn in part from and had links to older elite groups. These groups continued in the democratic era to serve as the constituency for the reformed state apparatus. The second coalition might be termed the “progressive coalition” because the role played by absolute monarchs in the first coalition was played in the second by a rationalizing middle class, a middle class whose ideology was Benthamite in England, Positivist on the continent, and Progressive in the United States. Although the leading elements of the first and second coalitions differed, their remaining members were drawn from similar groups. In order for a rationalizing bourgeoisie to bring about administrative reforms and thereafter to be successful in providing political backing for a reformed civil service, it had to draw support from elements of the groups—be they aristocrats, patricians, or less exalted public servants— that held positions under the earlier regime. The upshot of this analysis is that the circumstances under which a party first mobilizes a mass following has enduring implications for its subsequent behavior. To summarize: externally mobilized parties will tend to eschew the use of patronage (a tendency that will be stronger the greater the resistance the party must overcome in order to gain power) regardless of the social composition of their electoral base. Internally mobilized parties will tend to be patronage oriented unless they operate in a setting where either an absolutist or a progressive coalition became entrenched prior to the mobilization of the masses into politics. The circumstances of a party’s origins influence its subsequent behavior regardless of the social characteristics of the voters upon whose support the party relies because the formative experiences of a party have implications for the three factors that, as I noted earlier, determine whether a party will adopt or eschew a patronage strategy: (1) the demands and expectations of the party’s rank-and-file supporters; (2) the material and organizational resources available to the party; and (3) the orientations of the party’s leadership and its cadres and the interests of the elites who are allied with the party or who are capable of sanctioning it. Externally Mobilized Parties The relationship between party origins and subsequent behavior is quite straightforward in the case of externally mobilized parties: the more “external” the circumstances of the party’s origins—the fewer the allies it had within the preexisting regime, the greater the social and ideological distance between the party’s founders and that regime, the greater the resistance the party is compelled to overcome in order to gain power—the less likely it is the party will decay into a patronage machine after it does come to power. This relationship is clear cut because the birth pangs of such a party, if they are severe, can permanently shape its character. Consider the extreme case: parties whose founders have no significant allies among power holders or other elites. The leaders of such a party will not enjoy access to governmental or other sources of patronage and will find it necessary to rely upon other appeals to mobilize their supporters; it is precisely under these conditions, of course, that Puritanism emerged as the ideology of a politically excluded gentry (and Calvinism as


109

the ideology of a politically excluded provincial nobility) in the seventeenth century, that liberalism emerged as the ideology of a politically excluded bourgeoisie in the eighteenth century, that socialism emerged as the ideology of a politically excluded working class in the nineteenth century, and that anarchism emerged as the ideology of a politically excluded peasantry in the twentieth century.21 The fewer the allies such a party enjoys within the preexisting regime, moreover, the more strongly is that regime likely to resist the party’s claims, to the point of seeking to repress it. In this situation, rank-and-file members of the party will be compelled to endure considerable sacrifices in order to affiliate with the party; the party will find it necessary to construct a strong organization to link its members and leaders to one another; and, most important of all, the party will be compelled to mobilize an extensive mass base in order to attain power. These early experiences have enduring implications for the party’s standing along each of the three above-mentioned dimensions. First, voters who undertook major sacrifices to affiliate with a party prior to its coming to power are not likely to demand particularistic payoffs as a condition for their continued support after the party comes to power. Nor, once such voters are brought into the electorate, will they be available for bribery by machine politicians from other parties, for, as S.M.Lipset and Stein Rokkan have observed, the party that first brings a social group into the political system retains a privileged hold upon the members of that group for decades, and even generations, thereafter.22 Finally, if the party mobilized an extensive mass base prior to coming to power, it will not, in order to stay in office, find it necessary subsequently to seek additional support among groups of voters that did not share the experience of affiliating with the party when it was an opposition force, and whose devotion to the party, therefore, is not quite so strong. As for the second of the above-mentioned considerations—the material and organizational resources available to the party leadership—party leaders who construct a strong organization prior to gaining access to patronage (an organization that necessarily will be grounded upon ideological and solidary inducements) will not subsequently find it necessary to fuel that organization with patronage in order to keep it running and to maintain a linkage with their subordinates and followers. Moreover, a leader who commands such a strong party organization will not find it necessary to tolerate corruption and empire building on the part of his subordinates; he will be in a secure enough position to discipline and even to fire officials who use public authority to build personal political followings.23 Third, and finally, the leaders and cadres of such a party themselves will comprise a constituency for universalism, and the party will not be beholden to any constituency for patronage. The leaders and cadres of an externally mobilized party, as I have suggested, are likely to be committed to an ideology—a vision of society—and once they come to power, they are not going to be willing to fritter away in patronage the authority they now have to remold society, at least not if it was only with great difficulty that they got hold of that authority. And if these leaders acquired a mass following prior to gaining access to the resources of the state, and if the party did not rely upon the support of local elites and notables in order to acquire that following, it will not subsequently be dependent upon the support of patronage-oriented politicians in order to retain its popular base. This set of considerations can explain why the European socialist parties listed in the table above—parties that acquired mass followings before they gained access to


110

patronage—did not decay into political machines after winning control of municipal, or eventually national, governments. And this has been true of these parties regardless of the social composition of their electoral bases, even when their supporters were drawn from groups (uprooted peasants, recent urban migrants, Catholics, Latins) whose members in their private and social lives are said to have narrow, family-centered loyalties. At the other extreme are those parties that did not find it necessary to overcome substantial resistance from the powers-that-be as a condition of coming to power, a category that includes the nationalist parties that took control of most Third World nations in the immediate postindependence period. These parties were not compelled to mobilize a substantial mass following or to build a strong party organization before they were handed the keys to the public treasury. Consequently, when they for the first time did find it necessary to mobilize popular support—in elections that were sponsored by the colonial power or that were held after independence was granted—these parties were an element of the prevailing regime and did enjoy access to the public bureaucracy. Having no alternative means of reaching into the countryside (or even into the shanty-towns around the capital city) it is not surprising that they would attempt to acquire a following by making use of the patronage at their disposal. In other words, at the moment when they first undertook to acquire a mass base, these parties for all intents and purposes stood in the same relationship to potential supporters as do internally mobilized parties. It is this consideration that explains why students of political development have found patronage to be so prevalent in the party politics of Third World countries, including those countries (the Philippines, Ghana) listed in the upper left of the table above.24 (By contrast, those Third World parties that were compelled to fight their way to power, for example, Tunisia’s Neo-Destour, are less patronage oriented, even when they rely upon peasants or migrants for much of their popular support.)25 Internally Mobilized Parties In a similar vein, the way in which an internally mobilized party initially establishes a linkage with its popular base has important consequences for its subsequent character. In order to specify the nature of this relationship one must divide the universe of internally mobilized parties into two subcategories: parties that undertake to acquire a popular following after the formation of an absolutist or a progressive coalition, and parties that mobilize a mass base before a constituency for bureaucratic autonomy of one or the other of these types coalesces. Parties in the former subcategory will tend thereafter to eschew patronage appeals, parties in the latter subcategory will tend thereafter to be patronage oriented. The relationship between party origins and subsequent behavior, however, is somewhat looser in the case of internally mobilized parties than it is with externally mobilized parties because the implications of the early experiences of the former for the subsequent orientations of the party’s rank-and-file supporters and the subsequent character of the party’s organizational structure are not quite so strong as they are with parties whose formative years were marked by the stuff out of which enduring loyalties and strong institutions are made: the shared experience of intense struggle and conflict. The formative experiences of an internally mobilized party do, however, strongly influence whether the party in subsequent years will be more beholden to a constituency


111

for bureaucratic autonomy or a constituency for patronage, and for this reason in particular internally mobilized parties falling into the first subcategory are less likely to be patronage oriented than those falling into the latter subcategory. In the remainder of this article I will seek to substantiate this proposition. To anticipate, I shall argue that in nations where either an absolutist or a progressive coalition emerged prior to the mobilization of the masses into politics, internally mobilized parties will not have been in a position to use patronage in order to establish a linkage with a mass base but rather will have been compelled to rely upon other appeals (e.g., to patriotic or religious sentiment) and other forms of organization (e.g., churches, interest groupings). Because the party did not acquire a popular following by distributing patronage through a network of local politicians it will not thereafter be crucially dependent upon the support of such patronage-seeking politicians. And the very emergence of an absolutist or progressive coalition in the nation in question means that the agrarian and/or urban upper classes with whom internally mobilized parties are allied will have a stake in the rules and procedures that guarantee these groups control over the reformed bureaucracy and privileged access to positions within it. For these reasons the party leadership will have every incentive to avoid the use of patronage: they are allied with a constituency for bureaucratic autonomy and are not dependent upon the support of a constituency for patronage. These considerations explain why the European conservative parties listed in the table—Germany’s Christian Democratic Union/Christian Socialist Union and Britain’s Conservatives—today eschew the use of patronage (an absolutist coalition emerged in Germany and a progressive coalition in England before the creation of a mass electorate). More generally, I suspect (though I am not certain) that these considerations explain why patronage plays the smallest role today in the party politics of those European nations that, in Immanuel Wallerstein’s terminology, were governed by the “strong core states” of the sixteenth-century worldsystem.26 (I am prepared to argue, but will not do so here, that this consideration explains why patronage plays a relatively small role in the politics of cities in the South and West in the United States and why it plays a relatively small role in state politics in the American upper Middle West and trans-Mississippi West, even when the parties in question rely upon the votes of working class immigrant voters.) On the other hand, in nations where neither an absolutist nor a progressive coalition emerged prior to the extension of the franchise internally mobilized parties will enjoy access to patronage at the moment when they first undertake to acquire a popular following and, for the very reason that they are not confronted with a constituency for bureaucratic autonomy, they will have no incentive to avoid taking advantage of that access. Rather, they have every incentive to acquire popular support by distributing patronage to notables and politicians who have local followings. Such a strategy both follows the line of least resistance—it can be accomplished far more rapidly and with far less difficulty than building a party organization from scratch—and has the advantage of enabling the elites who are affiliated with the party to secure their hold over their clients and even to increase their power locally. These politicians and notables will become dependent upon the continued flow of patronage to maintain their power locally and within the party—they will develop into a constituency for patronage—and because the party is linked to its popular base through these politicians, the party will be dependent


112

upon this constituency to maintain itself in power. A party whose political fate is so tied to the patronage system—one that requires patronage to hold its organization together and to maintain the support of its voters and its cadre—will not be in a position to respond favorably to a constituency for bureaucratic autonomy if and when such a constituency does subsequently emerge. These considerations, I would argue, explain the patronage orientation of the governing parties in southern Europe (the most notable being Italy’s Christian Democrats) and they explain the significant role that patronage plays in the party politics of those states of the American East and lower Middle West (New York Pennsylvania, Illinois) listed in the above table, and in the major cities in these states (New York City, Philadelphia, Chicago). In each of these cases, moreover, patronage plays a role in binding elements of the middle class, as well as other voters, to the party in question. In the remainder of this article I will provide evidence for this argument concerning internally mobilized parties by analyzing three cases. The first, Germany, is a case where an absolutist coalition emerged prior to the creation of a mass electorate; the second, England, is a case where the formation of a progressive coalition preceded the full mobilization of the masses into politics; and the final one, Italy, is a case where neither of these coalitions emerged prior to the creation of a mass electorate. The order in which these events occurred, I shall argue, can explain the greater or lesser role that patronage plays in party politics in each of these nations today.

German Germany stands as perhaps the paradigm case of a nation in which an absolutist constituency for bureaucratic autonomy coalesced and became entrenched in the predemocratic era, and it illustrates quite clearly the consequences of this development for the subsequent evolution of party politics. In the sections below I will discuss the emergence of an absolutist coalition in Brandenburg-Prussia during the seventeenth and eighteenth centuries, and I will argue that the ability of this coalition to survive the changes of regime Germany experienced subsequently, and to defend the autonomy of the bureaucracy, has been responsible for the limited role that patronage has played in the political parties of the Second Empire, the Weimar Republic, and the Federal Republic— including those parties that have relied upon the support of voters who supposedly are most responsive to patronage appeals. The Emergence of an Absolutist Coalition The attack upon a decentralized administrative structure staffed by aristocratic patronage dates in Germany from the aftermath of the Thirty Years War.27 Germany was devastated in that war and the Hohenzollern rulers of Brandenburg-Prussia, especially Frederick William (1640–88), King Frederick William I (1713–40), and King Frederick II (1740– 86), attempted to recoup by creating a powerful standing army, an army that would defend the integrity of the Hohenzollern domains and lift Prussia into the ranks of the Great Powers.28 The Great Elector recognized, however, that the Estates of the Realm,


113

which dominated the existing governmental structure, would refuse to tax themselves in order to finance a large standing army, especially one under the control of the king. Consequently, he created an administrative apparatus that would enable him to bypass the estates, to extend his control over his domains, and to generate the revenues necessary to finance his army. In recruiting officials to serve in this bureaucracy the Hohenzollerns attempted to accomplish a number of things. First, they sought to find royal servants who would be personally loyal to the king and who would support the crown rather than the estates in the struggle for control over Prussia. Second, they wanted their officials to be competent, to be adept at performing the tasks they would be assigned. And finally, in making appointments the Hohenzollerns sought to build a social base for their regime, to induce acquiescence among those whose support was necessary if their venture was to succeed. These goals had conflicting implications for the recruitment process. The best way to insure that civil servants would be loyal to the crown would be to appoint commoners and foreigners to the bureaucracy, for they would be utterly beholden to the king for whatever status and power they enjoyed; the preservation of their status would be contingent upon the triumph of the crown in its struggle with the estates; and they had fewer ties to the estates than did officials who were aristocratic by birth.29 An effort to maximize the competence of the bureaucracy had roughly similar implications, both because such an effort implies that one not give favor to candidates by virtue of their birth and because most university graduates in Prussia were of bourgeois, rather than noble, birth.30 On the other hand, the necessity the Hohenzollerns faced of maintaining support for their regime and for their efforts at state-building had different implications for the recruitment process. The aristocracy was the class the monarchy could least afford to alienate, because they possessed the means to forcibly resist anything they regarded as tyrannical and because they were the class from which the officer corps of the Prussian army was to be drawn. For these reasons the kings of Prussia had an incentive to appoint Junkers to positions in the state bureaucracy.31 The Hohenzollerns pursued these several goals by recruiting civil servants from both the bourgeoisie and the aristocracy. The relative standing of these two classes within the bureaucracy fluctuated as political conditions changed.32 In particular, Napoleon’s victory over Prussia at the battle of Jena in 1806 discredited the army and the aristocracy and enabled a group of reformers led by Stein and Hardenburg and backed by the middle classes to come to power.33 The reformers of the Stein/Hardenburg era insisted that national regeneration could only be achieved if greater weight was accorded to the claims of talent and education than to those of birth. A system of examinations for entry into the bureaucracy was devised to determine whether candidates for civil-service appointments possessed these traits, examinations that required young men who wished to pursue a career in the bureaucracy to acquire a classical and legal education. Such an education could only be obtained by attending a classical gymnasium and, then, a university. This requirement limited careers in the upper civil service to the wealthy, but it did not discriminate between those whose family wealth came from land and those whose wealth came from commerce. The ideology of talent and education, of Bildung, and the civilservice examination system through which it was institutionalized, thus enabled the bourgeoisie to solidify its claim upon positions within the bureaucracy. It enabled them to


114

do so, however, without utterly alienating the aristocracy, for the civil-service examination system accorded the aristocracy representation within the bureaucracy far out of proportion to their numbers, and the ideology of Bildung provided a justification for the disproportionate power that, after Jena, they sorely needed. The civil-service examination system and the ideology of Bildung served a final purpose, a purpose that helps explain the political utility of these practices and principles to the groups that defended them. As Hans Rosenberg notes, these procedures and principles enhanced the autonomy and power of the bureaucracy, and hence the value of bureaucratic positions to those who could obtain them, because it gave the civil service control over its own recruitment and thereby prevented outsiders from influencing the bureaucracy through the hiring, firing, and promotion of civil servants.34 It was through these processes, then, that the Junkers and elements of the upper middle class obtained privileged access to positions in the bureaucracy and acquired a stake both in the regulations that gave them that access, and in protecting the autonomy of that bureaucracy against “outside interference.” In other words, these groups became a constituency for bureaucratic autonomy. The Second Empire This absolutist coalition retained its stake in the civil service under the Second Empire, as parliamentarians and party politicians became the major “outsiders” against whom the autonomy of the bureaucracy had to be defended. Otto von Bismark, the most talented representative of Prussian Junkerdom and the architect of German unification, was flexible in a number of respects, but he was intransigent on this point. Bismarck was prepared to make various concessions as circumstances dictated to industrialists, Catholics, and even the working class, but he was not prepared to concede to any of these groups, or to the politicians who spoke for them, the slightest control over the army or the civil bureaucracy.35 The constitution of the Second Empire provided that the chancellor and the cabinet were to be responsible to the Kaiser alone and, correlatively, that the Reichstag would have no control over the ministry and the bureaucracy. Practice did not deviate from constitutional precept during the Second Empire. Parliamentary politicians were scarcely accorded any access to the very top level of government let alone to the intermediate and lower levels of the bureaucracy: the majority of all cabinet ministers were recruited from the civil bureaucracy and the army rather than from the Reichstag. Large numbers of bureaucrats obtained seats in the parliament, while it was not possible for parliamentary and party politicians to obtain positions for themselves or their supporters within the bureaucracy.36 Under the Second Empire, then, Germany was ruled by what Otto Kircheimer has termed a “militarybureaucratic complex” and the social groups linked to this complex would not countenance any interference in what they regarded as their private preserve from the politicians who were their agents in the electoral and parliamentary arenas, let alone from politicians who belonged to opposition parties.37 The parties allied with the Junkers and/or the upper middle class and enjoying the closest ties to the government—the German Conservative party, the Reichspartei or Free Conservatives, the National Liberals—for this reason made no effort to use what power the Reichstag did possess


115

under the constitution—its budgetary and legislative powers—to extend parliamentary and party control over the bureaucracy. Rather than attempting to use patronage to appeal to the working classes and the peasantry, the Conservatives instead sought to win such support by allying first with Adolf Stoecker’s anti-Semitic Christian Social movement and then with the Farmer’s League.38 As for the nonregime parties of the Second Empire—the Social Democrats, the Catholic Center, and the Progressives—they simply were not in a position to raid the bureaucracy and obtain patronage even if their leaders had cared to do so. Weimar Germany’s defeat in the First World War led to the collapse of the monarchy and to the destruction of the institutions that had denied political parties any influence over the bureaucracy and that had prevented the members of all but the highest social classes from gaining entry into the upper reaches of the civil service. During the early months of the revolution of 1918 the political parties sought to take full advantage of this new situation: they attempted to penetrate and gain control over the administrative apparatus of the old regime and to use whatever influence they managed to obtain within it to strengthen their organizations.39 The revolution of 1918, however, was aborted. In order to crush a wave of left-wing insurrectionary activity, the early socialist governments of Weimar found it necessary to turn to military units from the old regime, and in order to carry on routine tasks of administration the government was compelled to rely upon the administrative apparatus and the bureaucrats it had inherited from the Reich. The military-bureaucratic complex that had governed the empire thus was able to survive into the new era. And because the German revolution was unfinished—because the economic and the political power bases of the Junkers and of the upper bourgeoisie had not been destroyed—the class coalition that defended the autonomy of this complex continued to be a potent political force during the Weimar period. The desire of the Republican parties to make peace with these political forces was sufficiently strong that they did not attempt to change in any fundamental way the structure, recruitment procedures, and social composition of the civil service.40 Indeed, in Ralf Dahrendorf’s judgment, the influence of administrative officials relative to other political actors was greater in the Weimar period than it had been under the previous regime.41 Unregenerate opponents of the Republic were so firmly entrenched in Weimar’s administrative apparatus that adventurers such as Adolph Hitler were able to conspire against the regime, as in the Munich Beer Hall Putsch, without being seriously punished by the Weimar judiciary.42 The survival of Germany’s absolutist coalition into the Republican era had a number of important and ultimately fatal consequences for political life in Weimar, two of which are of concern to us here. In the first place, this constituency for bureaucratic autonomy was allied with parties of the Right—especially with the German National People’s party (DNVP) and the German People’s party—and it thereby exercised an internal constraint upon whatever tendency these parties might have had to raid the bureaucracy in search of patronage. This constraint was particularly notable with respect to the DNVP, which sought to mobilize support among the peasantry and which therefore might have been


116

expected to rely upon a patronage strategy, but which relied instead upon nationalist and anti-Semitic appeals.43 It was, rather, the Center party, which did not have ties to the absolutist coalition, that appears to have been the most inclined of the parties of the Weimar period to use its positions within the government to influence bureaucratic appointments, in an effort to increase the representation of Catholics in the bureaucracy. This endeavor, however, met with only limited success.44 The survival of Germany’s absolutist coalition into the Republican era exercised an external constraint upon all of Weimar’s parties as well as an internal constraint upon the parties with which it was allied. Republican politicians dared not overturn practices and procedures that protected the autonomy of the bureaucracy and that gave the upper classes privileged access to high positions in it for fear that if they attempted to do so, the social groups that were linked to the bureaucracy would turn openly against the Republic. For this reason the parties of the Weimar coalition—the Socialists, the Democrats, and the Centrists—decided against purging the bureaucracy of anti-Republican elements in the wake of the Kapp Putsch in 1920.45 It is true that after the assassination of Walter Rathenau in 1922 a law was passed that increased the number of administrative positions that were subject to political appointment, and the Republican parties sought to “democratize” the civil service by appointing their adherents to these positions, but these parties hesitated to take full advantage of the opportunities provided by the 1922 law for fear of provoking the regime’s opponents.46 The proof of how discrete the Weimar parties had been came after the Nazi seizure of power. The Nazis in April 1933 promulgated the Law on the Reestablishment of the Career Civil Service, which provided for the removal from the bureaucracy of Jews, Communists, and “party-made” officials, that is, officials who had been brought into the civil service since 1918 without filling the normal requirements of the positions they held. But very few civil servants turned out to be liable for dismissal under the terms of this law.47 The existence of these severe internal and/or external constraints upon the freedom of action of political parties in Weimar Germany can explain why these parties were not, in the judgment of contemporaneous political scientists, especially patronage oriented.48 The Third Reich The Nazi revolution did not destroy the autonomy of the German civil service or sever the links between the bureaucracy and its traditional social base any more than had the revolution of 1918. Indeed, to the extent that the Nazis did purge from the bureaucracy elements that had entered it since 1918, the initial consequence of Hitler’s coming to power was more restorative than revolutionary. However, after the limited purges of 1933, Nazism exerted surprisingly little impact upon the civil service, as David Schoenbaum’s study indicates.49 The civil service was so successful in maintaining its autonomy because the vast majority of all civil servants inherited by the Third Reich conscientiously served the new regime and were loyal to it. The regime thus had little incentive to disrupt established bureaucratic routines or to insist that party considerations be accorded priority over technical proficiency in the appointment process. For this reason, as Schoenbaum notes, “despite the ostensible merger of State and Party, the outstanding loser in the struggle for


117

survival was the Party. Advantages to Alte Kampfer [Nazis who had joined the Party before 1933] were kept to a minimum; direct party interference in the traditional civil service sectors was eliminated almost immediately. The files of the Ministry of the Interior could be read as a series of jurisdictional conflicts in which, almost without exception, the civil service won and the party lost.”50 Because the Third Reich had pretty much left the German civil service alone and did not substantially alter its social composition, the heirs of Germany’s absolutist coalition were able to defend the autonomy of the bureaucracy when an open political system was restored in the Federal Republic after the Second World War. The Federal Republic The civil service of the Federal Republic is the lineal descendent of the bureaucracy that served earlier German regimes—the Third Reich, the Weimar Republic, the Wilhelmian Reich, and ultimately, Hohenzollern Prussia—both in law and in fact. As for juridical continuity, a federal law of May 1951 passed pursuant to Article 131 of the Bonn constitution granted all regular civil servants, including those dismissed after 1945 for their Nazi connections and those expelled by the East German regime, a right to reinstatement by the Federal Republic.51 And more than ten years after the collapse of the Nazi regime 60 percent of the individuals who occupied top civil-service positions in the Federal Republic had held comparable positions in the administrative structure of the Third Reich.52 In fact, the proportion of civil servants who survived the regime change in 1945 was identical to the proportion who survived the transitions from the Wilhelmian Reich to Weimar in 1918 and from Weimar to the Third Reich in 1933. I stated above that the constituency for bureaucratic autonomy in the Federal Republic today is the heir of Germany’s absolutist coalition, rather than that it is identical to the coalition that defended bureaucratic autonomy in earlier regimes, because I would not deny that German society has changed in many important respects in the past several decades. The combined impact of industrialization, wartime decimation, and postwar partition destroyed the Junkers as a social class. But the two centuries of bureaucratic absolutism that preceded (and brought on) many of these changes have had their effect. The creation of an absolutist coalition implied not only the militarization of the Prussian aristocracy but also the “feudalization” of the Prussian bourgeoisie—the emergence of a bourgeoisie d’état that aspired to assimilate into the aristocracy, that adopted its étatist values, and that has carried on this tradition even though the class that had previously exercised moral hegemony over them has passed from the scene. It is this class (and those who now aspire to enter it), with its orientation to public careers, its high level of corporate consciousness, and its conviction that it stands “above politics,” that has defended the autonomy of the civil service in the Federal Republic.53 This constituency for bureaucratic autonomy has been an enormously powerful force in West German politics. The ability of the civil service to survive the collapse of the Third Reich and the Allied military occupation with only minimal changes in personnel and structure is a reflection of that political strength.54 Upon assuming control of Germany the occupation authorities, especially the Americans, intended to bring about the “denazification” and the “democratization” of the civil service; that is, they intended to


118

purge and to reorganize the bureaucracy. The responsibility for carrying out these reforms, however, was left to the German authorities, and politicians in postwar Germany found it politically impossible to resist the pressure of defenders of the bureaucratic status quo. The Deutscher Beamtenbund (DBB), the major spokesmen for the higher civil servants, was particularly insistent in demanding that the “vested rights” of the Beamte be honored. One measure of its success is that only one thousand of the fifty-three thousand civil servants who were initially removed in denazification proceedings were in the end permanently excluded from the bureaucracy, and ultimately the Federal Republic passed the law of May 1951, which I described above. The power of Germany’s constituency for bureaucratic autonomy is such that, as Taylor Cole has observed, the civil service was the public institution in Germany that passed through the occupation with the fewest changes.55 The political power of Germany’s constituency for bureaucratic autonomy in the postwar era has rested upon a number of foundations. First, as the Socialists learned in 1918, as the American, British, and French occupation forces rediscovered in 1945, and as politicians in West Germany have recognized to this day, any regime that is not prepared to countenance a sustained period of disruption must delegate the task of administering the routine functions of government to the civil servants it inherits from the previous regime. The critical position that it occupies gives Germany’s traditional bureaucratic elite enormous bargaining power.56 Second, civil servants in Germany are well organized and rely upon pressuregroup activity to assert their claims and to defend their position in the politics of the Federal Republic.57 Third, the civil service is closely linked with other institutions in German society, most notably with the legal profession and the universities.58 A fourth and final practice that strengthens the defenders of bureaucratic autonomy in Germany is one that might be termed “reverse colonization.” Political parties are prevented by law from placing their agents in the career civil service; career civil servants, however, have colonized the parties and the public institutions—especially the Bundestag, the Bundesrat, and the Landtag—of the Federal Republic. This practice has its roots in the nineteenth century, and the DBB fought hard during the occupation, ultimately with success, against the Allies’ proposal that a law be passed, akin to America’s Hatch Act, that would severely restrict the political activities of civil servants. The DBB recognized that the practice of reverse colonization enabled the civil service to gain representation for its corporate interests in the inner councils of the Federal Republic.59 Because the defenders of bureaucratic autonomy have links to so many important institutions in German society, any party that expects to govern the nation finds it necessary to come to terms with this constituency—unless, as was true of the SED in the East, it is prepared to sponsor a social revolution. This is doubly true of a party such as the Christian Democratic Union (CDU) that is allied with the social groups that are advantaged by the traditional recruitment procedures of the German civil service. The relationship between these groups and the CDU goes a long way toward explaining its behavior toward civil-service reform and toward patronage in the postwar era. The stance of the CDU toward the civil-service reform proposals of the Allies was, as Karl Hochswender puts it, “moderately conservative.”60 Adenauer and the other top


119

leaders of the party rejected the idea of totally overhauling the civil service and insisted that the public personnel system of the Federal Republic be built upon the German Career Officials Act of 1937. As leaders of the ruling party in the Republic, they were, however, interested in increasing the control that supervisors were able to exercise over their subordinates within the bureaucracy and in increasing the technical proficiency of the civil service. Hence they supported reforms that would enable supervisors to deny regular salary increments to and to demote officials who failed to perform their duties satisfactorily and that would provide “open competition” for civil service positions, that is, would permit individuals who were not career officials to compete for appointments in the upper reaches of the civil service. These proposals, however, encountered enormous resistance among spokesmen for the Beamte within the CDU, especially among the CDU members of the Bundestag’s Civil Service Committee. The committee defeated the first, and the provision for open competition was able to pass only when some CDU members broke with a majority of their colleagues and voted with the SPD delegation in favor of it. The CDU’s alliance with Germany’s constituency for bureaucratic autonomy also can account for the fairly limited role that patronage played within the party during the decade and a half when it dominated the government of the Federal Republic. The CDU during this period was not totally indifferent to the identity of upper level bureaucrats, but as Hochswender notes (citing Theodor Eschenburg’s study of the scope of patronage in the politics of the Federal Republic) the concern of the CDU leaders in these matters was limited to two issues.61 First, in order to hold together the coalition governments that they led, the leadership of the CDU took care to see that the individuals appointed to upper level positions within the bureaucracy—the “politische Beamte”—as well as those appointed to the cabinet itself reflect the political composition of the government. Thus ambassadorships were allocated among the CDU’s coalition partners. Second, as a quasiconfessional party, the CDU pays heed to the religious affiliations of civil servants, especially those in sensitive positions. Thus the position of chief of the Cultural Division of the Foreign Ministry is reserved for a Catholic while the chief of the Cultural Division of the Ministry of the Interior invariably is a Protestant, and religious elements within the CDU have argued that in Catholic areas chief gynecologists in public hospitals and teachers in public schools should be Catholics. These appear to be the only arenas, however, in which the defenders of the traditional recruitment practices of the German civil service have been compelled to make concessesions to partisan and electoral pressures.

Great Britain An attack upon the patronage system may be spearheaded by a rationalizing bourgeoisie as well as by the forces of absolutism, and a coalition between such a grouping and older political and social elites—a progressive coalition—may thereafter serve as a nation’s constituency for bureaucratic autonomy. Parties will eschew patronage strategies in political systems where such a coalition emerges prior to the mobilization of a mass electorate. Great Britain was the first country where this occurred, and thus I will illustrate this proposition by discussing the British case.


120

The Emergence of a Progressive Coalition For reasons that need not be discussed here, a centralized bureaucratic state, headed by an absolute monarch and supported by a class of officials drawn from the aristocracy and haute bourgeoisie, was not established in England in the seventeenth century as it was in many nations on the continent.62 The English Civil War prevented such a development and confirmed that public administration would remain decentralized in Great Britain in the era prior to the extension of the franchise. The major offices of domestic administration—most importantly, the justices of the peace—and the major programs of domestic policy—the maintenance of public order, the administration of local justice, and poor relief—were controlled by local gentry and their dependents. Those public functions administered centrally were controlled by a parliament that itself was a “committee of landlords.”63 And appointments to these administrative positions were made in a manner calculated to cement the parliamentary majority of the incumbent ministry: the eighteenth century was the classic era of patronage in British politics.64 The governing classes of England, then, were not tied to a bureaucratic regime. In the absence of any such constituency for bureaucratic autonomy, party politics, or at least the politics of parliamentary parties,65 was patronage ridden. The leadership in the fight against the patronage system was assumed by elements of Britain’s professional upper middle classes, by a group that can appropriately be labeled a “rationalizing bourgeoisie.” Those who sought to reform the home civil service by ending patronage appointments and by introducing competitive examinations were part of a more general movement to rationalize and renovate the major public and quasi-public institutions in Britain—the universities, secondary schools, the administration of India, municipal administration, the system of poor relief, and so on. Within this broader group, however, one can distinguish two subgroups. The first of these was oriented toward the concerns of the older, more established segment of Britain’s middle classes. They sought to rationalize the nation’s traditional institutions—the ancient universities, the elite secondary schools, the imperial administration—so that these institutions would be more attuned to the interests of England’s liberally educated professional elite.66 Part and parcel of this effort was the proposal, embodied in the Northcote-Trevelyan Report of 1854, that admission into the highest grade of the civil service be determined by an examination geared to a literary education such as could be obtained only at Oxford and Cambridge. This proposal would have limited access to upper-level positions in the civil service to the classes that sent their sons to these two universities, namely the aristocracy, the gentry, and established members of the liberal professions. The second group of reformers was oriented more toward the concerns of the emerging entrepreneurial class. They attempted to rationalize the way in which the government dealt with the industrial sector of society. In particular, the leaders of this second stream pressed the government to assume the burdens of establishing the social preconditions for industrialization and of dealing with the externalities of industrialization. Men such as Edwin Chadwick, Horace Mann, and John Stuart Mill sought to socialize the costs both of educating, disciplining, and maintaining the health of the industrial labor force and of dealing with the “human exhaust of capitalism,” the poor, the criminal, the


121

dispossessed.67 This second group of reformers, like the first, sought to change the way in which civil servants were recruited, but since the goals they sought to accomplish differed, their proposals differed accordingly. Edwin Chadwick, John Stuart Mill, and the Administrative Reform Association (which was organized by a group of businessmen) proposed that civil-service examinations be instituted that would favor individuals who possessed technical skills, not those who were liberally educated.68 An examination system of this character, they argued, would increase the technical proficiency of government agencies, and it would loosen the grip of the upper classes upon the civil service, a grip that the Trevelyan-Northcote proposals would tighten. The Trevelyan-Northcote proposals of 1854 were defeated because they failed to win the support of the entrepreneurial classes and the aristocracy. When a plan was fashioned in 1870 that was backed by these two classes, civil-service reform was adopted with absolutely no controversy. The 1870 reform plan divided the civil service into three grades—administrative, executive, and clerical. Examinations for the administrative class were to be keyed to a liberal education, while examination for the executive class were to be keyed to an “English education,” that is, an education in the English language and in modern subjects, as opposed to one in ancient languages and classical texts. In other words, the 1870 program embodied the proposals both of the reformers who were oriented toward the established professional class and those who were oriented toward the entrepreneurial class, and consequently both segments of the bourgeoisie could unite behind it. As for the landed classes, their support for civil-service reform in 1870 can be attributed to the suffrage extension of 1867.69 Once the franchise had been extended to householders and rent payers, the aristocracy and gentry no longer could assume that the patronage system would continue to provide them with preferential access to the civil service; its tendency, rather, would be to open public employment to the most numerous class of voters, to democratize the civil service. The reform of 1870, however, by restricting appointments in the administrative class to those with a liberal education, would preserve the preferential access to the higher civil service traditionally enjoyed by the landed classes, for they could afford to give their sons such an impractical education. To be sure, an examination system for entry into higher civil service would exclude the indolent, and it would reward only those members of the aristocracy who had acquired habits of concentrated work and who had attached themselves to the service ideal. But under the prodding of schoolmasters such as Thomas Arnold and Oxbridge dons such as Benjamin Jowett, the British aristocracy in the nineteenth century did make these adjustments to bourgeois values and did adopt the ideology of service, an ideology that justified their retaining their privileges and power in the industrial era.70 These concessions staved off far greater losses. The Party Response: The Conservatives Patronage in the civil service finally was restricted in Britain, then, and a system of competitive examinations was instituted, when a reform plan was fashioned that could be supported by the landed and entrepreneurial classes as well as by professional elites. As such a constituency for bureaucratic autonomy emerged, the parties that depended upon


122

the support of these classes were compelled to adjust their behavior accordingly. This adjustment was bound up with the response of these parties to the suffrage extensions and kindred reforms71 that occurred at the same time—a conjunction of events that, as I have already suggested, was not in the least coincidental. The major parties in Britain did not adjust to these reforms without internal conflicts and difficulties. But they responded far more smoothly, and far more completely, than has been the case in nations where a constituency for bureaucratic autonomy only coalesced after the masses had been fully mobilized into the political system. In the paragraphs below I will provide evidence for my argument by discussing the case of the Conservative party. As late as 1868, the Conservatives in most constituencies did not command a party apparatus that had an existence apart from the patron-client chains that linked Tory squires to their tenants and dependents.72 Elections in the countryside were managed by landlords, and in the borough constituencies they were run by those figures from the underside of Victorian politics: wire-pullers, publicans, official treaters, paid messengers. The central affairs of the party were administered by the whips—the chief whip was simultaneously patronage secretary of the Treasury—and by a principal agent, who was the party leader’s personal attorney, and who was happy to involve himself in party business because it brought a large number of lucrative election cases to his law firm. The principal agent and the whips got candidates who were looking for constituencies and constituencies that were looking for candidates in contact with one another; they raised money from the party’s grands seigneurs and funneled it into constituencies; and they distributed patronage, both civil and ecclesiastical, to the party’s supporters.73 A movement to create a different kind of party organization was initiated in the late 1860s and early 1870s and is closely associated with the career of John E. Gorst.74 Gorst chaired the meeting in 1868 that established the National Union of Conservative and Constitutional Associations, an organization whose founders recognized that the future of Conservatism lay in mass party organization. In 1870 Gorst was appointed principal agent of the Conservative party by Disraeli, and he immediately set about to accomplish this goal by contacting prominent Conservatives in each constituency and encouraging them to organize a Conservative committee and to select local candidates. This nascent mass party organization ran the Conservative campaign in the election of 1874 and in large measure was responsible for the Conservative victory that year. After the 1874 victory, however, serious tensions arose between the leaders and cadres associated with this new party apparatus, on one hand, and the party’s Old Guard, those elements of the party whom Gorst termed “the old identity,” on the other hand.75 Gorst complained that once the party got into office, the leaders of the “old identity,” the whips, undermined the new structure by intervening in local constituencies apart from the network of Conservative committees and associations whose creation Gorst had supported: the whips backed candidates without consulting the local Conservative leadership; they distributed patronage to their local friends and allies; they financed corrupt electoral practices. This demoralized the cadres who had been attracted into party work in the early 1870s, and the structure Gorst had built up began to crumble. Gorst complained bitterly to Disraeli that the practice of “managing elections at the Treasury” would be fatal to the party, but to no avail.76 He resigned as principal agent in 1877. Unable to reform traditional party practices from within, Gorst joined an attack upon


123

them from without.77 In the late 1870s the aggrieved party cadres began voicing their complaints at the annual meetings of the National Union. The conflicts between the proponents of the new and the old styles of party politics reached its peak after the Conservatives were defeated in the election of 1880. Lord Randolph Churchill and his allies in the so-called Fourth party—one of whom was Gorst—took up the complaints of the dissidents, gained control of the National Union, and attempted to use it as a springboard for vaulting Churchill into the leadership of the party. The Fourth party episode ended in a compromise, “a sudden reconciliation which,” according to E.J.Feuchtwanger, “never has been quite satisfactorily explained.”78 In a broader perspective, however, this reconciliation is not difficult to understand: it was but another episode in the establishment of a modus vivendi between Britain’s landed and urban upper classes—an earlier episode of which was the civil-service reform of 1870—another step toward the formation of a coalition that it was very much in the interests of each side to enter. The top leaders of the Conservative party, Lord Salisbury and Sir Stafford Northcote, were compelled by the Conservative defeat in the election of 1880 to recognize that they could not afford to alienate the opponents of the patronage system within the party. These opponents—chiefly, but not entirely, drawn from the provincial bourgeoisie—for their own reasons were extremely anxious to come to terms with the groups that comprised the traditional base of the Conservative party. As James Cornford notes, the major reason why the cadres of the provincial Conservative and Constitutional associations were willing to undertake the burdens of party work was that it gave them the opportunity to interact with the Tory aristocracy and gentry as collaborators in a common endeavor.79 That incentive would only continue to be available if the two sides could come to terms with one another. In the course of the 1880s an accommodation was hammered out with which both sides could live. The traditional governing classes maintained their privileged access to the cabinet.80 Moreover, as both R.T.McKenzie and Samuel Beer stress, control over public policy continued to be vested in the government and the parliamentary party: in the realm of public policy-making, conservative governments do not profess to listen to the constituency party.81 On the other side, a uniform network of Conservative associations was extended throughout the country, the control of these associations over the conduct of elections was acknowledged, and the executive committee of the National Union was reorganized so that it could serve as the spokesmen for the cadres who manned these associations in the inner councils of the party.82 The Corrupt Practices Act 1883 outlawed many of the traditional techniques the whips had employed to influence local elections, and the Reform Act of 1884 reapportioned the shires and boroughs in which yet older forms of electoral control were practiced. Most relevant of all to us here, in the 1880s the Conservatives came increasingly to respect the autonomy and integrity of the civil service. Harold Hanham reports that in the 1870s virtually all high level appointments that Disraeli made in the Treasury were handed out as partisan rewards; by contrast appointments to these positions were granted in two-thirds of all cases as promotions to civil servants from within the Treasury by both Conservative and Liberal governments in the 1880s.83 As the political costs of violating the autonomy of the bureaucracy


124

increased, both of Britain’s major parties turned increasingly instead to the distribution of honorific positions, especially positions as magistrates, to their major local supporters.84 In the 1880s the Conservatives began to grant titles of nobility to merchants and manufacturers who made substantial contributions to the central party fund or who played major roles in the Conservative party organization, a practice that quite literally led to the fusion of Britain’s aristocracy and bourgeoisie.85 By these means the Conservatives could reward party luminaries without meddling in the allocation of civil-service positions and thereby alienating the constituency for bureaucratic autonomy upon whose support the party was becoming increasingly dependent. This set of accommodations proved to be remarkably stable. The pattern of party organization established by the Conservatives in 1885 has endured, as McKenzie notes, down to the present day.86 After a discrete and conflict-ridden transitional period, then, the Conservative party in Great Britain abandoned the practice of patronage politics. The constituency for bureaucratic autonomy within the party was able to prevail, and older practices were abandoned leaving surprisingly few residues, because it was supported by a large popular following. One strongly suspects that had universal suffrage been adopted in England prior to the formation of a constituency for bureaucratic autonomy the outcome of the struggle between the practitioners and opponents of patronage politics would have been quite different. In that event Britain would have recapitulated the experience of the United States during the Jacksonian Era. The wire-pullers within the Conservative party would then have been free to use patronage to acquire a genuinely mass following and their bargaining power vis-à-vis Salisbury and Northcote would have been as great as the leverage of the Fourth party and the National Union proved to be. The imaginative exercise, however, requires one to introduce so many counterfactual assumptions that it is not fruitful to pursue it further. In order to understand the conditions that enable patronage to survive into the contemporary period, it is more useful to analyze a case where the creation of a mass electorate did antedate the formation of a constituency for bureaucratic autonomy. Italy is such a case.

Italy The “Blocco Storico” Neither a progressive nor an absolutist coalition assumed control of the newly unified nation of Italy during the predemocratic era. It is not difficult to account for the absence of the former. Italy was simply too backward economically to follow the British path during the nineteenth century. It requires a somewhat lengthier explanation to account for the failure of Italy to follow the German precedent. On the surface, at least, there were some striking similarities between the Italy and Germany of, say, 1850. Both nations were politically fragmented; in both nations there was one state (Prussia in the case of Germany, Piedmont in the case of Italy) in which centralized bureaucratic institutions had taken root in the early modern period; and in both cases the process of unification involved the extension of that state’s institutions over the rest of the nation. There were, however, some crucial differences between the two cases. In Germany prior to unification


125

the region that was economically and socially most backward (East Elbia) was governed by the state whose institutions were extended over the newly unified nation. This meant, as I have argued above, that the most important segment of Germany’s landed classes had been integrated into, and had acquired a stake in the integrity of, the nation’s administrative apparatus. In Italy the region that in the mid-nineteenth century was economically and socially most backward (the Mezzogiorno) was governed not by the state that led the drive for unification, but rather by a regime (the Kingdom of the Two Sicilies) that was destroyed in the process. For this reason the landed classes of the south (which in the 1860s included members of the middle class who had seized land from the aristocracy, the Church, and the peasantry)87 had no stake in the integrity of the bureaucracy that now governed them. There was a second and related difference between the German and Italian cases. The forces of Prussian absolutism had to make very few concessions in order to absorb the other German states into the empire and to control the newly unified state thereafter. Under the constitution of the empire, Prussia dominated the upper house of the imperial legislature, the king of Prussia became the emperor of Germany, and the chancellor, the cabinet, and the bureaucracy were responsible to the emperor rather than the Reichstag. In this way the forces of Prussian absolutism were able to maintain their control over the entire executive apparatus. In order to build a majority in the Reichstag, one of the few major concessions Bismarck and his successors as chancellor had to make was to approve a tariff on industrial goods. This concession won them the support of the Rhenish industrialists and cemented the famous “coalition of iron and rye.”88 The forces of Piedmontese absolutism, by contrast, lost control of the government and bureaucracy of the new state fifteen years after the proclamation of the Kingdom of Italy. Nonetheless the interests of northern elites were more than adequately served by the new regime, and a strong basis existed for a coalition between them and the gentry of the south—the coalition Antonio Gramsci termed the blocco storico.89 What these elites wanted above all was to defend themselves against their Republican opponents and to control the new nation’s economic policies. The southern gentry sought protection for their landholdings against the threat of peasant insurrection, and in light of the depressed state of the southern economy, they also welcomed any employment opportunities the new regime could provide. The “programs” of the northern and southern elites were fully compatible, and served as the basis for the alliance between them. The central government policed the south, putting down “brigandage” and “banditry,” and it provided the gentry, who dominated southern representation in the legislature, with patronage, which they could use to maintain their control over their inferiors. In return for this assistance the southern gentry rallied to the government, and its deputies supported ministries whose commercial, financial, and industrial policies were oriented chiefly to the interests of northern capital. This logrolling arrangement underlay and was implemented through the practice of trasformismo, the process by which deputies who were elected as opponents of the government were “transformed” into its allies once they were granted patronage. The institution of trasformismo was perfected by the Liberal government of Depretis, which came to power in 1876 and was kept in office with the votes of southern deputies. Upon assuming office, Depretis removed those prefects whose loyalty he could not count upon and replaced them with individuals whom he


126

could trust to use their power to sustain the government. The techniques perfected by Depretis were used by the governments that ruled Italy for the next forty years.90 Party and Patronage: The Christian Democrats A constituency for bureaucratic autonomy did not emerge in Italy, then, prior to the democratic era; to the contrary, the nation was governed by a coalition—the blocco storico—that depended upon patronage in order to hold itself together. This coalition has remained the dominant political configuration in Italy to the present day, and terms of trade outlined above have remained the fundamental basis of accommodation between the two major partners in this coalition.91 This is not to say, however, that the present-day structure of Italian politics is identical to what it was in the era of Depretis, Crispi, and Giolitti. With the introduction of universal suffrage in 1913 and with the removal by the Church of its ban upon the participation of Catholics in Italian politics, two new actors have moved onto the political stage: the working classes, organized chiefly through the Socialist and then the Communist parties, and the Catholics, organized chiefly through the Popular party and then the Christian Democrats. Italy’s older political classes attempted initially to exclude these new actors from effective power; the ultimate consequence of this endeavor was Fascism.92 After the collapse of Fascism, however, Italy’s landed and industrial elites found it necessary to throw their support behind the Church and to work through the Christian Democratic party in order to continue to exclude from power the groups that constituted the greater threat to their hegemony. The fears of northern industrialists and southern landowners in the troubled years of the 1940s, then, were the same as the fears of their predecessors in the troubled years of the 1860s—revolution in the north and peasant insurrection in the south.93 The solutions adopted in the 1860s and 1940s were fundamentally similar—an alliance among the upper and middle classes of the two regions. Although the DC in this sense served the same function for these classes as did the parties of the Liberal era, it was more than simply one of these old parties in new dress, more than simply the Sinistra in a cassock and clerical collar. La Democrazia Christiana, after all, was the successor to Don Luigi Sturzo’s progressive Partito Popolare; it was allied with the Catholic trade-union movement in the north and it supported land reform in the south; and if it were to govern a nation whose electorate now exceeded twenty-five million, and do so without the votes of deputies from the working-class parties, it would need to establish a broad base of popular support.94 To acquire such support the DC mobilizes its constituency not through the old clientele structure but rather through an extensive network of party sections and affiliated organizations—a youth movement, a women’s movement, an organization of farmers, and a trade union federation.95 Although the mechanism through which the DC appeals to its supporters is that of a modern mass party, the substance of its appeal, at least in the south, is the traditional one of Italian politics. The DC has taken advantage of its control over the bureaucracy to place hordes of well-paid ushers and janitors on the public payroll.96 It provides subsidies to marginal businesses, to firms that could not survive in the absence of state support and whose owners and employees therefore become dependent upon the party for their continued prosperity and employment.97 Christian Democratic deputies and party functionaries, rather than lawyers and


127

landowners, now write letters of recommendation on behalf of candidates for admission to state schools. Sidney Tarrow summarizes this development by speaking of a movement from “clientelismo of the notable to clientelismo of the bureaucracy” and “from vertical to horizontal clienteles.”98 But how was this development possible? Why did the northern industrialists, who through the Confindustria financed the DC in the postwar period, tolerate the patronage, the corruption, the governmental inefficiency that was implicit in this method of operation? A possible answer to these questions is suggested by P.A. Allum.99 Allum notes that the Italian economic miracle of the postwar era was grounded upon two things: the ability of Italian industry to pay low wages in the period before 1963 and the monetary and credit policies embodied in the “Einaudi line.” Neither of these policies demanded much in the way of efficiency or effectiveness from the bureaucracy or were adversely affected by the patronage practices of the DC. The exclusion of the workingclass parties from the government was a requisite for the first of these, and this was accomplished in 1948, partly in response to American pressure. The monetary and credit policies of Einaudi were administered through the Bank of Italy, an august institution outside the pale of the DC’s patronage. During the fifteen-year period following the Second World War, then, the industrialists who were allied with the DC and who might have been expected to object strongly to the party’s patronage practices and to have fought for administrative reforms and greater governmental efficiency—who might have been expected, in other words, to lead a progressive coalition—did not have a strong incentive to do so. For this reason the DC was able to develop along the lines indicated above without fear of alienating this important set of allies. In the early 1960s, however, this state of affairs began to change. The extraordinary rate of economic growth that Italy had enjoyed as a result of the government’s monetary policies and industry’s ability to pay low wages slowed considerably by 1963.100 Labor became more restive and less compliant than it had been previously, and the leaders of Italy’s largest firms, who could afford to do so, were prepared to make economic concessions and to press for governmental reforms. Increasingly, the incompetence of Italy’s patronage-ridden bureaucracy, and the cost of sustaining that huge bureaucracy, has come to be regarded as a major source of the “Italian crisis” and its reform has come to be regarded as a prerequisite for pulling the nation out of that crisis.101 By the time these tendencies began to manifest themselves, however, the constituency for patronage within the DC had become sufficiently powerful to block any would-be reformers. The very freedom the party enjoyed in the postwar period to use patronage to mobilize popular support enabled the practitioners of patronage politics within the party to build a mass base for themselves. For this reason, Raphael Zariski notes, the DC has become the more factionalized the longer it has remained in office.102 And the patronageoriented factions within the party are not prepared to see reforms implemented that would deprive them of the fruits of power. As Sidney Tarrow observes, “because state agencies have been used by party and factional groupings to service their electorates, no government has been able to undertake a reform of the state without threatening their political livelihood.”103 Were a DC government to propose reforms that alienated this constituency for patronage it would risk splitting the party and losing much of its mass


128

base. The success of the DC’s constituency for patronage contrasts markedly with the failure, described above, of its counterpart within the British Conservative party. In order to find parallels to the DC one must, rather, turn to other parties that mobilized and organized an extensive popular following before the opponents of the patronage system coalesced. The American party system provides many parallel cases, and the factional struggles within the Christian Democratic party resemble in substance and outcome nothing so much as the conflicts between regular and reform factions within the Democratic parties of Illinois or New York.

Conclusion By way of conclusion I will very briefly discuss two cases, France and the United States, which would appear to stand as counter-examples to the argument presented in this article, and I will suggest how their deviations from the patterns outlined above can be explained. Over the course of the past two centuries France appears to have deviated in both possible directions from the patterns outlined above. An absolutist coalition emerged in the seventeenth and eighteenth centuries in France, well before the creation of a mass electorate, and yet patronage came to play a major role in the parliamentary and party politics of France during the second half of the nineteenth century and the first half of the twentieth.104 Another change of direction occurred after the Second World War. Patronage played a less prominent role in the political life of the Fourth Republic than it had in the Third, and it has moved yet further into the background in the Fifth Republic.105 As for the United States, the creation of a mass electorate in the Jacksonian period preceded by half a century the emergence of America’s constituency for bureaucratic autonomy during the Progressive era, and yet the Progressives triumphed and managed to put an end to the practice of patronage politics in many states and cities in the United States. How can these cases be reconciled with the argument spelled out above? In arguing that internally mobilized parties will avoid the use of patronage in nations where either an absolutist or a progressive coalition emerged prior to the mobilization of the masses into politics, and that such parties will be patronage oriented where the mobilization of the masses into politics antedated the formation of either one of these coalitions, I do not mean to suggest that the order in which these two events occurs utterly predetermines the subsequent character of a nation’s party politics. My claim, rather, is more modest: the relative timing of these two developments weights the dice toward one outcome or the other. The “expected” outcome can be reversed, but any effort to reverse it engenders a crisis of the regime and requires the total reorientation of the nation’s politics—perhaps even a revolution—to succeed. If and when such a crisis occurs, the challengers to the old order and the defenders of that order mobilize their forces and the outcome of the struggle depends upon the breadth of the popular support and the character of the institutional backing enjoyed by each side. Germany, as I argued at some length above, experienced a number of such crises—for example, in 1848 and


129

1918—but the forces of absolutism were able to prevail in each instance. In France and the United States the challengers to the pre-existing regime fared somewhat better. France It should be noted, however, that it required a series of revolutions and the toppling of a succession of regimes—three monarchies, two empires, and four republics—to bring about the changes in France mentioned above, and even then it is possible to see the residues of Bourbon absolutism in the France of today. The Revolution of 1789 neither destroyed the centralized state that Louis XIV created, as Tocqueville long ago argued, nor did it dislodge the forces of absolutism from positions in that structure, as the career of the most prominent representative of those forces, Tallyrand, illustrates; Napoleon, rather, built upon the foundations laid by the Bourbons.106 It was only after the Revolution of 1830 that the old families of the noblesse de robe were driven from the prefectoral corps, 107 but even the revolutions of 1830 and 1848, the rise and fall of the Second Empire, and the creation of the Third Republic together were not sufficient to destroy the absolutist coalition or to drive its members entirely from the citadels of power in France. As Theodore Zeldin notes in his study of politics in the Third Republic, Republican parliamentary politicians did indeed regard themselves as opponents of the state and of the classes who were identified with it, and these politicians used the institutions of the Republic, namely, the Chamber of Deputies, to gain control over and to colonize those governmental agencies (Interior, Ponts et Chaussées, Education, or more generally, the field administration of the state) that most closely concerned the classes they represented, the provincial middle classes.108 Republican ministerial politicians—that small group (e. g., Sarraut, Briand, Freycinet) that monopolized cabinet positions in the Republic—however, found the overhead agencies of the central administration (especially the Conseil d’Etat and the Inspection des Finances) so useful for assuring their control over the regime that they did not undermine them.109 There was a tacit arrangement between the government and the old forces of absolutism to exclude parliamentary politicians from influence over these agencies, as well as over the foreign, colonial, and military services, and recruitment into them was restricted to young men who had passed through the Ecole Libre des Sciences Politiques.110 This institution and these departments remained the preserve of the absolutist coalition and were hotbeds of anti-Republican sentiment, as the events of 1940 were to demonstrate. In other words, under the Third Republic the state in France was divided along functional lines, and each half was grounded upon a different social base. The patronage system of the Third Republic fed upon that half of the state from which the absolutist coalition had been driven by the revolutions of the nineteenth century. With the creation of the Fourth Republic the defenders of bureaucratic autonomy were able to erect a more extensive barrier between the bureaucracy and party politicians than had existed under the Third Republic by entering into an alliance with lower level officials—an arrangement negotiated by Michel Debre and Maurice Thorez!111As the base of support for a politically autonomous bureaucracy was extended in this way, party politicians in the Fourth Republic found it more difficult to acquire patronage than had their counterparts in the Third Republic. And the weakening of parliament in the Fifth


130

Republic has made it more difficult for them still. I recount this history in order to underline two points. First, I would call attention to the severity of the political disruptions that were required to bring about changes in the role that patronage played in French politics and the magnitude of the realignments in the entire structure of national politics that these changes entailed. Second, I would note that despite this succession of revolutions and changes of regime, the constituency for bureaucratic autonomy that coalesced in the predemocratic era in France was able at all times to retain at least a foothold in the state administration and to emerge triumphant again in the contemporary period after it broadened its base of support. The United States If in France under the Third Republic the state was divided along functional lines between a constituency for bureaucratic autonomy and a constituency for patronage, in the United States in the twentieth century it has been divided along geographic lines: the practitioners of patronage politics have been able to hold their own in most states to the east of the Mississippi and their opponents have had the upper hand in most states to the west. The situation in the eastern states accords with the predictions of the theory outlined above: the enactment of white manhood suffrage in the United States antedated the formation of a constituency for bureaucratic autonomy (the former occurred in the Age of Jackson, the latter in the Progressive era); the major parties in the United States thus were free to use patronage when they undertook to mobilize mass support; consequently, the forces that had a stake in the patronage system enjoyed broad popular backing; and thus these forces were able to offer substantial resistance to the constituency for bureaucratic autonomy that coalesced under the banner of progressivism in the early decades of the twentieth century. By contrast, the opponents of the patronage system enjoyed considerable success in the western states, and this appears to contradict the theory outlined above. This variation in the character of present-day party politics among the major regions of the United States can, however, be explained in terms consistent with the analysis offered in this essay. The sources of this difference does not lie in the social composition of the electorate, as Raymond Wolfinger has argued.112 Nor does it lie in differences in the social composition of the forces that led the attack upon the patronage system. The leadership of America’s constituency for bureaucratic autonomy is drawn from roughly similar groups in all regions of the nation—from the professional upper-middle classes and from those elements of the local notability and business community that do not enjoy privileged access to the locally dominant party—and it resembles in broad outline the social composition of the progressive coalition in Great Britain—if one considers the descendants of the New England patriciate to be the American equivalent of the English aristocracy.113 The source of this variation lies, rather, along the other dimension I have emphasized: the extent to which party politicians had mobilized a broad popular base prior to the emergence of a constituency for bureaucratic autonomy.114 Party organizations in all regions of the United States in the nineteenth century were patronage oriented, but in the states of America’s “periphery” these organizations were less tightly organized and less broadly based than the political machines that governed the older


131

states of the northeastern “metropole”: the politicians who were the agents of “foreign corporations” (most commonly railroads) in the “colonial” regions of the United States, by virtue of the vast magnitude of the resources at their disposal relative to those possessed by other political actors, were able to dominate state politics without mobilizing widespread popular support. William Herren’s Southern Pacific machine in California, for example, was not a mass party in the same sense as was the Democratic machine in New York.115 The Progressives in the western states, I would hypothesize, were able to ride to power by winning support among the voters who had not been mobilized by the pre-existing party system. What I am suggesting, in other words, is that although manhood suffrage existed throughout the United States at the time the forces of progressivism coalesced, the electorate in the peripheral regions of the United States had not been fully mobilized by a well organized mass party, and consequently the constituency for patronage in these states was not so broadly based as it was in the states of the Northeast. This can explain why the Progressives in the West were able quite convincingly to picture themselves as spokesmen for “the people” against “the interests,” whereas in the East the electorate was more inclined to regard reform as the cause of a narrow elite and to view machine politicians as representatives of the popular cause.116 An analysis of these conflicts, however, and a more detailed comparison of the American and European cases is beyond the scope of this article.

Notes This is a revised and condensed version of a paper that was delivered at the Conference on the Scope and Practice of Social Science History, April 23, 1976 at the University of Wisconsin, and that appeared in the occasional paper series of the Western Societies Program, Center for International Studies, Cornell University For their helpful comments I would like to thank Douglas Ashford, Peter Gourevitch, Davydd Greenwood, Peter Katzenstein, James Kurth, Peter Lange, Charles Maier, and Sidney Tarrow. 1. 2.

3.

4.

James C.Scott, Comparative Political Corruption (Englewood Cliffs, N.J.: PrenticeHall, 1972), p. 110. Ibid, chap. 6; Carl Landé, “Networks and Groups in Southeast Asia: Some Observations on the Group Theory of Politics,” American Political Science Review 67 (March 1973) 119; Edward Banfield and James Q.Wilson, City Politics (Cambridge: Harvard University Press, 1963), pp. 40–411,122–23; and idem, “Political Ethos Revisited,” American Political Science Review 65 (December 1971): 1048–62 ; Elmer Cornwell, “Bosses, Machines and Ethnic Groups,” The Annals of the American Academy of Political and Social Sciences 353 (May 1964) 27–39; Samuel P.Hays, “Political Parties and the Community-Society Continuum,” in The American Party Systems, ed. William Nisbet Chambers and Walter Dean Burnham (New York Oxford University Press, 1967), pp. 173–81. Scott, Comparative Political Corruption, pp. 99, 118; Banfield and Wilson, City Politics, chap. 9. See also James Q.Wilson, Political Organizations (New York: Basic Books, 1973), chap. 6; Leon Epstein, Political Parties in Western Democracies (New York: Praeger, 1967), pp. 27–31. R.H.Evans, Coexistence: Communism and Its Practice in Bologna 1945–1965 (South Bend, Indiana: University of Notre Dame Press, 1967).

Party and Patronage: Germany, England, and Italy 5.

6. 7. 8. 9.

10.

11.

12. 13. 14.

132

Michael Paul Rogin, The Intellectuals and McCarthy: The Radical Spectre (Cambridge: MIT Press, 1967), p. 187; Sidney Tarrow, “Economic Development and the Transformation of the Italian Party System,” Comparative Politics 1 (1969): 181. Gerald Brenan, The Spanish Labyrinth (Cambridge: Cambridge University Press, 1943), pt. 2. John H.Fenton, Midwest Politics (New York: Holt, Rinehart and Winston, 1966). Peter J.Larmour, The French Radical Party in the 1930s (Stanford: Stanford University Press, 1964). On patronage and party management in the United States in the mid-nineteenth century, see Harry J.Carmen and Reinhard H.Luthin, Lincoln and the Patronage (New York: Columbia University Press, 1943). On regional variations in the significance of patronage in presentday American politics, see Raymond Wolfinger, “Why Political Machines Have Not Withered Away and Other Revisionist Thoughts,” Journal of Politics 34 (May 1972): 397– 98. Luigi Graziano, “Center-Periphery Relations and the Italian Crisis: The Problem of Clientelism,” in Territorial Politics in Industrial Nations, ed. Sidney Tarrow et al. (New York: Praeger, 1978), p. 312; and Alessandro Pizzorno, “I ceti medi nei meccanismi del consenso,” in Il Caso Italiano, ed. F.L.Cavazza and S.R.Graubard (Milan: Garzanti, 1974), cited in ibid., n. 59. James Duesenberry, Income, Saving and the Theory of Consumer Behavior (Cambridge Harvard University Press, 1949); Robin Marris, The Economic Theory of Managerial Capitalism (New York: The Free Press, 1964), chap. 4. Richard M.Cyert and James G.March, A Behavioral Theory of the Firm (Englewood Cliffs, NJ.Prentice Hall, 1963), chaps. 3–5. Emma Rothschild, Paradise Lost: The Decline of the Auto-Industrial Age (New York: Random House, 1973) chaps. 2–3. If a party is to pursue a patronage strategy it must be able to offer to its supporters indulgences that exceed in value the benefits they can obtain from alternative sources. That is, whether a group will respond to a patronage appeal depends in the first instance not upon the characteristics of its members per se—every man, as Hobbes observed, has his price—but rather upon the relationship between the value of the benefits produced by the government, on one hand, and the opportunities available to individuals whose support the party is seeking, on the other. This condition for the existence of a patronage party is not very restrictive. Regimes as weak as those of Jacksonian America and Hanoverian England were capable of generating indulgences that could sustain patronage parties, because even these regimes, through the granting of offices and charters, were in a position to affect the important interests of a considerable portion of the policial class. It is only when the jurisdiction of a regime is not commensurate in scope with the major interests of its inhabitants that the party controlling the regime would not be in a position to pursue a patronage strategy. The major example of such a setting is the upper income, bedroom suburb. The regulations from which a suburban politician can exempt a business firm, and the purchases that a suburban government makes, are of insignificant value to businessmen whose firms operate in metropolitan, national, or even international markets. And the jobs such politicians control pay salaries far lower than the ones that the upper-income residents of these communities can obtain from the private market. A classic fallacy of ecological reasoning thus is involved in the effort to draw inferences concerning the behavior of upper- and middle-class individuals from observations of the character of polities in upperand middle-income communities such as Winnetka, Illinois or Scarsdale, New York. See


133

Banfield and Wilson, City Politics, p. 140. 15. Epstein, Political Parties, pp. 110–11. 16. Landé, “Networks and Groups,” p. 118. 17. Philip Selznick, Leadership in Administration (Evanston, III.: Row, Peterson and Co., 1957), chap. 2; Ira Katznelson, Black Men, White Cities (London: Oxford University Press, 1978), p. 24. 18. Samuel P.Huntington, Political Order in Changing Societies (New Haven: Yale University Press, 1968), p. 415. 19. Maurice Duvarger, Political Parties (London: Methuen, 1959), pp. xxiv-xxxvii; Huntington, Political Order, pp. 412–20. 20. Cf.Wolfram Fischer and Peter Lundgreen, “The Recruitment and Training of Administrative and Technical Personnel,” pp. 456–561 in The Formation of National States in Western Europe, ed. Charles Tilly (Princeton: Princeton University Press, 1975); and John A.Armstrong, The European Administrative Elite (Princeton: Princeton University Press, 1973). 21. H.R.Trevor-Roper, “The General Crisis of the Seventeenth Century “in Crisis in Europe, 1560–1660, ed. Trevor Aston (Garden City, N.Y.: Anchor Books 1967), p. 87; Isaac Kramnick, The Rage of Edmund Burke (New York: Basic Books, 1977), chap 2; Reinhard Bendix, Nation Building and Citizenship (Garden City, N.Y.Anchor Books, 1969), pp. 86– 89; Brenan, Spanish Labyrinth, pp. 195, 201. 22. Seymour Martin Lipset and Stein Rokkan, Party Systems and Voter Alignments (New York: The Free Press, 1967), pp. 50–52 23. As Carl Landé notes, the leaders of weak regimes are unable to prevent subordinate officials from using their authority to generate patronage and to build personal followings for themselves, because such leaders are dependent upon the support of these officials and their followings to keep the regime in power. This phenomenon can explain why the Christian Democratic party in Italy has become increasingly factionalized and patronage ridden the longer it is in office. See Landé, “Networks and Groups,” pp. 126–27; Raphael Zariski “Intra-Party Conflict in a Dominant Party: The Experience of Italian Christian Democracy,” Journal of Politics 27 (February 1965): 3–34. 24. See, e.g., the account of the development of the CPP machine in Scott, Comparative Political Corruption, pp. 124–31. 25. Clement Henry Moore, Tunisia Since Independence (Berkeley and Los Angeles: University of California Press, 1965). 26. Immanuel Wallerstein, The Modern World-System (New York: Academic Press, 1974), chap. 5. 27. Perry Andersen, Lineages of the Absolutist State (London: NLB, 1974), pt. 2, chap. 3. 28. F.L Carsten, The Origins of Prussia (Oxford: Clarendon Press, 1954), pt. 3. 29. Cf.Fischer and Lundgreen, “Recruitment and Training,” pp. 457 ff. 30. Hans-Eberhard Mueller, “Bureaucracy and Education: Civil Service Reform in Prussia and England and Strategies of Monopolization” (Ph. D diss, University of California at Berkeley, 1974), p. 158. 31. Hans Rosenburg, Bureaucracy, Aristocracy and Autocracy (Cambridge Harvard University Press, 1958), pp. 43 ff. 32. Ibid., pp. 66, 68 ff.; Mueller, “Bureaucracy and Education,” pp. 103, 290. 33. John R.Gillis, The Prussian Bureaucracy in Crisis (Stanford: Stanford University Press, 1971), pp. 29 ff. 34. Rosenburg, Bureaucracy, Aristocracy and Autocracy, chap 9. The social function of the


35.

36. 37. 38. 39. 40.

41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60.

134

gymnasiums and universities in Germany and of the ideology of Bildung is discussed in Armstrong, European Administrative Elite, pp. 135 ff., 165 ff. Theodore Hamerow, The Social Foundations of German Unification, 1858–1871: Struggles and Accomplishments (Princeton: Princeton University Press, 1972), chaps. 4,7, 9; Arthur Rosenberg, Imperial Germany: The Birth of the German Republic, 1871–1918 (Boston: Beacon Press, 1964), chap 1. Karl Hochswender, “The Politics of Civil Service Reform in West Germany,” (Ph.D.diss., Yale University, 1962), pp. 63–65. Otto Kircheimer, “Germany: The Vanishing Opposition,” in Political Oppositions in Western Democracies, ed. Robert Dahl, (New Haven: Yale University Press, 1966), p. 237. Lewis Hertzman, DNVP: Right Wing Opposition in the Weimar Republic, 1918–1924 (Lincoln: University of Nebraska Press, 1963), p. 17. Herman Finer, The Theory and Practice of Modern Government (London: Methuen &Co., 1932), 2:1389. The continuities between the civil service of Imperial and Weimar Germany are stressed by Fritz Morstein Marx, “Civil Service in Germany,” in Civil Service Abroad, ed. Leonard White (New York: McGraw-Hill, 1935), pp. 161–274. Ralf Dahrendorf, Society and Democracy in Germany (New York Doubleday, 1967), p. 252. Erich Eyck, A History of the Weimar Republic (Cambridge: Harvard University Press, 1962), p. 297. Hertzman, DNVP, chap. 5; Charles Frye, “Parties and Interest Groups in Weimar and Bonn,” World Politics 17 (July 1965): 641. Brian Chapman, The Profession of Government (London: Urwin University Books), pp. 284 f. Herman Finer, Governments of the Greater European Powers (New York: Holt, 1965), p. 628. Morstein Marx, “Civil Service in Germany,” p. 256. Fritz Morstein Marx, Government of the Third Reich (New York: McGraw-Hill, 1937); Hochswender, “Politics of Civil Service Reform,” p. 61. Finer, Theory, p. 480; James Pollock, “The German Party System,” American Political Science Review 23 (November 1929): 885. David Schoenbaum, Hitler’s Social Revolution (Garden City, N.Y.: Doubleday, 1966), chap. 7. Ibid., p. 221. Hochswender, “Politics of Civil Service Reform,” p. 39. Lewis Edinger, “Post-Totalitarian Leadership,” American Political Science Review 54 (March 1960): 66. Dahrendorf, Society and Democracy, chap 14. John Herz, The Government of Germany (New York: Harcourt, Brace, Jovanovich, 1972), p. 130. Taylor Cole, “The Democratization of the German Civil Service,” Journal of Politics 14 (February 1952): 17. Edinger, “Post-Totalitarian Leadership,” p. 80. Herz, Government of Germany, p. 128. Dahrendorf, Society and Democracy, p. 15. Hochswender, “Politics of Civil Service Reform,” p. 193. Ibid., p. 154.


135

61. Ibid., pp. 157–59, citing Theodor Eschenburg, Aemterpatronage (Stuttgart, 1961). 62. Anderson, Lineages of the Absolutist State, pt. 1, chap 5. 63. Barrington Moore, The Social Origins of Dictatorship and Democracy (Boston: Beacon Press, 1966), chap. 1. 64. Louis Namier, The Structure of Politics at the Accession of George III (London: MacMillan, 1961). 65. This is not true of extraparliamentary groupings such as the Yorkshire Association in the late eighteenth century or the Chartists in the early nineteenth. These formations were the prototypes of what I have termed “externally mobilized parties.” On the former, see Ian R.Christie, “The Yorkshire Association, 1780–4: A Study in Political Organization,: The Historical Journal 3 (1960): 144–61. On the latter, see Asa Briggs, Chartist Studies (New York: St. Martin’s Press, 1959). 66. Mueller, “Bureaucracy and Education,” p. 281. 67. The phrase is Theodore Lowi’s, The End of Liberalism (New York W.W.Norton, 1969), p. 243. 68. Olive Anderson, “The Janus Face of Mid-Nineteenth Century English Radicalism The Administrative Reform Association of 1855,” Victorian Studies 8 (March 1965) 231–42. 69. Mueller, “Bureaucracy and Education,” pp 324–30, 344. 70. I owe this insight to a discussion with Martin Bernal. 71. I.e., the redistribution of seats that was part and parcel of parliamentary reform in 1867 and 1884 and the Corrupt and Illegal Practices Act of 1883. On the latter see Cornelius O’Leary, The Elimination of Corrupt Practices in British Elections, 1868–1911 (Oxford: Clarendon Press, 1962), chap. 6. 72. James Cornford, “The Transformation of Conservatism in the Late Nineteenth Century,” Victorian Studies 7 (September 1963): 41 f. 73. E.J.Feuchtwanger, Disraeli, Democracy and the Tory Party (Oxford: Clarendon Press, 1968), pp. 106–13. 74. E.J.Feuchtwanger, “J.E.Gorst and the Central Organization of the Conservative Party, 1870–1882,” Bulletin of the Institute of Historical Research 2 (1959): 192–208. 75. Cornford, “Transformation,” p. 49. 76. H.J.Hanham, Elections and Party Management (London: Longmans, Green, 1959), p. 362. 77. Cornford, transformation,” p. 44. 78. Feuchtwanger, Tory Party, p. 87. 79. James Cornford, “The Adoption of Mass Organization by the British Conservative Party,” in Cleavages, Ideologies and Party Systems, ed. Erik Allardt and Yrjo Littunen (Turku: Publications of the Westermarck Society), p. 411. 80. W.L.Guttsman, The British Political Elite (New York: Basic Books, 1963), pp. 78 ff. See also James Cornford, “The Parliamentary Foundations of the Hotel Cecil,” pp. 268–311 in Ideas and Institutions of Victorian Britain, ed. Robert Robson (London: G.Bell and Sons, 1967). 81. R.T.McKenzie, British Political Parties (New York: St.Martin’s Press, 1963), p. 180; Samuel Beer, British Politics in the Collectivist Age (New York: Alfred Knopf, 1967), p.98. 82. Cornford, “Transformations,” p. 50 ff. 83. H.J.Hanham, “Political Patronage at the Treasury, 1870–1912,” Historical Journal 3 (1960): 77 ff. See also Maurice Wright, Treasury Control of the Civil Service (Oxford: Clarendon Press, 1969), p. 75. 84. John Vincent, The Formation of the Liberal Party (London: Constable, 1966), pp. 126–38.


136

85. H.J.Hanham, “The Sale of Honours in Late Victorian England,” Victorian Studies 3 (March 1960): 227. 86. McKenzie, British Political Parties, p. 179. 87. Denis Mack Smith, Italy: A Modern History (Ann Arbor University of Michigan Press, 1959), chap. 6. 88. Alexander Gerschenkron, Bread and Democracy in Germany (New York: H. Fertig, 1966). 89. See P.A.Allum, Italy: Republic without Government? (New York: W.W.Norton, 1973), pp. 4, 22. 90. Mack Smith, Italy, chap. 15. 91. Robert Fried, The Italian Prefects (New Haven: Yale University Press, 1963), p. 297. 92. With the signing of the Lateran Pacts in 1929, of course, Mussolini reached an accommodation with the Church. This presaged and made possible the political configuration that was to emerge after World War II. See Richard A.Webster, The Cross and the Fasces, (Stanford: Stanford University Press, 1960), chap 7. 93. Norman Kogan, A Political History of Postwar Italy (New York: Frederick A., Praeger, 1966), p. 30. 94. On Christian Democracy in Italy, see Webster, Cross and the Fasces, and Mario Einaudi and François Goguel, Christian Democracy in Italy and France (Notre Dame: Notre Dame University Press, 1952), pt.1. 95. Giorgio Galli and Alfonso Prandi, Patterns of Political Participation in Italy (New Haven: Yale University Press, 1970), chap. 5. 96. Graziano, “Italian Crisis,” table 7. 97. Pizzorno, “I ceti medi nel meccanismi del consenso.” 98. Sidney Tarrow, Peasant Communism in Southern Italy (New Haven Yale University Press, 1967), p. 341. 99. Allum, Italy, pp. 25–29. 100. Ibid., pp 29–32. 101. Luigi Graziano, “Clientelism and the Political System: The Sources of the Italian Crisis” (mimeo, 1975), pp. 5–7 and sources cited therein. 102. Raphael Zariski, “Intra-Party Conflict in a Dominant Party: The Experience of Italian Christian Democracy,” Journal of Politics 27 (February 1965): 3–34. 103. Sidney Tarrow, “The Italian Party System between Crisis and Transition,” American Journal of Political Science 21 (May 1977): 217. 104. Ezra Suleiman, Politics, Power and Bureaucracy in France (Princeton: Princeton University Press, 1974), pp. 808 ff. 105. See Sidney Tarrow, Integration at the Periphery: Grassroots Politics in Italy and France (New Haven: Yale University Press, 1977), for a striking picture of differences between present-day France and Italy on this score. 106. Alexis de Tocqueville, The Old Regime and the French Revolution (New York: Anchor Books, 1955), p. 2, chap. 2. 107. Nicholas J.Richardson, The French Prefectoral Corp, 1814–1830 (Cambridge: Cambridge University Press, 1966), pp. 204–6. 108. Theodore Zeldin, France 1848–1945 (Oxford: Clarendon Press, 1973), 1:576–79; Larmour, French Radical Party, p. 24. 109. Zeldin, France 1848–1945, pp. 603 ff., chap. 20. 110. David Thompson, Democracy in France Since 1870, 4th ed. (New York: Oxford University Press, 1964), pp. 59–63. 111. Suleiman, Bureaucracy in France, p. 46 f.


137

112. Wolfinger, “Political Machines.” 113. Ari Hoogenboom, Outlawing the Spoils: A History of the Civil Service Reform Movement (Urbana: University of Illinois Press, 1961); George Mowry, The California Progressives (Berkeley: University of California Press, 1951), chap. 4; Gerald W.McFarland, Mugwumps, Morals and Politics, 1884–1920 (Amherst: University of Massachusetts Press, 1975). 114. To be sure, the political cliques that dominated southern politics prior to the Progressive era were even more narrowly based than their counterparts in the western states, and yet patronage still plays a significant role in state politics in the South. The survival of patronage in the South is to be explained more by the weakness of progressivism in that overwhelmingly rural region than by the breadth of support enjoyed by their opponents. In the urban enclaves of the South progressivism was quite successful. See C. Vann Woodward, Origins of the New South, 1877–1913 (Baton Rouge: Louisiana State University Press, 1951), pp. 388 ff. 115. Cf.Mowry, California Progressives, chap. 1; and Martin Shefter, “The Electoral Foundations of the Political Machine: New York City, 1884–1897,” pp. 263–98, in American Electoral History: Quantitative Studies of Popular Voting Behavior, ed. Joel Silbey, Alan Bogue and William Flanigan (Princeton: Princeton University Press, 1978). 116. Glazer and Moynihan phrase it this way: The Catholic ascendary in New York had been based first on numbers, but second on a reasonably well grounded assumption that they normally, as Democrats, would best look after the interests of ordinary people, and would be especially concerned with the least well off, being themselves only recently emerged from that condition. The Protestant elite of the city had always challenged that assumption, asserting instead that the Tammany bosses were boodlers, pure and simple, or in a slightly different formulation such as that of Lincoln Steffens, were merely paid lackeys of the really Big Boodlers. Either way the charge was that they did not truly represent the people as they claimed to do. In three elections out of four the masses would choose to believe Charlie Murphy’s version rather than that of the New York Times. Nathan Glazer and Daniel P.Moynihan, Beyond the Melting Pot, 2nd ed. (Cambridge: MIT Press, 1971), p. [xiii]



138

SECTION TWO: States and Ethnicity


Commentary I.Katznelson’s ‘Working Class Formation and the State: Nineteenth Century England in American Perspective’ can best be seen as filling in some of the details missed out in the broader discussion offered by Mann in his article in the previous section. Even though the British state was essentially liberal, the fact that citizenship rights had to be won meant that workers were forced to organise as a class—albeit, given the ease of the struggle, as a class with labour solidarity rather than political consciousness. In contrast, the early, almost automatic achievement of political citizenship in the United States meant that there was no experience of being a class, ruled out of the public sphere. In these circumstances, ethnic loyalties took the place of class alignments. R.Bates’s ‘Ethnic Competition and Modernisation in Contemporary Africa’ turns to a very different set of social circumstances. The problem addressed by Bates is that of why ethnic competition increases during the process of modernisation. A very large part of the answer is to be found, he argues, in the classifying strategies of the modernising state. More particularly, the recognition and favouring of one particular ethnic group makes it entirely rational for other groups to, so to speak, brush up on and press claims on the basis of their own ethnic loyalties. All-in-all, the European pattern of development—in which extended kinship links were destroyed early on, thereby making people better fodder for state formation—looks as if it will not be repeated in Africa. It is all too easy to see that this is likely to impede further political, economic and social development. D.L.Horowitz’s magisterial Patterns of Ethnic Separatism’ attempts to produce a general typology so as to understand the incidence and character of ethnic separatism within the whole of the historic record. He concentrates on two variables—whether the group occupies an advanced or backward part of the region and whether the group itself is or is not socially advanced, especially in terms of educational achievement—in order to plot the likelihood and timing of four different types of separatist movement. This is both useful and convincing. But it is well worth noting, given the concerns of this collection, that the key triggering mechanism creating drives for secession in each case was that of actions on the part of the state. There was no Tamil separatist movement of any significance until the state sought to exclude Tamils from public life! More generally, separatist movements have limited chances so long as the state remains true to basic tenets of liberalism. Here again political action is occasioned by nothing less than the need to respond to state interference. Horowitz also interestingly notes, however, that the chances of separatist movements do not depend simply upon what happens inside a particular territory: quite as important is external recognition. Given current pressure in international society to diminish the norm of non-intervention, so slavishly obeyed for half a century after 1945, it is very likely that the world will see many more cases of successful ethnic separatism.


142

46 Working-CIass Formation and the State: Nineteenth-Century England in American Perspective* Ira Katznelson

* Source: P.Evans, D.Rueschemeyer, T.Skocpol, eds., Bringing the State Back In, 1985, ch. 8, pp.257–284.

When the House of Representatives investigated election fraud in New York City in 1868, it uncovered a massive scheme organized by Tammany Hall to sell counterfeit naturalization papers in order to register immigrants, who were not yet citizens, to vote. The committee report revealed that neighborhood saloons provided the locations for most of this activity. The testimony of Theodore Allen, the owner of “St.Bernard’s” on Thompson Street in Greenwich Village, was quite typical: I keep a public house, and a man by the name of James Goff and his brother, who were engaged in procuring naturalization certificates, used to come to my house a great deal…. I suppose 1000 were sent to Brooklyn that I saw them have. They contracted for these papers, they said, at 50 cents a head.1 Each Sunday evening, from the mid-1850s to the late 1880s, a debating society met at the Hope and Anchor Inn on Navigation Street in England’s Birmingham. Some twenty to forty speakers would argue their cases on a set topic. From the vantage point of the middle classes, Brian Harrison has noted, there was much to frighten the outside observer: the motion of 23 January 1859 for working class enfranchisement was unopposed, and the monarchy lost by thirty-nine votes to sixteen in a debate on republicanism on 22 March 1863. On two occasions, 26 August 1866 and 21 April 1867, the Society was visited by prominent London Reform Leaguers: these included George Howell who joined in the debates…. By twenty-three votes to fifteen, on 11 June 1871, the Society supported the Paris Communists. Nor was its radicalism exclusively political. On 21 August 1859, the strikers in the London building trade were supported by twenty-two votes to two; on 16 October 1864 the midlands miners, on strike at the time, were supported by forty-five votes to eighteen.2 The puzzle I wish to identify and explain can be discovered in the workingclass saloons and pubs of nineteenth-century city neighborhoods in England and the United States. In

Working-CIass Formation and the State: 143 both countries drinking places were important centers of transportation before the widespread introduction of steam-powered streetcars and trains; and after, they were places of refreshment for short- and long-distance commuters. Sometimes plain and sometimes extravagantly decorated, public houses were places of entertainment, especially for men. Some were music halls, some underworld hangouts, some houses of prostitution. All were places of dense sociability, embedded in the fabric of street and community. And, especially after midcentury, most were gathering places for crowds that were increasingly homogeneous in terms of social class. In both England and the United States public houses became the most common location of working-class political activity. In the United States, saloons provided hospitable gathering places for political parties. Saloon-keepers often acted as political entrepreneurs who built modestly durable and intensely local partisan machines capable of delivering votes in a predictable way in exchange for patronage. As an important study of party organization in the nineteenth and early twentieth centuries concluded, “The saloons were the nodal points of district organization of both parties. In the back rooms the bosses met to discuss their plans for carrying the districts, while out in front before the bar the captains were building up goodwill among the patrons.”3 These political parties, I have argued elsewhere,4 were organizations the political mobilization activities of which were disconnected on the whole from the trade union and work dimensions of working class life. The manifest content of saloon politics was not that of class but of ethnicity and territoriality. It was concerned with the links between the citizens of a specific residential area and the local level of government. Working-class pubs in England, by contrast, were part of a network of values and organizations that reflected and promoted the view that class pervades all social relationships, not just those at the work place. For the middle classes, pubs came into their own, much like American saloons, as places of mobilization of voters at election time, but for the great majority who could not vote, the meeting rooms of their pubs provided refuge for a host of oppositional political activities, including, most notably, trade union and Chartist groups. The contrasting political content of working-class activity in pubs and saloons in England and the United States is indicative of a general pattern that has been overlooked in most treatments of working-class formation in England. Implicit in much of this scholarship is a Continental contrast that makes problematic the cautious, reformist quality of workers’ collective expressions and actions, especially after the demise of the Chartist movement. By contrast, the class character of working-class dispositions, organization, and activity is taken for granted as the virtually inevitable product of capitalist-industrial development. The important, but in this respect typical, treatment of the emergence of class in the early nineteenth century by Asa Briggs thus begins by reminding us (no doubt correctly) that “the concept of social ‘class’ with all its attendant terminology was a product of large-scale economic and social changes of the late eighteenth and early nineteenth centuries” and stresses that “the change in nomenclature in the late eighteenth and early nineteenth centuries reflected a basic change not only in men’s ways of viewing society but in society itself.”5 A comparison with the United States demonstrates the dubious quality of even such an innocuous formulation and its confusion of necessary and sufficient conditions. As in


144

England, American workers eschewed class struggle outside of a gradualist, constitutional framework. In the United States, however, a split consciousness came to divide the working class: as laborers at the work place and as ethnics or residents of this or that territory in their residential communities. In England, by contrast, there was no equivalent divided consciousness, nor was class as a category of social understanding limited to the realm of work and labor. “Class” joined rather than divided the realms of work and off work. Especially characteristic of working-class struggles in early industrial England, Stedman Jones has argued, was “the closer intertwining of industrial and political demands. …It is difficult to separate political and industrial demands in the thirties and forties because working class leaders themselves rarely did so. Universal suffrage, according to Doherty, the spinners’ leader, ‘means nothing more than a power given to every man to protect his own labour from being devoured by others.”’6 Such rhetoric would have been quite unusual in the United States, and the talk of suffrage gives us a first hint as to how to think about why. These differences in the degree and character of class understanding and activity, captured in microcosm in the differences between the rhetorical and organizational content of political gatherings at the drinking places of working people, indicate how the dominant tradition of English working-class historiography debilitates our understanding by assuming what must be explained and, in so doing, makes comparative studies of English and other instances of working-class formation very skewed. At the extreme, this perspective simply treats England as the paradigmatic case of class formation: As the first working class to be made and to make itself, it revealed the ineluctable logic of the capitalist mode of production. In this way, historical analyses of the English working class have tended to mimic Marx’s identification of England at the economic level as heralding the path of Western capitalist development more generally. Historical treatments of American working-class formation have suffered, not without irony, from an imitation of this approach. A good number of radical social historians in the United States (who are better described as neoprogressives in the Beardian sense than as neo-Marxists) have sought to recapture and reconstruct the long-neglected past of the working class. The large, suggestive literature they have produced too often assumes the aptness of the “making” teleology in English work. The important issue is not whether class was made at all in America or in England or elsewhere, but rather the terms on which a working class was formed. How shall we go about the task of explaining the puzzle of variation between English and American patterns of working-class formation? My starting point is an aphorism of Irving Howe’s: “The working class is a reality, the proletariat an idea.”7 Approaches to problems of class formation that are too structural and too teleological are unable to deal with real working classes, precisely because they address theoretically constructed proletariats expected by imputation of interest to appear and act in history in particular ways.8 But which concrete historical factors shall we place our explanatory bets on? This essay develops a state-centered explanation. I argue in a brief examination of “labor aristocracy” approaches to English working-class formation that economy-and societycentered explanations tend to make unwarranted teleological assumptions, overemphasize the place of work, and underestimate the significance of spatial arrangements, especially

Working-CIass Formation and the State: 145 as they concern the connections between work and home. I next show that even when we compensate for the shortcomings of such explanations by stressing the reorganization of city space under the impact of industrialization, by underscoring activities in workingclass residential communities, and by making contingent what some theorists assume to be “natural,” they still fail to account for the kinds of variations we find in the politics of the neighborhood and the saloon. On key dimensions, especially the consequences of capitalist and urban development for working-class residential spaces, the United States and Britain were very similar. The differences in English and American patterns of working class formation, I try to show, are best accounted for by the impact of the organization and public policies of their respective states. This is an essay, in short, about how states and their policies shape and inform the creation of meaning about class expressed in language, dispositions, and organizations. At issue is how the political contexts created by state authorities established the vocabularies and institutional forms that workers would develop to shape and represent their demands directed both to employers and to the state in the early industrial period. In explaining the contrasting cases of England and the United States, I stress the importance of variation in patterns of interest representation and repression rather than the similarities between the two states (porous bureaucracies, constitutional continuity, and the significance of the law), which are often lumped together too crudely under the macrocomparative rubric of the “weak” state. Without attention to the differences between these states’ organizations and public policies, I argue, it is not possible to understand the key differences in class formation. How and why the state provided a major source of variation in key features of the development of the English working class, considered in contrast to the American, is the analytical centerpiece of the essay.

Society- and Economy-Centered Explanations If working classes are expected to develop “naturally” in certain militant or revolutionary ways but do not, historians are tempted to search for alibis.9 Since most treatments of working-class formation in England take working-class moderation as the object of their explanations, they characteristically try to explain why the counterfactual of a revolutionary class has not been realized. Although this puzzle and this approach are very different from my own, I begin here for three reasons: It is the dominant approach; attention to it will help us to clear some important analytical ground; and some work in this tradition will help us to reformulate a precise, if dissimilar, object of analysis. “Labor aristocracy” explanations of both a crude and subtle variety are the most common “alibi” approaches. Using such diverse criteria as wage levels, regularity of work, trade union membership, styles of consumption and culture, and commitments to distinctive sets of norms and values, scholars have sought to find a fault line dividing the working class. The analytical payoff is a political one. As Henry Felling has observed, “It is an essential feature of the Marxist theory of the labour aristocracy that this supposedly small section of the working class was conservative in politics and imposed its conservatism upon working class institutions, thereby concealing but by no means eliminating the underlying militancy of the mass of the workers.”10


146

This approach is beset by a number of fundamental empirical and analytical flaws. The term “labor aristocracy” is inexact. It shifts in meaning from one treatment to another. A number of microlevel treatments of the relations of craft and manual workers in specific industrial settings indicate just how difficult it is to specify work locations where aristocrats of labor actually confronted a mass of workers. Contrary to expectations of the theory, empirical evidence, especially for the late nineteenth century, suggests that “it was the more prosperous workers who were the more politically militant and radical, while the lower ranks displayed either apathy or conservatism.”11 And if the leadership of unions did display a certain narrowness of focus, stressing immediate gains in wages, working conditions, and in public policy rather than more fundamental social transformations, what were their strategic alternatives? The labor aristocracy approach invites comparison with continental Europe rather than with the United States. This orientation highlights certain questions and obscures others, especially those with which this essay is concerned. Leaving this problem aside for the moment, we might note that the construction of a European foil for the English case depends on an entirely problematic vision of what actually happened across the English Channel. Reformism is not just an English issue. There are quite remarkable similarities in all the capitalist societies of Western Europe and North America with respect to the limits in practice of working-class challenges to the existing order. The implicit Continental contrast in much scholarship about the English working class operates on an entirely misleading model of sustained, heroic, protorevolutionary activity by European working classes.12 Perhaps an even more basic problem in the labor aristocracy position (one stressed in an important critique by H.F.Moorhouse) is its tendency to assert a priori the significance of a segment of the working class. Such a designation is not possible to assess unless working class culture as a whole is examined. At most, an explanation stressing a privileged stratum of labor would have to be the outcome of a historical analysis of causes of variation in patterns of working-class formation in specific settings rather than a model of cooptation and sellouts imposed on historical analysis.13 This reading and critique of the labor aristocracy approach, and, by extension, other equivalent answers to the question of why the working class did not behave as an essentialist class model predicts, begins to clear the ground to an alternative approach. It beckons us to shift focus away from the gradualist problématique of the Continental contrast toward a focus on the emergence of a holistic, or global, kind of class consciousness in England that an American comparison implores us to explain. It insists that we look in the first instance not at a fragment of the working class but at the language and activity of the group broadly conceived, and it demands that we reject the notion that there is one most likely course of working-class development. Instead, we must turn to history and its variations, even as we understand that each case of workingclass formation is one of a family sharing common traits—but what history and what traits? Let us return to St.Bernard’s saloon on Thompson Street and to Birmingham’s Crown and Anchor for some guidance about how to proceed. These public houses shared a number of traits that highlight basic features of capitalist industrialization and urbanization that are directly connected to the variations in patterns of working-class

Working-CIass Formation and the State: 147 formation we have identified in the English and American cases. These public houses were located in neighborhoods that were predominantly residential. Such local places of work as existed were outside the home. The communities in which they were found were segregated by class. The pubs—like local fire companies, gangs, burial and insurance societies, and other neighborhood-based organizations—were part of the fabric of new kinds of working-class residential communities. Unlike preindustrial city neighborhoods, where people tended to work in home-based settings and where the social classes were jumbled together, the new urban space separated work from home and the classes from each other. The significance of this spatial reorganization is apparent when we realize that what distinguished the early development of the English from the American working class concerned primarily patterns of culture and activity, not at the place of work, but at the place of residence. In both countries working people constructed tools where they lived to respond to the bewildering environments in which they found themselves. Although the forms assumed by working-class organizations in the two societies were virtually identical, however, their rhetorical and institutional contents diverged radically. The contrast between the American pattern of divided consciousness and the English pattern of a more holistic understanding of the significance of class is, at its core, a difference in understanding about the meaning and role of working-class neighborhoods and their links both to work places and to the political process. For this reason, before we attempt to account for this divergence we need to look closely at the spatial reorganization of the early industrial city. First, however, I should like to make a small detour. An important book, Patrick Joyce’s Work, Society, and Politics: The Culture of the Factory in Late Victorian England,14 provides a coherent alternative to the labor aristocracy position on post-1840s working-class gradualism by focusing on the working class as a whole. His treatment differs fundamentally from this essay not only in the puzzles it explores but also in its explicit rejection of the spatial dimension I think so important. By detouring briefly through Joyce’s book, I will be able to clarify my argument and meet the most cogent objections I know to it. Joyce focuses on the factory rather than on the residential community in Lancashire and the West Riding. Building on a distinction originally proposed by Stedman Jones between artisans subject to the formal control of capitalists and factory workers subject to “real” substantive control in the work process, he tries to show how the consolidation of the industrial factory system created a common reality for workers and how their experience of work shaped their institutions and mentalities outside of work. A culture of deference forged in the factory, he argued, was extended to all facets of social and political existence. Joyce acknowledges that the factory and the neighborhood had become separate entities, but he maintains that the factory and its social relations were at the heart of neighborhood feeling and that the class segregation of working-class housing areas was nestled into a larger, well-ordered system of deference. In this view, working-class subcultures were but an aspect of an elaborate network of exchanges between the classes presided over by manufacturers who dominated not only the politics but the cultural expectations of their town and whose sway over working-class life was the product of


148

their deeply rooted, factory-based hegemony. He maintains that as the industrial process became more developed factory life penetrated more deeply into other spheres of life; the less complete the mechanization of production, the less powerful was the hold of the paternal culture of the factory. In advancing this claim, Joyce challenges not only the view, which I hold, that the growing separation of place of work from place of residence made possible an autonomous working-class cultural and political life, a matter I shall return to, but also the core argument of the labor aristocracy position, which he turns on end (and in so doing reproduces its corporeal shape). “Far from being the ‘moderates,’ “he writes, “the labour aristocrats were in the forefront of radical politics in this period…. The politics of labour…were chiefly the concern of those outside the cultural environment of the factory, and especially the craft and skilled sectors in the working class. If these were a labour aristocracy then it was independence rather than reformist ‘collaboration’ that was their political legacy.”15 In turning upside down the labor aristocracy position, Joyce adopts its tendency to overstate the divisions between craft and factory workers at the work place (writing about France, William Sewell has shown how factory transformations may in fact provide for new definitions and extensions of craft labor) and leaves unaccounted for the immense number of workers, like those in the engineering trades, who fall neatly into neither broad category.16 Furthermore, although it is clear that the “aristocrats” by definition provided leadership to the trade union and party politics of the late nineteenth century, it is by no means clear that they were “radical” as opposed to “gradual,” nor is it obvious that they represented an elite without a mass, factory-based following. For the trade unions such a claim borders on the ridiculous, and for the electoral process the question must remain opaque because most factory workers could not vote until the end of the First World War. From the perspective of the argument I am developing, Joyce’s mimetic treatment of the aristocracy thesis is not merely a matter of secondary interest. It points to the failure of both positions to make crisp distinctions among three axes of differentiation of working class members: (a) workers relatively privileged at work (whether because of their wages or the character of the labor process); (b) workers with different styles of life in residential communities segregated by distinctions not only between the working class and other classes, but within the working class itself; and (c) workers who provided the political leadership for trade unions, pressure groups, and political parties. The various attempts to collapse all of these axes into one grand axis of internal working-class division obscure the interesting contingent questions about the relationships among these three (sometimes overlapping) sets of actors. From my point of view these distinctions are crucial, because without them it is impossible to ask why, and under what circumstances, political leaders make demands at work directed at employers and, off work, directed at the state in the same terms (as broadly was the case in England) and when they press their demands in utterly different terms in each arena (as broadly was the case in the United States).

Transformations of Urban Space

Working-CIass Formation and the State: 149 What of Joyce’s challenge to the stress on the spatial separateness of work and community and its implications? He takes some pains to deprecate the significance of this separation. The factory town of the mid-nineteenth to late nineteenth century, he writes, “retained more of the village than it acquired of the city. Understood as the ‘walking city’ the factory town grew by cellular reproduction, the town slowly absorbing factory neighborhoods in its expansion.” Until the introduction of tramways and bicycles at the turn of the century, he continues, “the link between home and work remained firm until these severed it.”17 If Joyce is correct, then a state-centered explanation of the single, rather than divided, quality of class understanding in England is not necessary. We might then argue that differences in class formation between the United States and England simply are reflections of different patterns in the objective organization of spatial relations. In this approach the split consciousness of the American working class would be explained by changes in city space in the antebellum period, a subject I have tackled elsewhere.l8The more unitary English pattern, by contrast, would be accounted for by the more tight integration of work space and home space. It is not possible to sustain this line of reasoning, because Joyce is wrong. There was a major reorganization of space in the very large and middle-sized industrializing cities and towns in the middle decades of the nineteenth century. The most dramatic changes came about in the largest cities; indeed, one cannot help but wonder if the exclusion of Liverpool and Manchester (Lancashire’s most important centers of working class concentration) did not bias Joyce’s view of the relationship of work and off work. But even in smaller cities (we shall look at the Lancashire factory town of Chorley later) it would be difficult to overstate the importance of changes in and about space. The large-scale urbanization of British society is one of the most striking features of the nineteenth century. None of the ten largest cities less than doubled in population between 1801 and 1851 (compared with only one of the top French cities in the same period). At this moment of astounding growth, Leeds and Birmingham tripled in size; Manchester and Liverpool quadrupled. All of the ten largest cities had populations of over 100,000 by 1851, with Liverpool and Manchester well over 300,000. By midcentury Liverpool’s upper and middle classes “who could afford to do so had moved from their place of work or business or even beyond it into adjacent townships.” Their residential areas in new suburban rings were divorced from both the work and the residence places of workers. Within the heart of the city there were “distinctive zones of different economy and society.” Dividing the city into eight areas, Lawton has shown with great clarity how Liverpool space had become sorted out into spatial divisions of the classes from each other.19 Other research has carefully established that most of the housing constructed in early-nineteenth-century Liverpool was built for the new working classes in class-segregated residential neighborhoods. By midcentury, workshop houses had become domiciles of the past for the majority of Liverpool residents and those of other industrializing cities.20 Joyce might suggest that the big-city pattern is atypical, but very fine recent studies would put such an assertion in question. A.M.Warnes, for example, has studied the Lancashire town of Chorley between 1780 and 1850. In the manner of Sam Bass Warner’s study of Philadelphia, he takes three “snapshots” of the town in 1780, 1816,


150

and 1850.21 Before 1780 the town was an agricultural township, serving the market and commercial functions of its rural environment. Its growth began to accelerate around 1780 as a result of the weaving trade stimulated by the mechanization of spinning. Even so, its landscape continued to be defined by its preindustrial functions: “Nowhere had large employing units been formed; most people still lived at or immediately adjacent to their place of employment.”22 Between 1780 and 1816 the town’s population increased by 50 percent to 6,000, a growth stimulated mainly by employment in early, small textile mills. At this date factories and clusters of houses around them became part of the town’s topography, but the growth of the textile industry can be accounted for more by the enlargement of existing workshops than by their replacement by factories, and “residential location was still to a great extent determined by the location of employment and not usually by the social differences among the population.”23 Nevertheless, the 1816 social geography of the town was rather more complicated than that of 1780. A somewhat more differentiated pattern had developed, characterized by a more well defined commercial core, a more rapid pattern of growth in outlying parts, and scattered but discernible clusters of settlements associated with specific new occupations, including calico works and cotton spinning factories. By 1850 the town had been radically transformed. In the years since 1816 Chorley had become a factory town. Reporting on the same transformation stressed by Joyce, Warnes notes that handloom weavers, once the majority of the population, were reduced to under 10 percent of the work force. Using the enumerators’ books of the 1851 census, he estimates that “over one half of the economically active were engaged in branches of the textile industry that necessitated a journey to work.” Furthermore, he finds not only that there was a dramatic shift from domestic employment, but that “by 1851 the proximity of home and workplace was breaking up, at least for certain sections of the workforce,” and “the conditions for a recognizable pattern of social segregation were beginning to develop.”24 It is important not to overstate the modernity of this pattern. Only a negligible number of workers (about 4 percent) lived more than a mile from their work, and over half of the 1,813 people Warnes examined lived within a quarter-mile of their work places. Of the two in three workers who no longer labored where they lived, the majority had to move only short distances to overcome the separation. “Nevertheless there is no doubt that the average distance travelled to work had sharply increased since 1816, not only because more workers lived away from their work but also because of the increasing variety of employment.” In a striking finding Warnes reports that, the newer the factory, the farther away were the locations of residence of its workers.25 He also hypothesizes (contra Joyce’s emphasis on tied loyalties) that with the diversification of work places there may have been a reduction of commitment by workers to a particular factory or employer, as well as a diversification of family employment in different factories. These changes probably increased the average distance of family members from work. Residential differentiation based on occupation was being replaced by a “tendency for those with similar incomes, educational levels, or other status variables, to congregate in limited parts of the town.”26 Paradoxically, belonging to certain occupations (spinning, printing, and bleaching) made such a divorce between

Working-CIass Formation and the State: 151 occupations and residence possible. Joyce concludes his treatment of the residence community with a caution. The impersonality and class segregation of the twentieth-century town are so often read back into the nineteenth century that we forget just how unlike the towns of these two centuries were. As words of prudence these imprecations are well taken, but for the purposes of our discussion they largely miss the point. Writing about the industrial revolution more generally, Sewell has shrewdly observed that “what now appears as the hesitant beginnings of a long and slow development seemed to be a major departure to contemporaries: Even a few steam engines or blast furnaces or spinning mills could make a powerful impression on people who had never seen them before. From their point of view, modern industry was a distinctive feature of their age; theirs was an industrial society as no previous society had even seen.”27 So with the spatial concomitants of industrialization. What mattered not at all to contemporaries of the nineteenth century was the degree of difference between the nineteenth- and twentieth-century city; impressed on their lives and consciousness was the new urban form that introduced patterns of separation between work and off work not just for the very wealthy but for the majority. In some towns places of work and residence may have been only a short walk away, but they were places apart, and except in the case of employer-provided housing (by best estimates only a small fraction of working-class housing, even in single-factory towns) workers came to cluster in communities separate from the locus of work, from other classes, and from other workers with different (Weberian) class attributes. This revolution in space produced a fundamental change in relationships of authority and in forms of local organization. With the exception of company towns, where factory masters held sway over the totality of working-class life, the role of capitalists was constricted to the work place, freeing workers to live without direct supervision by their bosses in their neighborhoods. This new freedom was undergirded by a new division between markets for labor and markets for housing. In the workshop structure of employment and residence, the two had moved together to the same rhythm. With the sundering of this link, working-class communities came to be shaped more and more by real-estate speculation and by the strategies of builders and landlords. They became stratified by income and styles of life. And they came to be conscious environments that people of all classes sought to shape and control.28 In these new environments, working-class people created new institutions, new relationships, new patterns of life. The contingent connections between these and broader patterns of politics and class formation have too long been neglected in favor of such explanatory shortcuts as the labor aristocracy hypothesis. There is, however, one major exception to the rule of the neglect of space in this approach to class formation. It is worth our attention, because it gives us important clues about how to shape our analytical questions and our approaches to answering them. This exception is John Foster’s Class Struggle and the Industrial Revolution,29 the most explicit (I would say rigid) Leninist version of the aristocracy thesis. Its great merit is to have introduced social space into its arguments about working-class reformism. Foster’s study of Oldham is rich with hypotheses about cross-class alliances, demographic and production patterns, and the formation of a specific kind of labor aristocracy. I shall leave


152

most alone, but I do wish to highlight Foster’s unusual sensitivity to the implications of spatial transformations in the early industrial town and his attempt to incorporate space as a constitutive element of the social structure. The transformations of the textile industry in Oldham were not merely changes in the organization and techniques of production. When production was organized in cross-class households, where work and off work were tightly integrated, a patriarchal model of class and authority based on face-to-face paternal relations and the direct supervision of the various spheres of social life integrated the social order. The material and ethical claims of this system could not survive the industrial process that shattered household patterns of production. Foster shows that during the first half of the nineteenth century Oldham’s workers came less and less frequently to live where they worked. As the town’s districts came to be designated increasingly either as places of work or as places of residence, members of the working class came to live apart from their employers and their supervision. By devising their own social and cultural institutions where they lived, working people fashioned an independence from the patriarchal claims of their employers. These institutions, rooted in the relatively free space of the working-class community, provided the potential basis for the construction of relatively independent sources of class and group dispositions and collective action.30 Although Foster’s contribution more than shares in the dubious features of the labor aristocracy position, his joining of space, class, and politics has the particular merit of making conditional the role that working-class neighborhood institutions might play in the larger story of working-class life. Because of his focus on the traditional historiographical problem of working-class reformism, Foster stresses the contingent aspects of these social and cultural institutions with respect to whether they became autonomous organizations capable of building working-class solidarity against capitalists and the state in spite of divisions among workers or whether they became differentiated between the community institutions of the “aristocrats” and the rest of the working class (with the former becoming closely tied to bourgeois ideology, politics, and authority). What Foster does not deal with are the related, but different contingent questions of whether and under what circumstances workers will utilize their newly autonomous institutions in the residence community to mount demands not only against their employers but against the state in class terms. Put another way, what will be the relationship between the politics of labor and the work place and the potentially separate emerging politics of community?31 Of all the Western industrializing countries, the urban-spatial histories of the United States and England in the nineteenth century were the most alike.32 So too was the array of territorially based working-class institutions that provided a dense infrastructure to community life.33 As workers in both countries experienced the separation of work and community, they developed ways of construing and acting on these experiences in their new residential spaces and institutions. Precisely because of the relatively tight controls that employers continued to exercise in the posthousehold factory (a point stressed by Joyce), the residence community came to loom so important as the place where workers could create meanings and practices in partial freedom.

Working-CIass Formation and the State: 153 Let us return once again to some of these local institutions. Historians of American neighborhood and political party life have long been impressed by the combination of sociability and political action of the local party machine. The same dual role of leisure and instrumental activity characterized Chartist meetings in workingmen’s clubs, friendly societies, and public houses. Rowley writes, for example, of annual dinners held to “commemorate the birthday of Tom Paine, enlivened by radical songs, toasts, and recitations,” and, more generally, of heavy drinking during business meetings.34 Whereas such activities underscored a divided working-class consciousness in the United States, they helped promote a more global consciousness of class in England. Quite unlike the American patterns, where even early trade unions developed meeting places at or very near work places, English trade union branches, many secret and illicit, usually met in working-class neighborhoods, most frequently in pubs. The organization of industrial disputes was often pub-based, and pub landlords frequently acted as strike coordinators or union treasurers. Many unions were identified with specific public houses, and some important Chartist pubs (like Nottingham’s Seven Stars and the King George on Horseback, the landlords of which acted as treasurers of their local Chartist cells)35 simultaneously were prominent meeting places for industrial and political organizers. Friendly societies, the main self-help community organizations of English workers, also gathered in ale houses. Intensely local in character, many of these societies were trade unions by other names that sought to protect their funds and their members from legal recrimination.36 The quality of the “trade union consciousness” created by this compressed set of working class organizations in the residence community obviously is rather different from the “trade union consciousness” that came to prevail in the United States, where from the very earliest moments of trade union history labor unions were disconnected from other institutions and locations of working-class life and political activity. From the vantage point of the United States, the key issue for studies of English working-class formation is whether we can account for the broad cultural pattern identified by Felling as a combination of political and social “conservatism, associated however with a profound class consciousness and quite commonly a marked sense of grievance.”37 This distinctive mix of political caution, stressed by most historical analysts, and a keen, pervasive sense of the relevance of class divisions to all spheres of life (a situation described by Eric Hobsbawm in the language of “high classness”)38 was very different from the American pattern with respect to the second, but not the first, characteristic. Obviously, the English pattern was the result of a historically contingent process. What were its causes? One way to approach this question is to ask why both trade union and political agitation in England but not the United States were pressed into institutions of residence community. There, in the voluntary organizations created in the “free space” of communities separated from work places, English workers learned to put claims to their employers and to the state in a rhetoric and idiom of class. What accounts for this development? Obviously, the industrial urban, spatial, and organizational patterns that English workers shared with their American counterparts cannot explain the divergence between the two histories of working-class formation.


154

Equally clear is the inability of society- and economy-centered explanations, rooted more in what these societies share than in their differences, to account for these variations. In the remainder of this essay I propose that important differences in the organizational structures and the public policies of the states in the two countries account for the key divergences we have identified. The claims I shall make, I hasten to add, are provisional—in Charles Tilly’s words, “more solid than working hypotheses, perhaps, but less firm than theses one nails up, to challenge all the world, at the end of a long inquiry.”39 State Structures and State Policies A focus on the state as a major source of variation in patterns of class formation in the English and American cases at once confronts an important feature of the macrocomparative literature on the state: the tendency to lump these two countries together under the rubric of the Anglo-American “weak state.” J.P.Nettl’s seminal essay proposing that “stateness” be treated as a variable cites Marx favorably for observing that the United States and England were excluded from the “necessity of violent overthrow of the state because there was no state as such to overthrow.” In neither country is the state “instantly recognizable as an area of autonomous action,” and in both, the law, rather than being an emanation of the state, has a great deal of autonomy.40 Most analyses of the “weak” American state, from Tocqueville’s Democracy in America to more recent treatments such as Huntington’s well-known analysis of the diffusion of the Tudor polity,41 root the U.S. experience in the British for the obvious historical reasons. In this largely persuasive line of analysis, England is understood to be an aberrant case of state formation in early modern Europe. The Glorious Revolution of 1688, though radical from a Continental perspective, ratified the traditional prerogatives of Parliament against those of an absolutist alternative. Sovereignty was defined not by a conflation of the civil and the civic within the state but in terms of representation and, ultimately, a system of political parties linking state and society through electoral mechanisms. Although English sovereignty crystallized in a single representative body, indicating a pattern of centralization very different from the American federal system, where sovereignty ultimately resided in the “people,” both countries shared a very clear contrast with the state patterns of continental Europe. These states, unlike those of Prussia or France, did not have a monopoly of access to technical, professional information. Furthermore, for reasons Barrington Moore explains, “England’s whole previous history, her reliance on a navy instead of an army, on unpaid justices of the peace instead of royal officials, had put in the hands of the central government a repressive apparatus weaker than that possessed by strong continental monarchies.”42 Colonial America, of course, shared in and inherited the characteristics of the British state, whose Crown bordered on an abstraction and which was relatively undeveloped as an autonomous entitity demarcated from civil society.43 Although one might be tempted to point to an apparent correlation between strong states and militant working classes in continental Europe and to the seeming correlation

Working-CIass Formation and the State: 155 between weak states and reformist tendencies within the working classes of England and the United States, the consequences of the weak states in both England and the United States compared with the absolutist and bureaucratic regimes of the Continent are not self-evident. Propositions along these lines would have to depend on caricature to the point of distortion. Continental working classes have hardly displayed a uniform propensity for revolutionary or protorevolutionary activity, just as the English working class has known moments of militant contention (E.P.Thompson, for one, thinks that in 1832, as in 1819, the year of Peterloo, “a revolution was possible” because “the government was isolated and there were sharp differences within the ruling class”),44 and American workers have challenged capital at the work place with great courage and at high risk and cost. If it is difficult to move directly from an assessment of a given state as “strong” or “weak” to statements about revolutionary versus reformist working-class orientations to the regime, it is even more difficult to make this kind of causal claim if we want to explain whether class, as a category of understanding and action, joined the spheres of work and community or was limited to the sphere of work. The main attempt I know of to apply this line of reasoning to the English case, that of Kenneth Dyson, is flawed in ways that an American comparison reveals. In his view, England’s weak state “helped to cement polarization in the industrial and political systems…. Accordingly, the experience of a relatively ‘unbridled’ capitalism gave it a bad reputation, created a powerful and isolated working class culture, and undermined willingness to cooperate both in industry and in politics. Britain acquired a peculiar class structure: the obstinacy and distrust associated with attitudes of ‘them’ and ‘us’ were not directed at the state but at bastions of privilege and exclusiveness associated with society.”45 We shall see, from an American vantage point, that it is misleading to say that working-class dispositions and structures of feelings in England were not honed in opposition to the state. The American experience, moreover, in which “we”-”they” distinctions based on class were constricted largely to employee-employer conflicts at the work place, appears to belie this line of argument. But appearances can be deceptive, for if we inquire about the precise organizational forms that the “weak state” took in England and the United States; if we examine the different clusters of political rights the two working classes possessed in the early industrial period; if we look at other bundles of public policies; and if we explore the effects that these differences had on the content of working-class organizations located where workers lived, we shall be able to make more persuasive connections between the characteristics of these states and their activities and divergent patterns of working-class formation. The most important political right in the nineteenth century, of course, was the right to vote. Here the differences between the United States and England were marked. Virtually all adult males in the United States could vote by the early 1830s, but in England only one in five adult males composed the eligible electorate after the reforms of 1832, one in three after 1867, and only three in five even after 1885. How important was this difference for our puzzle? Accounts of the genesis of the franchise and the story of its expansion have long been familiar themes in treatments of the formation and civic incorporation of Western


156

working classes. Reinhard Bendix, elaborating the scholarship of T.H. Marshall, has suggested that there is a common sequence of the extension of voting citizenship in the West that helps to explain the shared political gradualism of the various national working classes. By contrast, H.F.Moorhouse explains English reformism (the standard object of analysis), not so much by civic incorporation through the franchise, but by the long period of exclusion of workers. The differences between the protagonists of this debate, however, represent two sides of the same coin,46 since all the participants in the discussion agree on what is not in doubt: that franchise extensions and their timing have had an important effect on working-class propensities to act in certain ways. They disagree only about how. Here, too, a comparative American perspective is chastening, for the history of the franchise in the United States reminds us that gradualist outcomes do not depend only, or necessarily, on a protracted period for the extension of citizenship rights to individuals. What is so striking about the familiar discussions of the franchise and class is their avoidance of the questions about class formation that are at the heart of the comparative puzzle with which we are concerned. The quick, nearly conflict-free extension of the franchise to men in the United States and the much more protracted, conflict-ridden expansion of the right to vote in England are central to explanations of the differences between the global consciousness of class on one side of the Atlantic and the divided consciousness of class on the other. But the impact of the variation of political rights on class formation was not a direct or simple one. To understand it, we have to take up the franchise in combination with another key difference between the two countries concerning the organization of the state: that of a federal versus a unitary state. Nettl, it will be recalled, who introduced the concepts of “weak” and “strong” states into contemporary comparative-historical political sociology, stressed that the United States and England shared relatively diffuse state organizations, but he paid insufficient attention to the differences in the ways their states were organized. The diffuse federal organizational structure of the United States took much of the charge out of the issue of franchise extension, for there was no unitary state to defend or transform. Once suffrage restrictions were lifted (in tandem with other democratizing reforms, such as an increase in the number of public offices and in the regularity and frequency of elections), the United States had the world’s first political system of participatory federalism. Within this framework of state organization and democratic political rights, new kinds of political parties with a mass base were constructed. These political organizations, reaching into virtually every ward and neighborhood in the country, put together the spatial, ethnic, religious, and political identities of the various subgroups of the American working class. This act of organization and social definition took place where workers and their families lived. It created direct links between the political system and voting citizens, who were organized into politics on the basis of many identities but rarely those of class as such, and this pattern created an institutional and participatory structure that was set apart from the organizations workers created to put their demands to their employers. American participatory federalism was not just a system of voting based on intensely local solidarities. It was also a system of governance, taxation and delivery of services.

Working-CIass Formation and the State: 157 During the antebellum years in the cities, where the bulk of working people lived, municipal services provided the main content of a local politics of patronage and distribution. This period saw the introduction of professional police forces, the bureaucratization of municipal charity and poor relief, and the establishment of mass public school systems, as well as a massive program of publicly licensed construction. Political parties focused on the connections between these services and the various neighborhoods of the city. Local politics became a segmented and distributive politics of community. In this politics, workers appeared in the political arena not as workers but as residents of a specific place or as members of a specific (nonclass) group. Although there were many class-related economic issues on the national and state political agendas, including internal improvements, tariffs, control of banking, and slavery, the process of voting for most workers was insulated from these concerns, because it was focused elsewhere. By contrast to the United States, the interplay of state and class in England was radically different in each of the respects I have noted. The unitary (not federal) English state concentrated distributive public policies at the center as a result of the passage of such public acts as the Poor Law of 1834; the Public Health Acts of 1848, 1866, 1872, and 1875; the Police Acts of 1839 and 1856; and the Food and Drug Acts of 1860 and 1872. Whether this growth was mainly the result of the initiatives of humanitarian and Benthamite civil servants and parliamentarians or the result of attempts to defuse classbased opposition of industrial capitalism (these are the two main poles of a lively historiographical debate),47 the growth of the central government was staggering. In 1797 most of the 16,000 employees on the central government payroll were customs, excise, and post office personnel. By 1869 this number had grown to 108,000, reflecting the new activities undertaken by the administrative agencies of the state. For the century as a whole, public expenditure grew fifteen times in real terms.48 If patronage and property were the two hallmarks of the pre-nineteenth-century civil service, by midcentury patronage had been sacrificed to “save the main pillar, property.”49 The professionalization of governmental administration at the center, a process that marked the period from 1830 to 1870, further focused attention by members of all social classes on the center of the state apparatus in London. As Parliament and Whitehall took on new responsibilities for social policy and the regulation of working conditions and as they reorganized their affairs to rationalize the enforcement of regulations and the delivery of services, they potently reached into and affected workingclass life, at the work place but especially in the residence community. Each new parliamentary act and each new wave of administrative expansion and reform focused ever the more sharply the attention of the working class on the state and its activities. From the perspective of a working person (and from an American perspective), the state that he or she confronted was not a “weak state.” The comparative intensity of the tie between the working class and the central state was powerfully reinforced by the second aspect of what Harold Perkin has aptly called the nineteenth-century “battle for the state”: the struggle for the franchise.50 Just as the services and the regulations of government penetrated working-class life and thus reinforced class identities at the residence place as well as at the work place, so the demand by English workers for the right to vote could have been constructed on no


158

other basis than that of class, for the working class was excluded on explicit class criteria. Only with universal suffrage, Heatherington wrote in the Poor Man’s Guardian in December 1831, would “the term classes merge into some comprehensive appellation, and no bloodshed will ensue.”51 Although working-class reform organizations before and after 1832 were local in organization, unlike American political party machines they directed their claims of citizenship to the center, that is, to Parliament. One effect of the interplay between the state and workers where they lived was the creation of a common fault line, based on class, in all parts of English society: One of the distinguishing features of the new society, by contrast with the localism of the old, was the nationwide character of the classes, in appeal if not always in strength. At some point between the French Revolution and the Great Reform Act, the vertical antagonisms and horizontal solidarities of class emerged on a national scale from and overlay the vertical bonds and horizontal rivalries of connection and interest. That moment…saw the birth of class.52 When every caution has been made, the outstanding fact of the period between 1790 and 1830 is the formation of the working class. This is revealed, first, in the growth of class-consciousness: the consciousness of an identity of interests as between all these diverse groups of working people and as against the interests of other classes. And, second, in the growth of corresponding forms of political and industrial organization. By 1832 there were strongly based and self-conscious working class institutions—trade unions, friendly societies, educational and religious movements, political organizations, periodicals— working class intellectual traditions, working class community-patterns, and a working class structure of feeling.53 English working-class voluntary organizations turned outward in two respects. Rather than reinforcing local particularities based on intraclass differences of territory, income, or craft, they linked the activities and sensibilities of workers to each other across these lines. Furthermore, they joined to the concerns of the residence community the class issues of political participation, public policy, and trade unionism. In this respect, the Chartist movement provided the most important post-1832 concretization and deepening of these tendencies, which contrast so sharply with the role played by neighborhoodbased working-class voluntary organizations in the United States. Chartism provided the unifying pivot of English working-class dispositions and organization. This movement took a working class that had oscillated from the turn of the century between economic and political action and fused both kinds of action in a national network of community-based associations. For some two decades, Chartism created a distinctively English global consciousness of class harnessed not only to the campaign for votes, but to poor law agitation, to trade unionism, to factory reform, to Owenite socialism, to an unstamped press, to millenarianism, and to machine breaking. In all these activities Chartism reflected the great diversity of a differentiated working class. Where there was a substantial number of artisans, Chartist associations tended to stress values of self-help and independence. Where handloom weavers predominated, as in Lancashire and the West Riding, the form and content of agitation tended to be more

Working-CIass Formation and the State: 159 strident. Where domestic industry predominated, as in the East Midlands, workers were more likely to seek allies and guidance from middle-class reformers.54 Overall, however, there were consistent themes to Chartism. These included the attempt to build an independent political voice for laborers based on class understanding, as well as the regular elaboration of links between economic problems and political representation. At its most vigorous, Chartism swallowed up other working-class movements and gave them a common definition. Its key feature, J.F.C.Harrison has argued, was “its class consciousness and temper…. Chartists of many shades of opinion emphasized their movement was concerned to promote the interests of working men as a class.” They “assumed the need for class solidarity and their leaders talked the language of class struggle.”55 From the point of view of the historiographical problematique of gradualism, Chartism seems to be another instance of working class reformism. From the vantage point of the American comparison, the important feature of Chartism is its scope and depth as a class institution and its posing of a democratic, egalitarian alternative to the existing political and economic order based on a class analysis. As Trygve Tholfsen has stressed so tellingly, the People’s Charter was not just a political document but a coherent classbased set of demands connecting all spheres of society. “Implicit in the Charter was both a demand for the transformation of the structure of politics and the broader principle that working men ought to exercise control over every aspect of their lives…. What set Chartism ideologically apart from middle-class liberalism…was the conviction, often only tacit, that class was the crux of the problem of progress and justice.”56 This view was reflected at the level of organization. Unlike American political parties, which were class specific in the neighborhood but were otherwise interclass institutions, the Chartist “party” was entirely independent of the Whigs and Tories, and over time, the Chartists drew away from middle-class allies. If Chartism disconnected the working class from other classes, it joined the political to the economic aspirations of workers, and it was affected as a movement by economic conditions and struggles. The dominant classes and politicians who successfully managed to use the state to deny workers the vote in 1832 tried to crush trade unions both before and after. As a consequence, it was possible for Chartism to become the common core of the working class and thus to impose itself as the common sense of a holistic rather than a divided kind of class consciousness, in tandem with a trade unionism quite different from its American counterpart. Just as the absence of political citizenship pressed political agitation into the autonomous institutions of working-class localities, so the state’s stance with respect to union activity produced relatively weak unions and compelled workers to bring their work-place organizational efforts into the protected space of the residential community. The Combination Acts of 1799 and 1800 made union organization very risky and thus drove labor organization underground, to the pubs and friendly societies of hospitable neighborhoods. Even after the acts were repealed in 1825, the state used the common law to interdict strike activity. After 1825 trade unions were harassed, though not legally suppressed; unions, as such were no longer unlawful, but action in restraint of trade was. The Combination Act of 1825 did not by itself make strikes illegal, but it left to judicial opinion whether a strike


160

was in restraint of trade. This legal situation did not prevent the development of unions; on the contrary, there was an explosion of public unionization attempts, sometimes on a grand scale, in the late 1820s and 1830s. But the fact that “after 1825 the common law was invoked against trade unions to an unprecedented extent” produced a situation that compelled unions to keep their planning and activities as secret as possible, hidden from the authorities in community-based associations.57 Moreover, the legal climate of repression combined with the pressures on union continuity provided by the operation of business cycles made unions especially fragile institutions. At moments of economic depression the state could leave to economic forces the role of restraining unions, but when the cycle proved favorable and unions grew more bold, the state could invoke the doctrine “to the effect that any overt positive action by groups of workers was likely to be a wrong in the nature of Conspiracy, even where the strike was quietly conducted.”58 As a result of these political forces, English organized labor was too fragile to provide an independent basis of action against capital. English workers were pushed out of the arena of work into politics and into the residence community at the same time. Although the trade unions themselves never joined Chartism, “the greater number of trade unionists,” Slosson found, “declared for the Charter.”59 In the United States, by contrast, the comparatively mild character of state repression against trade unions made a clearly defined, public, work place-based organization of workers possible. There, too, workers were liable for combinations under the common law, but in the American federal system there was no national legislation or central direction to antiunion prosecutions. Charges brought in one locality were not brought in another, and prosecutors found convictions difficult to obtain in a system of trial by peers rather than by magistrates. American unions were buffeted by economic crises throughout the nineteenth century and, in the century’s later decades, were faced with a wave of public as well as private attempts to repress labor organization. In the early period of class formation, however, the problems that unions had with the law were quite secondary. In times of economic prosperity workers were not inhibited on the whole from joining unions at the work place; in the middle 1830s, for example, some two in three craft workers in New York City were unionized.60 The organizational forms of the American and English states and their constitutional and public policies, in short, had very different consequences for the political content of local working-class associations and, by extension, for patterns of working-class formation in each country. In the United States, political agitation for the vote was unnecessary, and the community provided the location for the organizations at the base of inter-class political parties that appealed to voters by mobilizing nonclass solidarities. Unions, in turn, were allowed a separate existence. In their embrace a working class was formed as labor. Not so in England, where only the institutions of the residence community were available to workers through which to put demands to employers and to the state. Pressed together by the exigencies of law, repression, state organization, and public policy, the locality-based voluntary organization fused the separate facets of working class life into a common, deeply felt consciousness of class.61

Working-CIass Formation and the State: 161

Notes I am pleased to acknowledge the many helpful comments I received on earlier drafts of this chapter from the participants at the Social Science Research Council Conference, “States and Social Structures: Research Implications of Current Theories,” held at the Seven Springs Conference Center, Mount Kisco, New York, February 25–27,1982, where it was first presented (especially those of Ken Sharpe and Theda Skocpol) and also from J.David Greenstone, Michael Harloe, Richard Harris, and Henry Felling. 1. M.R.Werner, Tammany Hall (New York: Doubleday, Doran, 1928), p. 135. 2. Brian Harrison, “Pubs,” in The Victorian City, ed. H.J.Dyos and Michael Wolff (London: Routledge & Kegan Paul, 1973), p. 180. 3. Roy V.Peel, The Political Clubs of New York City (New York: Putnam, 1935) p. 38. 4. Ira Katznelson, City Trenches: Urban Politics and the Patterning of Class in the United States (New York: Pantheon Books, 1981). 5. Asa Briggs, “The Language of ‘Class’ in Early Nineteenth Century England,” in Essays in Labour History, ed. Asa Briggs and John Saville (London: Macmillan, 1967), p. 43. Briggs has also written one of the few well-crafted comparisons of English and Continental patterns of class formation. See his “Social Structure and Politics in Birmingham and Lyons,” British Journal of Sociology 6 (January 1950): 67–80. 6. Gareth Stedman Jones, “Class Struggle and the Industrial Revolution,” New Left Review, no. 90 (March-April 1975): 56, 58. 7. Irving Howe, “Sweet and Sour Notes: On Workers and Intellectuals,” Dissent, Winter 1972:264. 8. I might add that many scholars who find themselves on the culturalist side of the structuralist-culturalist debates raging among social historians share with Briggs the idea that a working class that experiences a society arranged in class ways will be made, and will make itself, into a class. This formulation is implicit in E.P.Thompson’s magisterial volume, The Making of the English Working Class (London: Victor Gollancz, 1963), just as it inhabits lesser contributions that take working-class reformism as the main object of analysis. The three journals in which these discussions are most accessible are History Workshop, Radical History Review, and Social History. 9. One of the problems this approach to English working-class formation poses is worthy of comment in passing. By taking the class orientations of workers for granted, debates about reformism and militancy are oddly divorced from the actual content of the society’s discourse, organizations, and competing capacities of classes. The result, as Eley and Nield acutely observe, is an imitation by Marxist historiography of the liberal “intellectual schematization—if workers go on the barricades they are ‘revolutionary,’ if they do not they are ‘reformist’ or ‘integrated,’ “a perspective that “misses the realities that drive workers to action or condemn them to inaction.” In accepting this dichotomy much Marxist historiography has appended to it the assumption that a formed working class should develop in certain ways. If it does not, a “theoretical alibi” for the working class as a whole has to be found (Geoff Eley and Keith Nield, “Why Does Social History Ignore Politics?” Social History 5 [May 1980]: 260–61, 258). 10. Henry Pelling, “The Concept of the Labour Aristocracy,” in Popular Politics and Society in Late Victorian Britain, ed. Henry Pelling, (London: Macmillan, 1968), p. 41. Rooted in an


11. 12.

13. 14. 15. 16. 17. 18. 19. 20.

162

article by Engels in 1885 and in Lenin’s treatment of imperialism, the concept of a labor aristocracy was introduced into contemporary historical scholarship by Eric Hobsbawm. Work in this tradition has examined wages and conditions of work (including the divide between craft and factory workers), urban neighborhood differentiation (emphasizing spatial segregation, housing conditions, voluntary organizations, and leisure activities—the focus I find most congenial), and politics (with a look at the positional location and world views of labor union and party leadership). For the development of the concept of the labor aristocracy, see Friedrich Engels, “England in 1845 and 1885,” in On Britain, ed. Karl Marx and Friedrich Engels (Moscow: International Publishers, 1934), pp. 67, 99; and E.J.Hobsbawm, “The Labour Aristocracy,” in Labouring Men, ed. E.J. Hobsbawm (London: Weidenfeld & Nicholson), 1964. The classic treatment of working-class reformism is that of Walter Bagehot, The English Constitution (London: Collins, 1971). Other approaches include a stress on value systems and political culture, understood in the limited sense of the aggregation of individual orientations to politics (see Gabriel Almond and Sidney Verba, The Civic Culture [Boston: Little, Brown, 1965]); a macrohistorical emphasis on the distinctive sequence of British political development (see Otto Kirchheimer, “The Transformation of the Western European Party Systems,” in Political Parties and Political Development, ed. Joseph LaPalombara and Myron Weiner [Princeton, N.I.: Princeton University Press, 1966], pp. 17–200); and on the debilitating character of available class alliances and ideologies (see Perry Anderson, “The Origins of the Present Crisis,” in Towards Socialism ed. Perry Anderson, and Robin Blackburn [Ithaca, N.Y.: Cornell University Press, 1966], pp. 11–52). Pelling, “Concept of the Labour Aristocracy,” p. 56. See H.F.Moorhouse, “The Marxist Theory of the Labour Aristocracy,” Social History 3 (January 1978): 61–82; H.F.Moorhouse, “The Significance of the Labour Aristocracy,” Social History 6 (May 1981): 229–33. Moorhouse, “Marxist Theory,” p. 73. Patrick Joyce, Work. Society, and Politics: The Culture of the Factory in Late Victorian England (New Brunswick, N.J.: Rutgers University Press, 1980). Ibid., p. xv. William H.Sewell, Work and Revolution in France: The Language of Labor from the Old Regime to 1848 (Cambridge: Cambridge University Press, 1980), pp. 154–61. Joyce, Work. Society, and Politics, pp. 118–119. Katznelson, City Trenches, chap. 3. R.Lawton, “The Population of Liverpool in the Mid-Nineteenth Century,” Transactions of the Historical Society of Lancashire and Cheshire 197 (1955): 93, 94. James H.Treble, “Liverpool Working Class Housing, 1801–1851,” in The History of Working Class Housing: A Symposium, ed. S.D.Chapman (Newton Abbot, United Kingdom: David & Charles, 1971), pp. 167–209. Other studies confirm the Liverpool pattern. Pritchard, for example, has treated the spatial transformation of Leicester, where in 1840 the integration between work and residence was very tight but by 1870 had been shattered and a modern form of spatial organization had developed. Vance has done the same for Birmingham, and Prest for Coventry. Olsen, Wohl, Sheppard, and a good many others have shown that even in London, where craft production survived as the dominant form through the nineteenth century, there was a “systematic sorting-out of London into single-purpose, homogeneous specialized neighborhoods,” an increased crowding of the working classes into wholly separate neighborhoods, and, overall, the large-scale separation of work and home (R.M.Pritchard, Housing and the Spatial Structure of the City [Cambridge: Cambridge


21.

22. 23. 24. 25. 26.

27.

28.

29.

30.

31.

University Press, 1976]; James Vance, Jr., “Housing the Worker: Determinative and Contingent Ties in Nineteenth Century Birmingham,” Economic Geography 43 [April 19667], pp. 95–127; John Prest, The Industrial Revolution in Coventry [Oxford: Oxford University Press, I960]; A.S.Wohl, “The Housing of the Working Classes in London,” in London 1808–1870: The Infernal Wen, ed. S.D.Chapman and Francis Sheppard [London: Seeker & Warburg, 1971]; Harold Pollins, “Transport Lines and Social Divisions,” in London: Aspects of Change, ed. Ruth Glass [London: Macgibbon & Kee, 1964], pp. 27–51; Donald Olsen, “Victorian London: Specialization, Segregation, and Privacy,” Victorian Studies 17 [March 1974]: 267). Sam Bass Warner, Jr., The Private City: Philadelphia in the Three Periods of Its Growth (Philadelphia: University of Pennsylvania Press, 1968); A.M.Warnes, “Early Separation of Homes from Workplaces and the Urban Structure of Chorley, 1780 to 1850,” Transactions of the Historical Society of Lancashire and Cheshire 122 (1970): 106. Ibid., p. 111. Ibid., p. 119. Ibid., pp. 120, 122. Ibid., p. 132. Ibid., p. 133. A parallel discussion can be found in T.C.Barker and J.R.Harris, A Merseyside Town in the Industrial Revolution St Helen’s 1750–1900 (Liverpool: Liverpool University Press, 1954). Sewell, Work and Revolution in France, p. 143. In this reminder of the danger of reading modern patterns into the past, he echoes geographer David Ward, who tells us that the social geography of English and American cities in the nineteenth century was one of comparatively weak patterns of spatial differentiation as compared with twentieth-century cities (David Ward, “Victorian Cities: How Modern?” Journal of Historical Geography 1[2] [1975]:135–51). For a discussion, see Vance, “Housing the Worker.” One of the few textbook histories to recognize the importance of spatial transformations is that of Eric Hopkins, A Social History of the English Working Classes (London: Edward Arnold, 1979). Its first two chapters are called “Life at Work” and “Life at Home.” John Foster, Class Struggle and the Industrial Revolution: Early Capitalism in Three English Towns (London: Weidenfeld & Nicholson, 1974). A suggestive discussion of these issues can be found in the review of the Foster book by Mary Anne Clawson in Contemporary Sociology 6 (May 1977): 289–94. This division between employers and workers in space was accompanied as well by another division: between workers themselves. With the separation of labor and housing markets as a concomitant of the division between work and community, workers with different capacities to consume goods and services (the hallmarks of class for Max Weber) came more and more to be segregated from each other in different areas of the city. The result, Foster argued, was the development of a geographically segmented working class that produced a tendency toward fragmentation and, in his terms, false consciousness. In this view, the ties of occupation, income, and space conjoined to make possible the development of a distinctive privileged stratum of the working class, an aristocracy of labor, whose propensity for alliances with the bourgeoisie undermined the “revolutionary consciousness” that had prevailed in the age of Chartism. The spatial dimensions of a labor aristocracy approach have also been developed suggestively by Geoffrey Crossick and Robert Gray. Crossick’s study of artisans segregated in space from industrial workers in mid-nineteenth-century Kentish London illuminates the


32.

33.

34. 35. 36. 37. 38. 39.

40. 41. 42.

43.

164

contradictory elements of the values and language structure of these “aristocrats” whose emphasis on sobriety, thrift, and respectability seemed to imitate dominant middle class Victorian values yet who also asserted an independent, even insular, working-class culture. Gray’s work likewise stresses the corporate class consciousness (here he borrows from Gramsci) of the spatially separate “superior” artisans, whose neighborhoods, housing conditions, and styles of life reflected “feelings of exclusiveness and superior social status” intertwined with “a strong sense of class pride. ‘There followed, in this view, a complicated pattern of implicit negotiated accommodations between this stratum of the working class and the middle classes that established the terms of incorporation of the working class into social and political life (Geoffrey Crossick, “The Labour Aristocracy and Its Values: A Study of Mid-Victorian Kentish London, Victorian Studies 20 [March 1976]: 301–28; Geoffrey Crossick, An Artisan Elite in Victorian Society [London: Croom Helm, 1978];R. Q.Gray, The Labour Aristocracy in Victorian Edinburgh [Oxford: Oxford University Press, 1976]; Gray, “Styles of Life,” p. 429). For my ability to make this comparative claim I am indebted to discussions with members of the Research Planning Group on Working Class Formation, funded by the Council for European Studies. For recent U.S. evidence, see Theodore Hershberg, Harold E.Cox, Dale B.Light, Jr., and Richard R.Greenfield, “The Journey to Work: An Empirical Investigation of Work, Residence, and Transportation: Philadelphia, 1850 and 1880,” in Philadelphia: Work Space, Family and Group Experience in the 19th Century, ed. Theodore Hershberg (New York: Oxford University Press, 1981, pp. 128–73. For comparisons see the discussions in Thompson, Making of the English Working Class, and in Amy Bridges, “A City in the Republic: New York and the Origins of Machine Politics” (Ph.d.diss., University of Chicago, 1980). Also see Ronald Aminzade, Class, Politics, and Early Industrial Capitalism: A Study of Mid-Nineteenth Century Toulouse, France (Albany, N.Y.: State University of New York Press, 1981), pp. 82, 94. Another complementary discussion of the importance of residence community ties is that of Michael Anderson, Family Structure in Nineteenth Century Lancashire (Cambridge: Cambridge University Press, 1971), p.103. J.J.Rowley, “Drink and the Public House in Nottingham, 1830–1860,” Transactions of the Thoroton Society of Nottinghamshire 79 (1975): 76. Ibid. P.H.J.H.Gosden, The Friendly Societies in England, 1815–1875 (Manchester: Manchester University Press, 1961). Henry Pelling, “Then and Now: Popular Attitudes since 1945,” in Politics and Society, ed. Pelling, p. 165. E.J.Hobsbawm, “From Social History to the History of Society,” Daedalus 100 (Winter 1971): 20–45. Charles Tilly, “Reflections on the History of European State-Making,” in The Transformation of National States in Western Europe, ed. C.Tilly (Princeton, N.J.: Princeton University Press, 1975), p. 47. J.P.Nettl, “The State as a Conceptual Variable,” World Politics 20 (July 1968): 577, 584. Samuel P.Huntington, Political Order In Changing Societies (New Haven, Conn.: Yale University Press, 1968). Barrington Moore, Jr., Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modem World (Boston: Beacon Press, 1966), p. 444 (cited in Tilly, “Reflections,” p. 632). The discussion in this paragraph was stimulated by my reading of Kenneth Dyson, The State


44. 45.

46. 47.

48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59.

60.

Tradition in Western Europe (New York: Oxford University Press, 1980), especially pp. 36– 44 and 186–202. Also see Alastair Reid, “Politics and Economics in the Formation of the British Working Class: A Response to H.F.Moorhouse,” Social History 3 (October 1978): 347–61. Thompson, Making of the English Working Class, p. 737. Dyson, State Tradition, pp. 249, 250. Recent work on the French working class (here I am thinking of Hanagan as well as Aminzade and Sewell) strongly suggests that it is not possible to think about the impact of the state on class formation and mobilization without considering political rights and repression over time (Michael P.Hanagan, The Logic of Solidarity: Artisans and Industrial Workers in Three French Towns, 1871–1914 [Urbana: University of Illinois Press, 1980]). Reinhard Bendix, Nation-Building and Citizenship: Studies of Our Changing Social Order (New York: Anchor Books, 1969). For discussions, see Oliver MacDonagh, Early Victorian Government, 1830–1870 (London: Weidenfeld & Nicholson, 1977); Harold Perkin, The Origins of Modern English Society (London: Routledge & Kegan Paul, 1969); Paul Richards and R.A.Slaney, “The Industrial Town and Early Victorian Social Policy,” Social History 4 (January 1979): 85–101; Robert D.Storch, “The Policeman as Domestic Missionary: Urban Discipline and Popular Culture in Northern England, 18501880,” Journal of Social History 9 (Summer 1976): 481–97; D.C.Moore, “Political Morality in Mid-Nineteenth England: Concepts, Norms and Violations,” Victorian Studies 13 (September 1969): 1–16; John Milton-Smith, “Earl Grey’s Cabinet and the Objects of Parliamentary Reform,” Historical Journal 15 (March 1972): 63– 82; Henry Parris, “The Nineteenth Century Revolution in Government: A Reappraisal Reappraised,” Historical Journal 3 (March, 1960): 1–37; and David Roberts, “Jeremy Bentham and the Victorian Administrative State,” Victorian Studies 2 (March 1959): 193– 210. Perkin, Origins of Modern English Society, pp. 123–24. Ibid., p. 310. Ibid., pp. 308–19. Cited in Patricia Hollis, ed., Class and Class Conflict in Nineteenth Century England, 1815– 1850 (London: Routledge & Kegan Paul, 1973), p. 80. Perkin, Origins of Modern English Society, p. 111. Thompson, Making of the English Working Class, pp. 212–13. J.F.C.Harrison, The Early Victorians, 1832–1851 (London, Weidenfeld & Nicholson, 1971), p. 156. Ibid., p. 157. Trygve Tholfsen, Working Class Radicalism in Mid- Victorian England (New York: Columbia University Press, 1977), p. 85. D.F.Macdonald, The State and the Trade Unions (London: Macmillan, 1960), p. 23. Gerald Abrahams, Trade Unions and the Law (London: Cassell, 1968), p. 28. Peter William Slosson, The Decline of the Chartist Movement (London: Frank Cass, 1967; originally 1916), p. 46. Also see David Jones, Chartism and the Chartists (London: Allen Lane, 1975); and Brian Brown, “Industrial Capitalism, Conflict, and Working Class Contention in Lancashire, 1842,” in Class Conflict and Collective Action, ed. Louise A Tilly and Charles Tilly (Beverly Hills, Calif.: Sage, 1981). Katznelson, City Trenches; also see Scan Wilentz, Chants Democratic: New York City and the Rise of the American Working Class, 1788–1850 (New York: Oxford University Press, 1984), p. 20.


166

61. For an important treatment of these issues, see Tholfsen, Working Class Radicalism. Also useful is H.I.Dutton and J.E.King, “The Limits of Paternalism: The Cotton Tyrants of North Lancashire, 1836 1854,” Social History 7 (January 1982): 111–41. For a fine-grained comparison of Birmingham and Sheffield that shows important variations in class formation within England, see Dennis Smith, Conflict and Compromise: Class Formation in English Society, 1830–1914 (London: Routledge & Kegan Paul, 1982).


167

47 Ethnic Competition and Modernization in Contemporary Africa* Robert H.Bates

* Source: Robert H.Bates, Comparative Political Studies, 1974, vol.6, pp. 457–484.

It has been argued that modernization promotes potentially disintegrative forces in developing areas, and in particular, often gives rise to powerful ethnic groupings (for example, see Geertz, 1963; Melson and Wolpe, 1970; Huntington, 1968). In this article, we elaborate this hypothesis in the context of the developing nations of black Africa. We note that important empirical studies have recently found that modernization and ethnic competition can and do covary in contemporary Africa (Morrison and Stevenson, 1972). And we attempt to explain why this should be so. The first part of the paper is definitional: we clarify our usage of the major terms of our argument. The second part is evidential: drawing on the social science literature from Africa, we attempt to demonstrate that the goods of modernity are desired but scarce; that modernization therefore creates new patterns of stratification; and that both these phenomena engender ethnic competition in Africa. We then move to the task of explanation. Given that modernization promotes competition, why is it that this competition assumes ethnic form? In seeking answers to this question, we devote the last two sections of the paper to the formation and persistence of ethnic groupings.

Definitions The major terms of our argument are modernity, modernization, ethnic competition, and ethnic group. In keeping with conventional usage, we define modernity operationally and call those nations more modern which attain higher levels of the following variables: education, per capita income, urbanization, political participation, industrial employment, and media participation. In practice, we will restrict our attention to the first three of these variables. We feel justified in using a single term—modernity—to refer to these distinct variables, for it has been repeatedly demonstrated that they are highly interrelated and that their interrelation derives from their tapping a single underlying dimension.1 “Modernization” we define as the process of attaining modernity. While modernity may be a highly interrelated and coherent set of variables—a system, as Lerner (1958) states—modernization is an incremental process. It takes the form of successive attempts to maximize the specific variables that comprise modernity. From the point of view of those who are modernizing, modernization is represented by the striving for more

Ethnic Competition and Modernization in Contemporary Africa

168

plentiful and better-paying jobs, the search for urban employment, the creation of opportunities for more education, and the quest for a host of other immediate benefits. By ethnic competition, we mean the striving by ethnic groups for valued goods which are scarce in comparison to the demand for them. Our definition of an ethnic group is, perforce, complex. Like all groups, ethnic groups are organized about a set of common activities, be they social (e.g., age-grade initiations), economic (e.g., cattle herding), or political (e.g., organizing behind a common candidate); they contain people who share a conviction that they have common interests and a common fate; and they propound a cultural symbolism expressing their cohesiveness. The primary factor that distinguishes ethnic groups is the cultural symbolism which they employ. The symbolism is characterized by one or more of the following: collective myths of origin; the assertion of ties of kinship or blood, be they real or putative; a mythology expressive of the cultural uniqueness or superiority of the group; and a conscious elaboration of language and heritage. In addition, ethnic groups differ from other groups in their composition: they include persons from every stage of life and every socioeconomic level.2 It is important to note that ethnic groups need not be tribes. By our usage, the term tribe denotes a group, generally rural, which is bound by traditional political structures to which people are linked by the mechanisms of traditional political obligation. Ethnic groups, we argue, are generally most active in the urban, modern sector and not in the rural, traditional countryside. Ethnic groups need not be based on traditional political institutions; rather, many are based upon newly created organizations, forged in the competitive environment of the modern era. And the ties that bind the members of ethnic groups are often material interests, and not traditional obligations. We concede that ethnic groups may expand into the traditional sector and gain the backing of tribes—this in fact can be a politically dangerous “stage” in their “evolution,” and we will examine some of the circumstances that can promote such urban-rural linkages. But, nonetheless, we wish to distinguish ethnic groups from tribal groups, and to consider the origins and dynamics of the former quite independently of what is known and asserted about tribal institutions and traditional political behavior in Africa.

The Argument and Evidence for its Plausibility Modernity is a cluster of desired goods. This is not to state that it is uncritically accepted, nor that African people do not decry the costs of modernization. The development of such philosophies as Negritude, Humanism, and the multitudinous versions of African Socialism by African intellectuals, and the spread of urban prophet churches and antiwitchcraft movements among some of the African masses, suggest the sensitivity of many to the costs of modernization.3 Nevertheless, it is obvious that the components of modernity are strongly desired. The Desire for Modernity The widespread demand of Africans for education has been amply documented; the promise of universal primary education is a political necessity in most African states


169

(Abernethy, 1969). Africans have demonstrated a willingness to alter life styles and social structures in pursuit of higher incomes. For example, Scudder and Colson (forthcoming) have documented the transition of one isolated and impoverished community, the Gwembe Tonga of Zambia, through three major agricultural systems; the agricultural transition of the Kipsigis and Kikuyu are well known (for example, Barwell, 1956); so too is the wholesale adoption of new cash crops by the Baganda. Studies of the flows of labor migrants in Africa also document our point. The overwhelming consensus of students of this phenomenon is that the cause of these migrations is the desire for employment and the income it makes possible (Gugler, 1968). The desire for modernity is also evidenced in the rates of urbanization in Africa. The rate of growth of urban populations is truly impressive. According to available figures, the urban population of the black states of Africa increased, on average, by 178% between 1955 and 1965, while the natural rate of increase of the population as a whole averaged 2.62% per annum.4 While most studies report that the influx of urban migrants represents a search for higher incomes, recently collected materials lend greater credence to a popular hypothesis that has often been denigrated by scholars of the phenomenon: that urbanization also represents a quest for the modern style of living to be found in the urban sector.5 Competition for Modernity The desire for the components of modernity in Africa is thus fully as incontrovertible as are the dissatisfactions to which modernity gives rise. Modernity is valued; equally as important, the goods it represents are scarce in proportion to the demand for them. The inevitable result is that people compete. This competition is best illustrated in the competition for income and for several of the resources which create wealth: land, markets, and jobs. Land In the agricultural societies of Africa, particularly where the population is dense, the penetration of a money economy gives rise to an intense competition for land. As Colson (1971:194) states: “By themselves such changes had an impact on local systems of land rights as men began to evaluate the land they used in new ways. They also led to an increasing number of legal battles over land; for men were encouraged to establish longterm rights in particular holdings either for immediate use or for subsequent gain.” While much of the competition for land is intraethnic, much of it is interethnic as well. Hill (1963: appdx. 3) has published accounts of several major interethnic land disputes in Ghana (see also Twaddle, 1969; Chubb, 1961:42ff.). The dispute between the Kikuyu and Masai over control of the former white highlands has created a major cleavage in the political life of Kenya (for example, see Burke, 1966). And a major source of urban conflict is the tension between those indigenous ethnic groups who have alienated their lands and those immigrant groups who have benefited from the occupation of urban real estate.6


170

Markets The competition for control of markets is equally intense. One of the best analyses of this phenomenon is Cohen (1969), who documents the intensive rivalry between the Yoruba and Hausa for control over trading routes to the interior along the southern coast of Nigeria. Lloyd (1956b) documents the rivalry between Itsekiri and Urhobo for marketing facilities in Warri Such conflicts characterize Eastern Africa as well (for example, Lonsdale, 1970; Ghai, 1970). Jobs Equally as pervasive is the competition for jobs. Parkin (1969) discusses the rivalry between Luo and Baganda for employment in the industrial and service sectors of Kampala; Grillo (1969), in his analysis of the East African Railways, notes a similar rivalry between Luo and Abaluhya. Competition for urban employment has been noted between the Bamileké and Douala in the Cameroons; between Nyanja and Bemba speakers on the copper belt of Northern Rhodesia; between indigenous Africans and strangers in Abidjan; and between Kasai Baluba and Bena Lulua in Kasai province in the Congo (Johnson, 1970a; Carter, 1960: ch. 10 and throughout; Anderson et al., 1967:120– 142). Exacerbating these tensions has been the expansion of the production of educated employables at a rate in excess of the expansion of job opportunities, a phenomenon that has been studied in Ghana, Nigeria, and elsewhere (Foster, 1965; Callaway, 1963). Modernity and Stratification Not only does modernization thus create competition; but also, because the elements of modernity are valued and scarce, they form the basis of a new stratification system in Africa. This is not to state that traditional criteria of social ranking are totally relinquished; indeed, all evidence is to the contrary, and we shall later argue that the interplay between the two stratification systems is crucially significant in explaining the emergence of ethnic groups in the modern era. Nonetheless, those who are modern can successfully claim higher social rank in contemporary African society than those who lack the attributes of modernity. The creation of a new stratification system is documented in studies of occupational prestige in Africa. These studies find that modern roles, such as those of the teacher or clerk, are given high prestige and that in fact they are generally ranked higher than the roles of traditional societies, such as those of craftsman or hunter. The results of these studies go beyond suggesting that modern occupations are prestigeful, however, to emphasizing that modern conceptions of stratification are being utilized by African peoples. Thus, studies report a close correspondence between the ranking of occupations by Africans in the Congo and those produced in the more developed countries (see Xydias, 1956; Mitchell and Epstein, 1959; Mitchell and Irvine, 1965). Hicks (1966), in a study of occupational prestige in Zambia, finds the criteria for the rankings to be similar to those reported in industrialized societies. The prestige of an occupation can largely be


171

accounted for, he indicates, by the degree of responsibility, service value, income, education and the nature of the working conditions associated with it. The numerous studies of elite formation in Africa also suggest that those who possess the attributes of modernity can successfully lay claim to high status in many indigenous societies. Thus, Lloyd (1956a, 1955, 1953), in a series of articles, notes the rise of wealthy traders and educated clerical workers as a new elite in Yoruba society; Austin (1964) discusses the same phenomenon among the Ashanti and, like Lloyd, documents the conflicts between the new elite and the traditional ruling classes.7 The same pattern has been reported in East Africa. Studies of the Gisu, Chagga, Luo, and Abaluhya describe how the spread of education and cash cropping generated new groups of literates and wealthy traders and how these new segments of the population successfully lay claim to elite status (La Fontaine, 1969; Twaddle, 1969; Stahl, 1969; and Lonsdale, 1970). Stratification and Competition Crucial to the emergence of ethnic competition is that societies as well as individuals tend to be evaluated along the dimension of modernity. Those groups which are more wealthy, better educated, and more urbanized tend to be envied, resented, and sometimes feared by others; and the basis for these sentiments is the recognition of their superior position in the new system of stratification.8 Thus, for example, in Calabar, the indigenous Efik took readily to education, while the immigrant Ibo lacked both education and the wealth which would follow; the result was tension and hostility, sentiments which were exacerbated by the Ibo’s attempts to close the gap (Morrill, 1963). This case finds its parallel in the famed rivalry between the Ibo and the Yoruba. Exposed to education at an earlier date, prosperous because of cocoa, and indigenous to the urban areas which became the primary locus of administrative and commercial development, the Yoruba claimed top rank in the structure of modern society in Nigeria. The envy of the Ibo, and the tensions which resulted from their attempts to catch up with the Yoruba. are too well known to warrant detailed discussion here. So too with the northern peoples of Nigeria; their fear of the superiority of the southern peoples in the modern sector led them to an explicit policy of “northernization” whereby they gave privileged access to educational and employment opportunities to residents of the north. Their unequal status in modern society led to another point of conflict with the southerners, this time over the date of self-government in Nigeria. As related for the Birom of northern Nigeria (Baker, 1956:94): “Birom leaders expressed unease at the thought of rapid achievement of self-government—in fact at the idea of selfgovernment before the Birom have produced enough professional men and traders and artisans to be able to claim all the occupations of control which exist in their Division.” As a result, the Birom aligned with the Northern Peoples Congress instead of with one of the several Southern parties which were competing for their allegiance, for the Congress favored a later date for self-government. This pattern is also found in East Africa. Thus, for example, in Kenya, the less modernized pastoralists feared the perpetuation of the disparity between themselves and the more educated, urbanized, and wealthy agriculturalists which would result were Kenya to become independent under the political control of the agriculturalist tribes. As a


172

result, the pastoralists less fervently pressed for self-government in Kenya than did the agriculturalists; and when they saw that their efforts were to fail, they sought to fragment power through a federal constitition. This conflict was perhaps the major basis for the political competition between KANU and KADU in Kenya (see Burke, 1966; Rothchild, 1969). A similar situation existed in Uganda, where it has been argued (Bennett, 1969) that the UPC represented a coalition of less modernized ethnic groups in common opposition to the prospects of a Baganda hegemony. As Zolberg (1964:5) succinctly states, in the case of the Ivory Coast: “Many [of the changes introduced under colonial rule] have reinforced old differentiations between tribes by adding to them new ones based on modern attributes, such as wealth and education.” And as one Nigerian commentator states (Akiwowo, 1964:162): Tribalism, then, exists when members of tribal groups must compete as groups for places in the class, status and power systems of the new nation. In a manner of speaking…it is a form of social indecision regarding the strategy of equitable distribution of advantages available to people in the new African nations.

Basis for the Formation of Ethnic Groups The basic question that arises from this discussion is: Why should the competition for the components of modernity and for status positions as defined by modernity involve ethnic groups at all? At least three answers can be given to this question. The first is that both the distribution of modernity and ethnic groups tends to be governed by the factor of space: Where modernization takes place often largely determines who gets modernized. The second is that administrative and ethnic areas often coincide. And the third is that it is often useful for those engaged in the competition for modernity to generate and mobilize the support of ethnic groupings. The Factor of Space It is the geographers who most forcefully portray the spatial patterns of modernization. Originating in “nodes” or “central places,” modernity then spreads or “diffuses” into the more remote regions of the territory, they report. They also demonstrate that the level of modernity slopes downward with distance, with the central places being the most modernized, the proximate areas being the next most developed, and the hinterlands lagging behind (for example, Gould, 1968; Soja, 1968). While there is considerable debate over whether territoriality is a required component of the definition of an ethnic group, there is no denying that the members of an ethnic group tend to cluster in space, nor can it be questioned but that colonial policy made every attempt to assign ethnic groups to stable and rigidly defined areas. The result of this correspondence in spatial orderings is that those ethnic groups which are most proximate to the locus of the impact of modernity tend to be the most modernized; and thus the competition for the benefits of modernity and for status positions in the modern sector can become organized on ethnic lines .


173

The evidence for this assertion is persuasive. Soja (1968), in his analysis of modernization in Kenya, finds that the Kikuyu, being proximate to Nairobi and the highlands, are the most urbanized and educated, and among the more wealthy, of the ethnic groups in Kenya (see also Rothchild, 1969). Similarly, Coleman (1958) and Abernethy (1969) argue that the initial advantage of the Yoruba in Nigeria derived from their proximity to Lagos and from the early establishment of missions in Lagos and Abeokuta. The Ibo, being more remote from these areas, were initially less exposed to the centers of modernization; and being less proximate to the locus of mission activity, they lagged badly in the attainment of education and well-paying jobs. The preeminence of the Baganda, as well as the tensions which have resulted, have also been explained in terms of their proximity to the administrative capital and largest town in Uganda. As one analyst reports (Kasfir, 1972:76): in Uganda (with the possible exception of the southeastern area) geographical distance from the capital city is sufficient to provide a rough indicator of the degree of modernity. Most Ugandans are fully aware that the Baganda profited more than others from their close proximity to the administrative center of the country. So pervasive a phenomenon does this appear that spatial proximity is sometimes offered as an alternative to classic notions of “cultural receptivity” in explaining differing rates of change. Thus, Kasfir (1972:76), discussing Apter’s structural-cultural theory of modernization in Buganda, notes that given the proximity of the Baganda to Kampala, the “argument…cannot be proved or disproved.” And Gugler (1968:465), in discussing the general resistance of pastoralists to the forces of change, comments that “the underlying more general factor [is] probably that many of these are difficult of access to schools and administration alike.” Space, Administration, and the Incentive to Organize Local administration serves as one of the primary agents of modernization in Africa. And the colonial powers, by delineating administrative boundaries along “tribal” lines, made it in the interests of their subjects to organize ethnic groupings so as to gain control over the administrative mechanisms which themselves controlled the modernization process. This assertion is best demonstrated in the studies of one of the primary sources of income in Africa—land. Colson (1971) notes that colonial policy produced two contradictory developments in land law. On the one hand, the growth of the cash economy furnished an incentive for individual ownership; on the other, the dominant mythology of the colonial administration, that land was “communally owned,” restricted permanent rights to land to the members of the local ethnic group. A clear implication of Colson’s analysis is that as the benefits of land ownership increased, as they did with the spread of cash cropping, so did the importance of retaining and affirming membership in ethnic groups. The political consequences of this rapidly became evident in the conduct of the local councils in which jurisdiction over land rights had in part been vested. The local councils began to function as ethnic organizations, legislating so as to protect the


174

benefits brought to the local population by the modernization process. Thus, Coleman (1958:59) reports that with the appreciation of land values, councils in Owerri, Ibadan, Benin, Idoma, Tiv, and Igala in Nigeria restricted access to land to the local ethnic group and divested strangers of the rights of permanent tenure. In this way, the material benefits to be derived from the land were channeled to the local residents and indigenous ethnic group (see also Chubb, 1961:25). The power of the local administration over economic resources extended beyond the control of land tenure to such other matters as access to markets and market stalls, the regulation of crop production and animal husbandry, the construction of roads for the export of produce, and the like. At the behest of those who had the greatest stake in the modern economy, often organized in “improvement unions,” many councils acted so as to bias the distribution of these resources for the benefit of the local population and away from immigrant strangers. Thus, La Fontaine (1969) reports that local leaders sought to distribute roads and payments to coffee growers around Mbale so as to benefit Gisu cash croppers exclusively. Lonsdale (1970) documents the attempts by the Kavirondo local councils to restrict access to markets to Abaluhya and Luo cash croppers. And Lloyd (1956b: 86), in his discussion of Urhobo and Itsekiri rivalry in municipal elections in Warri, notes that “it was said during the 1955 election campaigns that whichever tribe won the election would restrict the lease of stalls to its own members and thus give them a monopoly of the trade in the town.” Given that power over the distribution of many of the benefits of modernity is vested in the local administration, and given the correspondence between administrative and ethnic boundaries, it was therefore natural that persons would create politically cohesive ethnic groupings and utilize these to restrict the degree to which the administration could compel the sharing of the benefits of modernity with others. The demand of ethnic groups for their own districts and councils represents the logical culmination of this process, for by securing this demand they can more perfectly exclude others from the benefits of modernity and thereby reserve to themselves a larger portion. The Behavior of the Moderns A third major class of reasons for the formation of ethnic groups is that in the competition for the benefits of modernity, it has been in the interests of the most modern elements to sponsor the growth of “traditional” consciousness in Africa. Before explaining why this is so, it is instructive to indicate the extent of the evidence for the major role that moderns have played in organizing “traditional” groupings. The “educateds” often were the founders of ethnic unions. Thus, Lonsdale (1970:628) speaks of “the Christian establishment” of mission-trained literates who helped to form the Kavirondo Taxpayers Welfare Association. Twaddle (1969:197) writes of the “‘new men’ created by missionary education” who helped to form the Young Bagwere Association. And the role of the former mission students in organizing the independent schools among the Kikuyu and promoting ethnic consciousness in that tribe has been discussed by many authors (for example, Rosberg and Nottingham, 1966:105–135). In terms of income, it is often those who are better off by dint of their occupations in the modern sector—the clerks, cash croppers, and traders—who form ethnic unions. Thus,


175

for example, Ottenberg (1955:4) notes that it was those who “work as clerks for the local British Administration, who teach in local schools, who work for traders, or who are traders themselves” who founded ethnic unions among the Afikpo Ibo. Sklar (1963:72) notes that both the Egbe Omo Oduduwa and Ibo State Union “were created by representatives of the new and rising class—lawyers, doctors, businessmen, civil servants.” And Bennett (1969:80) notes that the Bahaya Union was led largely by relatively prosperous cash croppers and members of cooperative societies whose economic interests were threatened by the government’s policies toward coffee cultivation and land management. The role of urban dwellers in the formation of ethnic unions stands out most clearly in the literature from West Africa. There, ethnic unions were most often formed in urban centers and only later exported to the rural areas whose names they often bore. As stated by Offodile (1947:937): “It is significant to observe that almost all the tribal unions now existing in Nigeria were found, not in the very towns of the tribes represented, but outside their own villages, and sometimes outside their own tribal territories.” Thus it was in Lagos where the Ibo came into competition with the Yoruba and where Ibo tribal consciousness was formed. It was in Leopoldville that Abako was founded; only later was it exported to the majority of the Bakongo in the surrounding rural territories (La Fontaine, 1970:192). And even in the case of such minor tribal unions as the Afikpo Town Welfare Association, the origin of the organization lay in the city: “An educated Afikpo man working at a trade-union post in Aba, who had traveled widely in the course of his work…realized the need for a protective union to aid Afikpo people…. At his own expense he had 400 membership cards printed to organize Afikpo people” (Ottenberg, 1955:15–16). In explaining the behavior of the moderns, we can note numerous motivations for the formation of these unions, but one stands out above all others: the perception by the moderns that they must organize collective support to advance their position in the competition for the benefits of modernity. To this we now turn. Social and Economic Competition and the Formation of Groups We have noted that there are good reasons for one ethnic group to be more advanced than another; in part, the spatial diffusion of modernization makes this inevitable. We have also noted the forces that can account for an ethnic group consolidating itself into a position of advantage. For one, because modernity creates status differentiation, the members of an advantaged ethnic group are motivated to defend their leading position; they devise methods for retaining their privileged positions, such as by biasing local council legislation in the ways we have described. Secondly, because modernity is desired, the less favored members of an ethnic group place immense pressure on their more advantaged brothers to share the benefits derived from their advanced positions. Thus, family loyalties are activated to secure jobs;9 the income of the more prosperous is claimed by kin, often to meet school fees that will in turn secure future prosperity;10 and urban dwellers find their households being used by country folk in search of urban employment (see, for example, Lloyd, 1967; Caldwell, 1969:130; Comhaire, 1965; Sutherland, 1956). Under the pressure of the less advantaged, a sense of‘ obligation


176

resembling that usually extended to immediate kin is thus broadened to include fellow village dwellers and even persons from other villages and districts; and the language of relationships, such as the use of putative kin terms, is broadened to suggest this expansion. The result of these pressures is that the more advantaged members of the group are forced to draw into their sphere others of their kind. And the social-climbing less advantaged generate a mythology of consanguity in search of modern benefits. The initially advantaged group thus consolidates itself in the modern sector and comes to view itself as an ethnic grouping in the process. The result of this process of consolidation is to create among the members of other groups a sense of threat and disadvantage. In the competition for jobs, it is the more modern elements of these groups who most directly experience this threat and perceive it in ethnic terms. They come to understand that they are placed at a disadvantage by their inability to activate the sense of ethnic obligation so as to gain access to the modern sector. Moreover, they perceive that their individual progress is closely determined by the collective standing of their group; they therefore initiate programs of collective advancement in response. The creation of ethnic support by the competitors for jobs has been noted by Grille (1969) in his analysis of the Railway African Union in Uganda. High office in the union often led to promotion to more advanced jobs in the railway company, and so was much desired by railway employees. In order to enhance their mobility prospects by gaining union office, contenders for advancement would sometimes make appeals to tribal loyalty. As stated by Grillo (1969:318), “although those seeking office ... may have little sense of tribal interest, tribalism may be one of the weapons used in the struggle.” A similar pattern has been noted in Kenya, where accusations are made regarding the “Kikuyuization” of the government services and ethnic pressures mounted by the leaders of the less advanced groups in protest over job discrimination (Rothchild, 1969). However, the most striking illustration of the creation of ethnic action is to be found in West Africa. Perceiving that their individual fate in the struggle for modernity was tied to the collective standing of people from their own areas, the most modern members of the less modernized groups organized large-scale programs of advancement among their people. Abernethy (1969:107–108) furnished the best discussion of the phenomenon: The struggle for employment was bound to produce frustration, and those not chosen for the best jobs found it easy to blame their plight on the advantages possessed by members of other groups. Of course, different groups clearly did have differential access to education, which in turn was the key to job mobility. … What was the best course of action open to the urban migrant who was acutely concerned lest his ethnic group fall behind others in the struggle…? Certainly the rural masses had to be informed of the problem. If the masses were not aware of their ethnicity, then they would have to learn who they really were through the efforts of “ethnic missionaries” returning to the homeland. These “missionaries” would also have to outline a strategy by which the ethnic group, once fully conscious of its unity and its potential, could compete with its rivals. Clearly the competition required enrolling more children in school,


177

particularly at the secondary level for the graduates of a good local secondary school would be assured of rapid…mobility within modern society. In this manner, Abernethy accounts for the formation of ethnic unions among the Ibibio, Ibo, and Urhobo in Nigeria. Political Competition A similar pattern obtains in politics. In the political arena, it is not just power that is at stake, but also the benefit which power can bring: control over the distribution of modernity itself. There can be no doubt but that electoral competition arouses ethnic conflict. The tensions arising from the 1964 elections in Nigeria were one of the precipitates of civil war in that country. The primary instruments of the virulent ethnic conflict in the Congo were the numerous political parties formed by politicians to contest the 1960 elections (Young, 1965). And it was elections, or the anticipation of them, that precipitated ethnic conflict in Zanzibar, Rwanda, and Ghana (see the discussions in Austin, 1964; Lofchie, 1970; Lemarchand, 1970; Maquet, 1970). Perhaps the main reason for these conflicts is that in the competition for power, ethnic appeals are useful to politicians. Given that most constituencies tend to be dominated by the members of one ethnic group—a result of the politics of apportionment and delimitation-an ethnic appeal is an attractive and efficacious weapon in the competition for office. Moreover, because ethnic groups contain persons of all occupations, socioeconomic backgrounds, life styles, and positions in the life cycle, the appeal of common ethnicity can generate unified support where other issues would be divisive. As a result, in the competition for power, and for the benefits of modernity and the prestige it confers, politicians will stimulate the formation of competitively aligned ethnic groups. As stated by Sklar (1967b: 6): “It is less frequently recognized that tribal movements may be created and instigated to action by the new men of power in furtherance of their own special interests.”11 Naturally, this is not to state that the politicians are alone to blame for the rise of ethnic conflict. Indeed, while they do instigate ethnic conflicts, they often behave like captives of the forces which they helped to create. This leads us to the last question with which we plan to deal in this paper: Having accounted for the formation of ethnic groups, how can we explain their persistence?

The Persistence of Ethnic Groups Ethnic groups persist largely because of their capacity to extract goods and services from the modern sector and thereby satisfy the demands of their members for the components of modernity. Insofar as they provide these benefits to their members, they are able to gain their support and achieve their loyalty. The capacity of ethnic groups to extract goods and services from the modern sector is best demonstrated in their relationship with those who have achieved positions of


178

prominence in that sector. Ethnic groups exert powerful social pressures upon the modern elite in order to satisfy the demands of their members. Perhaps the most persuasive evidence for these assertions is the reaction of the modern elite itself. Its members experience their positions not only as privileged but also as onerous; they feel that they are at the center of tremendous social pressures. As Uchendu (1965:9) states: “My town demanded leadership from me. But this leadership is a trying as well as a thankless experience. My town has a passionate desire ‘to get up.’ “Chona (1968), currently VicePresident of the Republic of Zambia, argues the same point: modern leaders are subject to concerted pressures and forced to act as spokesmen for ethnic interests. Rather than blaming members of the national elite for instigating “tribalism,” he states, the citizens of Zambia should blame the “local leaders in the villages and towns” who “travel…to Lusaka to meet leaders” in order to urge them to serve parochial interests (see also Chona, 1970). Chona concludes, “unless the local leaders in the villages and towns stop being competitive against other groups and begin to regard top leaders as national leaders we shall not find a lasting solution to [the problem of tribalism].” The demands upon the modern elite are predictable. Characteristically, they contain demands for material resources: financial contributions from the moderns for the construction of new facilities and for the creation of educational funds. Some groups even levy taxes upon their more prosperous urban members.12 The members of the ethnic groups also seek advice on how to better their incomes. In response, ethnic groups retain members of the modern elite as “economic advisers …to instruct their members in the ups and downs of trade; they have also advised in favor of joint stock enterprise. Different fields of trade are surveyed and analyzed for the benefit of the members” (Offodile, 1947, 941). The demands of ethnic groups are also for service: the use of the skills of the moderns, be they technical, educational, or political, on behalf of “their people.” The capacity of ethnic groups to extract goods and services from the modern elite derives from several sources. They control the allocation of strong inducements. For example, because elite skills are a desired commodity, ethnic groups are able to win the use of these skills by making their acquisition contingent upon ethnic service. Thus, Lloyd (1956b: 86) reports for the Itsekiri that “several lawyer-politicians who were sent to England with community funds are expected on their return to repay the debt either in cash…or by winning tangible benefits for their people.” Other inducements include prestige: symbols of status are conferred or withdrawn by ethnic groups in recognition of services performed for the group. Thus, Plotnicov (1970:289) documents the conferral of status by ethnic groups in Jos; members of the modern elite, he writes, have the skills for dealing with government and the wider community. Their knowledge in legal and economic matters, secretarial and bookkeeping procedures, and their general sophistication in modern and urban affairs is indispensable. Their value to the ethnic group…[is] recognized through the granting of high…offices…and sometimes…titles as well, which further reinforce the modern elite’s powers. Ethnic groups also possess strong sanctions, most notably the capacity to withdraw elite


179

status. This is vividly revealed in the political sector, where, for example, in recent elections the Luo, Sukuma, and Lozi turned out of office several of their most renowned political leaders, some of cabinet rank. The reported reason for the imposition of these sanctions was the elite’s failure to serve local interests (see Scott and Molteno, 1969; Africa Report, 1970, 1965; Morgenthau, 1965; Africa Diary, 1970). There is another reason for the ability of ethnic groups to extract goods and services from the incumbents of the modern sector. For many moderns, what is prestigeful is still defined in terms of traditional criteria. For many, modernity becomes a resource which they utilize to attain prestige within the traditional sector. Evidence for the continued presence of traditional notions of stratification is contained in the very studies which emphasize the preeminence of modern stratification systems. Thus, for example, Hicks (1967:213) finds that the standard deviation of the rankings of modern occupations by Africans is higher than that of the rankings by Europeans; he attributes this in part to the use by Africans of two sets of stratification criteria, one traditional and the other modern. Similarly, Foster found that Ghanaians give higher ranks to traditional political chiefs and councilors than they give to many modern occupations “despite the fact that often holders of these [offices] had little education and received little…pay. Clearly the respondents in Foster’s study were using at least two dimensions to rate the full list of occupations: one being the western dimension and the other a traditional dimension” (Hicks, 1967:219, discussing Foster, 1965). The acceptance of traditional stratification patterns leads many moderns to convert their success in the modern sector into prestige in the traditional order by utilizing their wealth to obtain prestigeful positions in their ethnic groupings. Many studies cite the purchasing of traditional titles by the successful entrants into the modern sector (Ardener, 1953).13 Others note the use of wealth to purchase traditional offices.14 Still others record the use of income derived in the modern sector to practice clientage and to finance largescale traditional ceremonies, both of which are conventional means of enhancing one’s standing in the traditional order. As Balandier (1965:392–393) states:15 [a wealthy person] can…make “sociological investments”; in this case, he uses new economic conditions to achieve or to reinforce a traditional type of prominence. The size of his “clientele” and the extension of his generosity will reveal his degree of success; his profit will be expressed in prestige and authority…the economic “game” is still only a method to achieve goals determined by the old social and cultural system. The ethnic groups are thus able to extract goods and services from their members in the modern sector. We have already indicated how they use this capacity to satisfy the demands of their members for the components of modernity: income, education, and urban employment. The result of this process is the generation of political support. Thus, as Sklar (1967a: 528) notes, the tribally dominated parties of Nigeria seized power at the regional level where the regional party leaders operated highly effective systems of patronage, dispensing jobs, contracts, commercial loans,…and scholarships. Young people


180

in all parts of the country were pressured in various ways to support the regional government parties…. Opinion follows interest, and many young adults furthered their careers by adopting regionalist principles and tribalist ideologies. Related to our point is a last observation: that ethnic groups represent a successful mechanism for generating expanding benefits in the modern era. We have repeatedly alluded to the various components of modernity—income, education, and urbanization— and noted how ethnic groups are created as part of the process of attaining them. What we wish to stress here is that many ethnic groups, and in particular those that span the rural and urban sectors, combine these components into a mutually reinforcing system—one that produces expansion and growth and therefore increasing payoffs. We represent this system in Figure 1. We have already noted the major linkages in this system. We have also noted the manner in which ethnic groups furnish these linkages, by organizing crop marketing, education, and job placement, and by channeling the benefits of urban employment into the rural sector in the form of economic advice, remittances, and educational programs. Here we need but stress the phenomenon of “positive feedback” that perpetuates this system. To emphasize but one kind of evidence for this feedback, we mention the studies by Caldwell. He notes that those rural families which are wealthier tend to generate more educateds, that the educateds are more prone to migrate, and that they therefore obtain higher incomes. He also notes that the families which contribute toward education fees get back more money from the educateds than they contributed in the first place (Caldwell, 1969:106) As a result, the income of the rural families expands even more and the cycle continues. As we have noted, in some cases, ethnic organizations establish these feedbacks, and in other cases they lay hold of them and institutionalize them. In both cases, they have helped to organize the components of modernity into the kind of system described by Lerner (1958) and Deutsch (1961) That ethnic groups have devised organizations which enable them to cope with, and indeed to utilize and benefit from, the process of modernization helps to explain why they have prospered in the modern era. For insofar as they serve the demonstratedly strong demands of the African peoples for modernization, they can win their support and thereby gain a heightened capacity to persist. Figure 1


181

Conclusion In this essay, we have attempted to lay the foundation for an explanation of ethnic group formation and ethnic competition that is based on the modernization of African societies. No doubt there are other reasons for the existence and behavior of ethnic groups; no doubt, too, modernization can take place without ethnic groups forming and competing. But the two phenomena do intersect, both empirically and intellectually, and they do affect and shape each other. In this essay we have examined this point of intersection and tried to explain how the process of modernization can promote ethnic group formation. In doing this, we have suggested a way of looking at ethnic behavior that emphasizes that it is a dynamic and rational behavior; and that it represents an attempt to deal with, organize, and benefit from the modernization of societies. We realize that our argument is but the beginning of an understanding of the phenomenon, and not an explanation of it. But we feel that we have made the phenomenon amenable to theorizing, and amenable to analysis in terms of theories applicable in Africa, the United States, and elsewhere. If we have indeed succeeded in that, then we feel that this paper will have served the purpose for which we intended it.

Notes Author’s note: This paper is a revised and shortened version of a paper presented to the Program of Eastern African Studies of Syracuse University and is published with their permission. The assistance of the Program is gratefully acknowledged. 1. The seminal works in this vein are Deutsch (1961) and Lerner (1958). For data from Africa,


2.

3.

4.

5. 6.

7.

8.

9.

182

Soja (1968) finds that district level measures in Kenya of education and urbanization, and various indicators of income, load heavily on a single major factor. For discussions of the definitional problems of the term, see Melson and Wolpe (1970), van den Berghe (1965), Cohen (1969), and Mercier (1965b). To be noted is that to affirm the reality of ethnic groups is not to ignore their internal devisiveness. Conflicts between major segments and areal groups, between commoners and persons of royal blood, and between clans and villages—all these cleavages do exist. Nonetheless, internal conflicts do not necessarily weaken the capacity of the larger ethnic groups to mobilize their members for collective purposes; the conflict-laden, internally divided Ibo are a case in point. The fact of internal division and conflict should therefore not be taken as evidence of the absence of effective ethnic collectivities nor discredit the validity of our enterprise. One of the major social costs of modernization is the development of new stratification patterns, and thus inequality; another is the generation of new individual opportunities, and thus the promotion of individual self-interest at the expense of traditional social obligations. At the elite level, African Socialism, and its other variants, represent an attempt to speak to these problems (see Friedland and Rosberg, 1964; Nyerere, 1968; and Kaunda, 1969). At the mass level, witchcraft and anti-witchcraft movements represent parallel attempts to deal with these problems. For analyses of the relationship between social change and attempts to control witchcraft, see the contributions in Middleton and Winter (1963). See also the case of the Bakweri Cooperative Union of Farmers, in which the prosperous cash-crop-growing peasantry diverted their earnings into paying for the services of anti-witchcraft experts to defend against the jealousy of their less successful brethren (Ardener, 1958). Figures from the data collected by the Council for Inter-Societal Studies Northwestern University. The figures represent the percentage increase in the population of cities of 20,000 or more residents. As with all quantitative data on Africa, these must be treated with caution. For example, it is probable that in the case of Malawi, the population of a single town passed beyond the 20,000 mark during this period, with the result that all of a sudden over 20,000 people were added to the urban population; given the small number of people previously counted as urban, the effect of a small change in the population of this town could produce a massive change in the overall urbanization figure. Nonetheless, as with all statistics, while specific figures can be unreliable, the general tendency which they indicate can be reliable, and the general tendency toward rapid urbanization in Africa is indisputably strong (see Fischer et al., 1969; Paden, 1968). The economic theory has been most vigorously counterpoised against the “bright lights” theory by Gulliver (1965); but see Caldwell (1969). For other examples, see Lloyd (1956b) and the accounts of the Gisu’s conflict with the Ganda and Bagwere for control of Mbale contained in La Fontaine (1969) and Twaddle (1969). See also the discussion of Onitsha and Ibadan politics contained in Sklar (1963:15 ff. and 285 ff.). Caldwell (1969:143), in his study of returned urban migrants in Ghana, and Post (1963:376– 436), in his study of elections in the Eastern Region of Nigeria, clearly suggest the prestige given the urbanized and educated by the less modernized rural populations (see also Hershfield, 1969). For general discussions of elite formation, see Lloyd (1966), MortonWilliams (1956), Wilson (1966), Mercier (1965a), and Okigbo (1965). Our general point here is made also by Sklar (1967b; Sklar and Whitaker, 1964) in his argument that ethnic conflict is conducted in terms of gesellschaft principles at the elite level and in terms of gemeinschaft principles at the level of the masses. See, for example, the analysis of departmental employment practices in the East African


10.

11.

12.

13.

14.

15.

183

Railways depot in Kampala contained in Grillo (1969) and the description of the procedures for getting jobs contained in Okonjo (1967). See, for example, Comhaire (1965); Garlick (1958); and Caldwell (1968:106). See also the numerous discussions (Mercier, 1966; Tuden and Plotnicov, 1970) of “the failure of class formation” in Africa which underscore the extent of the use of kinship networks to extract the benefits accruing to the more modern elements of society (see also the references in note 7). As noted in one study in Uganda (Kasfir, 1972:63), the politicians strive to create ethnic unity where none existed before: “it is not always clear whether an ethnic group is united or whether its leaders are simply putting forward claims in the hopes of galvanizing potential members…. In a survey among rural Iteso carried out in 1956 asking who was the leader of the Iteso people, over 50% of the respondents stated there was none. The political elite of Teso, however, called for union of all Iteso in Kenya and Uganda, purging the language of Luganda words, and giving traditional names to thirty-five towns.” For example, see Offodile (1947) for a description of the levying of educational funds by the Ibo State Union, and Hershfield (1969) for an analysis of the contribution of urban dwellers to the construction of civic facilities in rural villages. See also the discussion of the impact on rural development of financial contributions from the urban dwellers in Caldwell (1969:161 ff.). See also the comments on title-taking in the Awzaw order made in Basden (1966:265). Plotnicov (1969:272, in commenting on an article by A.L.Epstein) noted, “an efflorescence of traditional title-taking among urban Ibo” in Jos during the period prior to the civil war. Sklar (1963:107–110) notes the tendency toward title-taking among the Yoruba and relates it to the elite’s desire to recruit ethnic support. For a commentary on this practice which is very close to the one here, note Uchendu (1965:92–93). Lloyd (1956a) notes the renewed attractiveness of traditional offices in the Western Region of Nigeria. Whitaker (1967) notes the tendency of those with high positions in modern society to utilize their resources to achieve higher offices in traditional political structures. Plotnicov (1970:280) notes the same pattern in Jos. For a discussion of the practice of clientage by moderns see Whitaker (1970:374 ff. and 415 ff.) and the discussion of the post-coup era in Rwanda by Lemarchand (1970). As for the financing of ceremonials, see, for example, Lloyd (1953:14) and Galletti et al. (1956:523 ff.).

References Abernethy, D.B. (1969) The Political Dilemma of Popular Education: An African Case. Stanford: Stanford Univ. Press. Africa Diary (1970) “Some Ministers, many M.P.’s defeated in election.” 10, 3:4788– 4789. Africa Report (1970) “Kenya’s new cabinet.” 15, 4:8–9. Africa Report (1965) “September elections are lively.” 10 (October): 17. Akiwoko, A.A. (1964) “The sociology of Nigerian tribalism?” Phylon 23 (Summer): 155–163. Anderson, C.W., F.R.Von Der Mehden, and C.Young (1967) Issues of Political Development. Englewood Cliffs, N.J.: Prentice-Hall.


184

Ardener, S. (1958) “Banana co-operatives in the Southern Cameroons,” pp. 10–25 in Nigerian Institute of Social and Economic Research, Conference Proceedings. Ibadan, Nigeria: University College. Ardener, S. (1953) “The social and economic significance of the Contribution Club among a section of the southern Ibo,” pp. 128–142 in West African Institute of Social and Economic Research, Second Annual Conference Proceedings. Ibadan, Nigeria: University College. Austin, D. (1964) Politics in Ghana, 1946–1960. London: Oxford Univ. Press. Baker, T.M. (1956) “Political control amongst the Birom,” pp. 88–94 in West African Institute of Social and Economic Research, Fifth Annual Conference Proceedings. Ibadan, Nigeria: University College. Balandier, G. (1965) “Traditional social structures and economic changes,” pp. 385–395 in P.L.van den Berghe (ed.) Africa: Social Problems of Change and Conflict. San Francisco: Chandler. Barwell, C.W. (1956) “A note on some changes in the economy of the Kipsigis tribe.” J. of African Administration 8, 2:95–101. Basden, G.T. (1966) Among the Ibos of Nigeria. New York: Barnes & Noble. Bates, R.H. (1970) “Approaches to the study of ethnicity.” Cahiers D’études Africaines 10, 40:546–561. Bennett, G. (1969) “Tribalism in politics,” pp. 59–87 in P.H.Gulliver (ed.) Tradition and Transition in East Africa: Studies of the Tribal Element in the Modern Era. Berkeley and Los Angeles: Univ. of California Press. Brokensha, D. (1966) Social Change at Larteh, Ghana. Oxford: Clarendon. Burke, F.G. (1966) “Political evolution in Kenya,” pp. 185–239 in S.Diamond and F.G. Burke (eds.) The Transformation of East Africa: Studies in Political Anthropology. New York: Basic Books. Caldwell, J.C. (1969) African Rural-Urban Migration: The Movement to Ghana’s Towns. New York: Columbia Univ. Press. Caldwell, J.C. (1968) Population Growth and Family Change in Africa: The New Urban Elite in Ghana. New York: Humanities Press. Callaway, A. (1967) “Education expansion and the rise of youth unemployment,” pp. 191–211 in P.C.Lloyd et al. (eds.) The City of Ibadan. Cambridge: Cambridge Univ. Press. Callaway, A. (1963) “Unemployment among African school leavers.” J. of Modern African Studies 1, 3:351–371. Carter, G.M. (1960) Independence for Africa. New York: Frederick A.Praeger. Chona, M. (1970) “Chona lashes out at Zambia’s tribalists.” Zambia Mail (December 28). Chona, M. (1968) “Who is responsible for tribalism?” Zambia News (May 5). Chubb, L.T. (1961) Ibo Land Tenure. Ibadan, Nigeria: Ibadan Univ. Press. Cohen, A. (1969) Custom and Politics in Urban Africa: A Study of Hausa Migrants in Yoruba Towns. Berkeley and Los Angeles: Univ. of California Press. Coleman, J.S. (1958) Nigeria: Background to Nationalism. Berkeley and Los Angeles:


185

Univ. of California Press. Colson, E. (1971) “The impact of the colonial period on the definition of land rights,” pp. 193–215 in V.Turner (ed.) Profiles of Change: African Society and Colonial Rule. Cambridge, Eng.: Cambridge Univ. Press. Colson, E. (1966) “Migration in Africa: trends and possibilities,” pp. 107–113 in I.Wallerstein (ed.) Social Change: The Colonial Situation. New York: John Wiley. Comhaire, J.L. (1965) “Economic change and the extended family,” pp. 117–127 in P.L. van den Berghe (ed.) Africa: Social Problems of Change and Conflict. San Francisco: Chandler. 107–127 in P.L.van den Berghe (ed.) Africa: Social Problems of Change and Conflict. San Francisco: Chandler. Deutsch, K.W. (1961) “Social mobilization and political development.” Amer. Pol. Sci. Rev. 55, 3:493–514. Doornbos, M.R. (1970) “Kumanyana and Rwenzururu: two responses to ethnic inequality,” pp. 1088–1136 in R.I.Rotberg and A.A.Mazrui (eds.) Protest and Power in Black Africa. New York: Oxford Univ. Press. Enloe, C.H. (1973) Ethnic Conflict and Political Development. Boston: Little, Brown. Fischer, L.F., R.C.Mitchell, D.G.Morrison, and H.M.Stevenson (1969) “Modernization and political instability in Africa: an empirical assessment of macro theory.” Presented to the annual meeting of the African Studies Association, Montreal, October 18. Foster, P. (1965) Education and Social Change in Ghana. Chicago: Univ. of Chicago Press. Friedland, W.H. and C.G.Rosberg, Jr. [eds.] (1964) African Socialism. Stanford: Stanford Univ. Press. Galletti, R., K.D.S.Baldwin, and I.O.Dina (1956) Nigerian Cocoa Farmers: An Economic Survey of Yoruba Cocoa Farming Families. London: Oxford Univ. Press. Garlick, P.C. (1958) “The matrilineal family system and business enterprise,” pp. 121– 131 in Nigerian Institute of Social and Economic Research, Conference Proceedings. Ibadan, Nigeria : University College. Geertz, C. (1963) “The integrative revolution: primordial sentiments and civil politics in the new states,” in C.Geertz (ed.) Old Societies and New States. New York: Free Press. Ghai, D.P. (1970) “The Bugandan trade boycott: a study in tribal, political, and economic nationalism,” pp. 755–770 in R.I.Rotberg and A.A.Mazrui (eds.) Protest and Power in Black Africa. New York: Oxford Univ. Press. Gluckman, M. (1960) “Tribalism in modern British Central Africa.” Cahiers D’études Africaines 1, 1:55–70. Gould, P.R. (1968) “Problems of structuring and measuring spatial changes in the modernization process: Tanzania 1920–1963.” Presented at the annual meeting of the American Political Science Association, Washington, D.C., September 2–7. Grillo, R.D. (1969) “The tribal factor in an East African trade union,” pp. 297–321 in P.H. Gulliver (ed.) Tradition and Transition in East Africa: Studies of the Tribal Element in the Modern Era. Berkeley and Los Angeles: Univ. of California Press.


186

Gugler, J. (1968) “The impact of labour migration on society and economy in SubSaharan Africa: empirical findings and theoretical considerations.” African Social Research 6 (December): 463–486. Gulliver, P.H. (1969) “The conservative commitment in northern Tanzania: the Arusha and the Masai,” pp. 223–242 in P.H.Gulliver (ed.) Tradition and Transition in East Africa: Studies of the Tribal Element in the Modern Era. Berkeley and Los Angeles: Univ. of California Press. Gulliver, P.H. (1965) “Incentives in labor migration,” pp. 427–436 in P.L.van den Berghe (ed.) Africa: Social Problems of Change and Conflict. San Francisco: Chandler. Harries-Jones, P. (1969) “‘Home-boy’ ties and political organization in a copperbelt township,” pp. 297–347 in J.C.Mitchell (ed.) Social Networks in Urban Situations. Manchester: Manchester Univ. Press. Hershfield, A.F. (1969) “Ibo sons abroad: a window on the world.” Presented at the annual meeting of the African Studies Association, Montreal, October 15–18. Hicks, R.E. (1967) “Similarities and differences in occupational prestige ratings: a comparative study of two cultural groups in Zambia.” African Social Research 3:206– 227. Hicks, R.E. (1966) “Occupational prestige and its factors: a study of Zambia railway workers.” African Social Research 1:41–58. Hill, P. (1963) The migrant cocoa-farmers of southern Ghana: a study in rural capitalism. Cambridge: Cambridge Univ. Press. Horowitz, D.L. (1971) “Three dimensions of ethnic politics.” World Politics 23 (January): 232–244. Huntington, S.P. (1968) Political Order in Changing Societies. New Haven Conn.: Yale Univ. Press. Johnson, W.R. (1970a) The Cameroon Federation: Political Integration in a Fragmentary Society. Princeton, N.J.: Princeton Univ. Press. Johnson, W.R. (1970b) “The Union des Populations du Cameroun in rebellion: the integrative backlash of insurgency,” pp. 671–692 in R.I.Rotberg and A.A.Mazrui (eds.) Protest and Power in Black Africa. New York: Oxford Univ. Press. Jones, G.I. (1968) “The political organization of Old Calabar,” pp. 116–157 in D.Forde (ed.) Efik Traders of Old Calabar. London: Dawsons of Pall Mall. Jones, G.I. (1963) The Trading States of the Oil Rivers. London: Oxford Univ. Press. Kasfir, N. (1972) “Cultural sub-nationalism in Uganda,” pp. 47–148 in V.A.Olorunsola (ed.) The Politics of Cultural Sub-Nationalism in Africa. Garden City N.Y.: Anchor. Kaunda, K.D. (1969) “Humanism in Zambia,” pp. 9–33 in B.De Gaay Fortman (ed.) After Mulungushi: The Economics of Zambian Humanism. Nairobi: East African Publishing. Kilson M. (1966) Political Change in a West African State: A Study of the Modernization Process in Sierra Leone. Cambridge, Mass.: Harvard Univ. Press. La Fontaine, J.S. (1970) City Politics: A Study of Leopoldville, 1962–63. Cambridge:


187

Cambridge Univ. Press. La Fontaine, J.S. (1969) “Tribalism among the Gisu: an anthropological approach,” pp. 177–192 in P.H.Gulliver (ed.) Tradition and Transition in East Africa: Studies of the Tribal Element in the Modern Era. Berkeley and Los Angeles: Univ. of California Press Lemarchand, R. (1970) “The coup in Rwanda,” pp. 877–923 in R.I.Rotberg and A.A. Mazrui (eds.) Protest and Power in Black Africa. New York: Oxford Univ. Press. Lerner, D. (1958) The Passing of Traditional Society: Modernizing the Middle East. Glencoe, III.: Free Press. Lloyd, P.C. (1967) “The élite,” pp. 129–150 in P.C.Lloyd et al. (eds.) The City of Ibadan. Cambridge: Cambridge Univ. Press. Lloyd, P.C. (1966) The New Elites of Tropical Africa. London: Oxford Univ. Press. Lloyd, P.C. (1956a) “The changing role of Yoruba traditional leaders,” pp. 57–65 in West African Institute of Social and Economic Research, Third Annual Conference Proceedings. Ibadan, Nigeria: University College. Lloyd, P.C. (1956b) “Tribalism in Warri,” pp. 78–87 in West African Institute of Social and Economic Research, Fifth Annual Conference Proceedings. Ibadan, Nigeria: University College. Lloyd, P.C. (1955) “The development of political parties in western Nigeria.” Amer. Pol. Sci. Rev. 49 (September): 693–707. Lloyd, P.C. (1953) “Some modern changes in the government of Yoruba towns,” pp. 7– 20 in West African Institute of Social and Economic Research, Second Annual Conference Proceedings. Ibadan, Nigeria: University College. Lofchie, M.F. (1970) “The Zanzibari revolution: African protest in a racially plural society,” pp. 924–967 in R.I.Rotberg and A.A.Mazrui (eds.) Protest and Power in Black Africa. New York: Oxford Univ. Press. Lonsdale, J.M. (1970) “Political associations in western Kenya,” pp. 589–638 in R.I. Rotberg and A.A.Mazrui (eds.) Protest and Power in Black Africa. New York: Oxford Univ. Press. Maquet, J. (1970) “Rwanda castes,” pp. 93–124 in A.Tuden and L.Plotnicov (eds.) Social Stratification in Africa. New York: Free Press. Melson, R. and H.Wolpe (1970) “Modernization and the politics of communalism: a theoretical perspective.” Amer. Pol. Sci. Rev. 64 (December): 1112–1130. Mercier, P. (1966) “Problems of social stratification in West Africa,” pp. 340–358 in I. Wallerstein (ed.) Social Change: The Colonial Situation. New York: John Wiley. Mercier, P. (1965a) “Evolution of Senegalese elites,” pp. 163–178 in P.L.van den Berghe (ed.) Africa: Social Problems of Change and Conflict. San Francisco: Chandler. Mercier, P. (1965b) “On the meaning of ‘tribalism’ in Black Africa,” pp. 483–501 in P.L. van den Berghe (ed.) Africa: Social Problems of Change and Conflict. San Francisco: Chandler. Middleton, J. and E.H.Winter (eds.) (1963) Witchcraft and Sorcery in East Africa. London: Routledge & Kegan Paul. Mitchell, J.C. (1962) “Wage labour and African population movements in Central Africa,” pp. 193–248 in K.M.Barbour and R.M.Prothero (eds.) Essays on African


188

Population. New York: Frederick A.Praeger. Mitchell, J.C. (1956) “The Kalela dance: aspects of social relationships among urban Africans in Northern Rhodesia.” Lusaka, Northern Rhodesia: Rhodes-Livingstone Institute, Rhodes-Livingstone Papers 27. Mitchell, J.C. and A.L.Epstein (1959) “Occupational prestige and social status among urban Africans in Northern Rhodesia.” Africa 29, 1:22–39. Mitchell, J.C. and S.H.Irvine (1965) “Social position and the grading of occupations.” Rhodes-Livingstone J. 38:42–54. Morgenthau, R.S. (1965) “African elections: Tanzania’s contribution.” Africa Report 10, 11:12–16. Morrill, W.T. (1963) “Immigrants and associations: the Ibo in twentieth century Calabar.” Comp. Studies in Society and History 5 (July): 424–448. Morrison D.G. and H.M.Stevenson (1972) “Integration and instability: patterns of African political development.” Amer. Pol. Sci. Rev. 66 (September): 902–907. Morton-Williams, P. (1956) “A discussion of the theory of elites in 3 West African (Yoruba) context,” pp. 25–32 in West African Institute of Social and Economic Research, Fourth Annual Conference Proceedings. Ibadan, Nigeria: University College. Mulford, D.C. (1967) Zambia: The Politics of Independence, 1957–1964. London: Oxford Univ. Press. Nyerere, J.K. (1968) Ujamaa: Essays on Socialism. London: Oxford Univ. Press. Offodile, E.P.O. (1947) “Growth and influence of tribal unions.” West African Rev. 18 (August): 937–941. Okigbo, P. (1965) “Social consequences of economic development in West Africa,” pp. 415–426 in P.L.van den Berghe (ed.) Africa: Social Problems of Change and Conflict. San Francisco: Chandler. Okonjo, C. (1967) “Stranger communities: the western Ibo,” pp. 97–116 in P.C.Lloyd et al. (eds.) The City of Ibadan. Cambridge: Cambridge Univ. Press. Ottenberg, S. (1955) “Improvement associations among the Afikpo Ibo.” Africa 25 (January): 1–28. Paden, J.N. (1968) “Modernization, stability, and national integration in Africa.” Presented at the annual conference of the American Political Science Association, Washington, D.C., September 3–7. Parkin, D.J. (1969) “Tribe as fact and fiction in an East African city,” pp. 273–296 in P.H. Gulliver (ed.) Tradition and Transition in East Africa: Studies of the Tribal Element in the Modern Era. Berkeley and Los Angeles: Univ. of California Press. Plotnicov, L. (1970) “The modern African elite of Jos, Nigeria,” pp. 269–302 in A.Tuden and L.Plotnicov (eds.) Social Stratification in Africa. New York: Free press. Plotnicov, L. (1969) “Comments,” pp. 271–274 in G.Breese (ed.) The City in Newly Developing Countries. Englewood Cliffs, N.J.: Prentice-Hall. Plotnicov, L. (1967) Strangers to the City: Urban Man in Jos. Pittsburg: Univ. of Pittsburg Press.


189

Pons, V. (1969) Stanleyville: An African Urban Community Under Belgian Administration. London: Oxford Univ. Press. Post, K.W.J. (1963) The Nigerian Federal Election of 1959. London: Oxford Univ. press. Rabushka, A. and K.A.Sheplsle (1972) Politics in Plural Societies. Columbus, Ohio: Charles E.Merrill. Rosberg, C.G., Jr. and J.Nottingham (1966) The Myth of “Mau Mau”: Nationalism in Kenya. New York: Frederick A.Praeger. Rotberg, R.I. (1967) “Tribalism and politics in Zambia.” Africa Report 12 (December): 29–35. Rothchild, D. (1969) “Ethnic inequalities in Kenya.” J. of Modern African Studies 7 (December): 689–711. Scott, I. and R.Molteno (1969) “The Zambian general elections.” Africa Report 14, 1:42– 47. Scudder, T. and E.Colson (forthcoming) “The Kariba Dam Project: resettlement and local initiative,” in M.H.R.Bernard and P.Pelto (eds.) Technical Innovation and Cultural Change. New York: Macmillan. Sklar, K.L. (1967a) “Nigerian politics in perspective.” Government and Opposition 2, 4: 524–539. Sklar, K.L. (1967b) “Political science and national integration—a radical approach.” J. of Modern African Studies 5 (May): 1–11. Sklar, K.L. (1963) Nigerian Political Parties. Princeton, N.J.: Princeton Univ. Press. Sklar, K.L. and C.S.Whitaker, Jr. (1964) “Nigeria,” 597–654 in J.S.Coleman and C.G. Rosberg, Jr. (eds.) Political Parties and National Integration in Tropical Africa. Berkeley and Los Angeles: Univ. of California Press. Smock, A.C. (1971) Ibo Politics: The Role of Ethnic Unions in Eastern Nigeria. Cambridge, Mass.: Harvard Univ. Press. Soja, E.W. (1968) The Geography of Modernization in Kenya: A Spatial Analysis of Social, Economic, and Political Change. Syracuse, N.Y.: Syracuse Univ. Press. Southall, A.W. (1961) “Population movements in East Africa,” pp. 157–192 in K.M. Barbour and R.M.Prothero (eds.) Essays on African Population. New York: Frederick A.Praeger. Southall, A.W. and P.C.W.Gutkind (1957) Townsmen in the Making: Kampala and its Suburbs. Kampala, Uganda: East African Institute of Social Research. Stahl, K.M. (1969) “The Chagga,” pp. 209–222 in P.H.Gulliver(ed.) Tradition and Transition in East Africa: Studies of the Tribal Element in the Modern Era. Berkeley and Los Angeles: Univ. of California Press. Sutherland, A.C. (1956) “Private enterprise housing in the urban areas of Kumasi, pp. 119–125 in West African Institute of Social and Economic Research Fourth Annual Conference Prodeedings. Ibadan, Nigeria: University College. Tuden, A. and L.Plotnicov (1970) “Introduction,” pp. 1–29 in A.Tuden and L.Plotnicov (eds.) Social Stratification in Africa. New York: Free Press. Twaddle, M. (1969) “Tribalism’ in Eastern Uganda,” pp. 193–208 in P.H.Gulliver (ed.)


190

Tradition and Transition un East Africa: Studies of the Tribal Element in the Modern Era. Berkeley and Los Angeles: Univ. of California Press. Uchendu, V.G. (1965) The Igbo of Southeast Nigeria. New York: Holt, Rinehart & Winston. Van Den Berghe, P.L. (1965) “Introduction,” pp. 1–11 in P.L.van den Berghe (ed.) Africa: Social Problems of Change and Conflict. San Francisco: Chandler. Whitaker, C.S., Jr. (1970) The Politics of Tradition: Continuity and Change in Northern Nigeria, 1946–1966. Princeton, N.J.: Princeton Univ. Press. Whitaker, C.S., Jr. (1967) “A dysrhythmic process of political change.” World Politics 19 (January): 190–217. Wilson, G. (1941–1942) “An essay on the economics of detribalization in Northern Rhodesia.” Lusaka, Northern Rhodesia: Rhodes-Livingstone Institute, RhodesLivingstone Papers 5 and 6. Wilson, G.M. (1966) “The African elite,” pp. 431–461 in S.Diamond and F.G Burke (eds.) The Transformation of East Africa: Studies in Political Anthropology. New York: Basic Books. Xydias, N. (1956) “Prestige of occupations,” pp. 458–469 in D.Forde (ed.) Social Implications of Industrialization and Urbanization in Africa South of the Sahara. Paris: UNESCO. Young, C. (1965) Politics in the Congo. Princeton, NJ.: Princeton Univ. Press. Zolberg, A.R. (1964) One-Party Government in the Ivory Coast. Princeton, N J.Princeton Univ.Press.


191

48 Patterns of Ethnic Separatism* Donald L.Horowitz

* Source: Comparative Studies in Society and History, 1981, vol. 23, pp. 165–195.

In the analysis of ethnic separatism and secession,1 two approaches can be distinguished. One is to ask what forces are responsible for the general upsurge in secessionist movements, from Burma to Biafra and Bangladesh, from Corsica to Quebec, and from Eritrea to the Southern Philippines.2 Another approach is to ask what moves certain territorially discrete ethnic groups to attempt to leave the states of which they are a part (or at least to secure substantial territorial autonomy), whereas other groups, also regionally concentrated, make no such attempts. The first question calls for a general explanation of aggregate trends; it aims to compare the present with some period in the past. The second calls for an explanation that can discriminate among classes of cases; it entails comparison not across time, but across space. Both questions are important, but what will answer one will not necessarily respond to the other. Conditions responsible for a general upsurge of separatism tend to be more or less constant across countries.3 Yet, despite its frequency, secession is a variable phenomenon. Some movements emerge early in the life of a new state, seemingly with little provocation. Others develop only after a prolonged period of frustration and conflict. Some movements simmer for years, even decades, and in the end may come to nothing, whereas others burst quickly into warfare. But many movements never reach even a slow simmer, much less a quick boil. A variable phenomenon of this kind is not explicable by a constant. In explaining the upsurge of separatism, there is much to commend the view that it results from the widespread diffusion of the doctrine of national self-determination.4 Many aspects of ethnic conflict testify to the power of ideal in politics, but none so much as separatism. Self-determination, a doctrine traceable to eighteenth-century conceptions of popular sovereignty, flowered with the burst of nationalism in nineteenth-century Europe. After World War I, it was applied to remake the map of Eastern Europe. The process was repeated after World War II, with the termination of colonial control in Asia and Africa. Nineteenth-century nationalism involved both the building of states like Germany and Italy out of smaller units and the growth of separatism, especially in the Hapsburg empire. The twentieth-century version entailed the dismemberment of empires and large states in favor of smaller units. But, in this process, self-determination was pursued incompletely, in Asia and Africa only to the level of preexisting colonial boundaries. Not surprisingly, the doctrine of self-determination is invariably invoked by secessionists. From this perspective, separatism is appropriately viewed as the working out of the logic of an idea—that “political self-expression,” usually on a territorial basis,

Patterns of Ethnic Separatism

192

is “a necessary concomitant of group distinctiveness.”5 Self-determination is a legitimizing ideology for ethnic separatism, perhaps a necessary condition for its emergence. Yet, if the idea of self-determination is well-nigh universal, the inaction of nonseparatist groups cannot be attributed to their failure to achieve full consciousness of nationhood.6 As we shall see, some very remote groups with low levels of literacy and few university graduates have invoked the doctrine of self-determination in support of their attempts at secession, whereas some centrally located, well-educated groups, attuned to worldwide ideological currents, have eschewed separatism. The powerful ideology of self-determination helps explain the emergence of a political environment hospitable to territorially divisive claims, but it cannot explain which groups will take up the cause. Equally helpful in understanding aggregate trends are utilitarian theories that account for the changing propensity to secede by identifying changes in the overall “balance of advantages in any particular period.”7 If indeed the balance of advantages, reckoned in terms of economic prosperity or security from external attack, has recently shifted “to the disadvantage of the sizeable state,”8 such a shift might constitute an implicit inducement to separatism. If neither prosperity nor security is advanced significantly by remaining in a large, multiethnic state, then it would not be surprising to discover an increase in the number of ethnic groups that choose to opt out. Yet, if changes of this kind have occurred, they constitute—like the ideology of selfdetermination—a generally applicable inducement to separatism, an alteration in the structure of incentives for all groups subject to the changes. But variations in the behavior of potential secessionist groups cannot be explained by a new “balance of advantages” that is constant across territories.9 What may be helpful, then, in accounting for changes over time is not helpful in identifying patterns of separatism: what kinds of group are likely to secede, and under what circumstances. To discern patterns of separatism, it is necessary to begin with variables across space rather than with variables across time, particularly with variables attuned to relations between peoples, including the anxiety and anger that so often attend those relations. Calculations of group interest play their part, though some ethnic groups opt for secession when it does not appear to be in their interests to do so. In decisions to secede, group interest is alloyed with enmity and offset by apprehension. The roots of those decisions are to be found in the texture of group relations. One fairly firm rule of thumb can be laid down at once. Whether a secessionist movement will emerge at all is determined by domestic politics, by the relations of groups and regions within the state. Whether a secessionist movement will achieve its aims, however, is determined largely by international politics, by the balance of interests and forces that extend beyond the state.10 Occasionally, considerations of the means available to support secessionist movements, including external assistance, may modify secessionist sentiment—though separatists are often surprisingly heedless of such prudential constraints. Although secession lies squarely at the juncture of internal and international politics, for the most part the emergence of separatism can be explained in terms of domestic ethnic politics. To this broad rule of thumb, there is a major exception. A group that might otherwise be disposed to separatism will not be so disposed, or will limit its activity in that


193

direction, if its secession is likely to lead not to independence, but to incorporation in a neighboring state, membership in which is viewed as even less desirable than membership in the original state. The cases in which this inhibition is likely to operate involve irredentism, where an international boundary divides members of a single ethnic group. The Baluchis and Pathans of Pakistan, for example, are likely to limit their separatist activity to the extent that it makes them vulnerable to incorporation in Afghanistan or, in the Baluchi case, Iran.11 The Ewe of Ghana are not likely to do anything that would risk their merger into Togo. Similar considerations, however, will not restrain the Malays of Southern Thailand, many of whom might indeed prefer to join Malaysia. This inhibition does not indicate under what conditions irredentism will occur; it merely highlights, in at least a few important cases, a limitation on domestically generated collective inclinations. At this point, an issue of definition intrudes, one well illustrated by the limited goals of some of the groups just mentioned. Should the terms separatism and secession be confined to movements aiming explicitly at an independent state, or should they be extended to movements seeking any territorially defined political change intended to accord an ethnic group autonomous control over the region in which it resides? Conceived in the latter way, separatism would include ethnic demands for the creation of separate states within existing states or for a broad measure of regional autonomy, short of independence. There is some ground for thinking that groups demanding complete independence may have the strongest sense of grievance. The contrast between Catalan and Basque claims in Spain is revealing on this score. Catalan ethnic sentiment runs as deep as Basque sentiment does, and probably has broader support. But Basque political organizations have more frequently turned to violence and more frequently demanded independence. Whereas Catalan organizations have aimed at autonomy within Spain, the Basques are divided. Franco’s severe repression of the Basques, many of whom had supported the Republicans, probably helps explain the unyielding character of some Basque organizations. In the Basque case, at least, there seems to be a clear and direct linkage between ethnic antipathy and declared political objectives. In many other cases, however, this linkage is more tenuous. The Kurds in Iraq consistently denied that their objective was independence. Even as they fought and died in the 1960s and 1970s, they eschewed anything beyond regional autonomy. The reason, presumably, was tactical: had they declared independence as their goal, the Iraqi Kurds would have engendered hostility from regimes in neighboring Syria, Iran, and Turkey, all of which have Kurdish minorities. In the 1974 warfare in Iraq, Iran supplied arms, food, and cross-border facilities for the Kurdish fighters, and this support particularly ensured that the movement demanded only autonomy. Demands can also shift from autonomy to independence and back again, depending on the state of negotiations between central regimes and separatists. The Moro National Liberation Front in the Philippines moved from autonomy demands to demands for separate statehood after the Philippine government had adopted a decentralization plan the Front found wanting.12 The Mizo National Front in India followed the same path, agreeing to a solution within the framework of Indian federalism in 1976 but, after a ceasefire broke down three years later, returning to warfare to achieve independence.13


194

Other movements, such as the Southern Sudanese, equivocated on their demands, using ambiguous terms like “self-determination” to cover internal differences.14 The Chad National Liberation Front, presumed to be fighting a war for the secession of the North, long refused to declare its objectives, and in the end more than half the country, including the capital, was in rebel hands. Tactics play a large role in the statement of objectives. The often tactical nature of demands, their elasticity and even fickleness, the willingness of independence movements to settle for very much less than statehood, and the occasional interest of secessionists in capturing the whole state if that proves possible—all of these argue for an inclusive conception of separatism and secession. Such a conception should embrace movements seeking a separate region within an existing state, as well as those seeking a separate and independent state.

Paths to Secession “Inevitably,” wrote Immanuel Wallerstein at the time of the Katanga secession, “some regions will be richer (less poor) than others, and if the ethnic claim to power combines with relative wealth, the case for secession is strong…. [E]very African nation, large or small, federal or unitary, has its Katanga.”15 Wallerstein was right to link the ethnic claim with the character of the region from which the ethnic group springs. These are the two conditions that matter most. But he limited the potential for secession unduly when he confined it to wealthy regions. In point of fact, there are several paths to secession, and rich regions are not the leading secessionists. They are far outnumbered by regions poor in resources and productivity. Despite strong feelings of alienation—or worse—neither Ashanti in Ghana nor the Western Region of Nigeria nor Buganda in Uganda, all prosperous regions, made a serious effort to secede. By contrast, wars have been fought by peoples in the poor regions of, among others, the Southern Sudan, the Southern Philippines, and Northern Chad. Why this is so we shall soon see. Table 1 provides a simple matrix of potential secessionists, principally in Asia and Africa. It includes groups that have and have not attempted to secede. The variables are straightforward. They are based on the positions of ethnic groups and regions relative to others in the state. For shorthand purposes, separatist ethnic groups are characterized as being either “backward” or “advanced.”16 Although these terms are subjective, they refer to group comparisons that have some objective bases. An advanced group is one that has benefited from opportunities in education and nonagricultural employment. Typically, it is represented above the mean in number of secondary-school and university graduates; in bureaucratic, commercial, and professional employment; and in per capita income. Certain stereotypes are commonly associated with these attributes. Advanced groups are generally regarded by themselves and others as highly motivated, diligent, intelligent, and dynamic. Backward groups, less favorably situated on the average in terms of educational attainment, high-salaried employment, and per capita income, tend to be stereotyped as indolent, ignorant, and not disposed to achievement.17 Needless to say, these traits are imputed to each group on the basis of the position of a minority of its members, and sometimes the basis in fact is greatly exaggerated. Nonetheless, group position and the


195

putative qualities associated with it are potent factors in ethnic conflict generally, and they certainly condition collective orientations to the possibility of secession.

Table 1: Potential Secessionists, by Group and Regional Position Backward groups Advanced groups Backward regional economies Southern Sudanese Ibo in Nigeria Karens in Burma Tamils in Sri Lanka Muslims in the Philippines Baluba (Kasai) in Zaire Muslims in Chad Lozi in Zambiaa Kurds in Iraq Nagas and Mizos in India Muslims in Thailand Bangalis in Pakistan Northerners in Ghanaa Basques in Spain Advanced regional economies Lunda in Zaire Bakonjo in Uganda Yoruba in Nigeriaa a Batéké in Gabon Baganda in Ugandaa a. Nonsecessionist groups. Separatist regions are characterized as backward or advanced according to the relative economic position of the particular region, as measured by regional income per capita excluding remittances from other regions (which would likely be terminated or reduced in the event of secession). Although I say “measured by,” data on regional income per capita are only sporadically available, and rarely on a reliable basis, for Asian and African countries. While the paucity of data excludes the possibility of analysis based on precise degrees of regional backwardness, advancement, or disparity between the two in given countries, identification of backward and advanced regions is not difficult. The same is true regarding group position. This characterization both of regions and groups ignores some common complexities. Table I assumes the existence of geographically concentrated ethnic groups, that may or may not become separatist. However, many groups that possess a geographically identifiable homeland are no longer geographically concentrated. Large numbers of group members may live outside the home region, a circumstance likely to have some impact on the emergence of separatism. Conversely, a secessionist region often contains more than one major ethnic group, and the groups may differ in their positions relative to groups outside the region. Similarly, the measurement of regional position by per capita income may obscure important elements of intraregional difference. Eritrea, for example, has had industrially developed cities but an exceedingly poor countryside: Which is the politically relevant reality? Then, too, although I shall speak of a backward region and an advanced region, as if any state had only two regions, rarely is a state so clearly bifurcated. I shall deal with some of these complexities at later points, but for the moment it is best to proceed with a simpler framework.18 The interplay of relative group position and relative regional position determines the


196

emergence of separatism. In stressing this interplay, I mean to reject direct causal relationships between regional economic disparity and ethnic secession. If degree of regional economic disparity alone determined the emergence of separatism, it would be reasonable to expect the preponderance of such movements in those states occupying the middle income levels, for in such states regional economic disparities seem to be greatest.19 But no such tendencies can be identified. Secession is attempted in lowincome states like Ethiopia and Chad, as well as in the Philippines and Nigeria, and, needless to say, it is also an issue in a number of economically developed countries. Relative regional position is a causal element in the emergence of secession, not because it predicts separatism in any straightforward way, but because it conditions the claims ethnic groups make and their responses to the rejection of those claims. The four categories of potential secessionists depicted in Table 1 differ from each other in several major respects. The demands the groups advance before separatist sentiment crystallizes, the events that move the groups to secession, the calculations that attend the decision to separate, and the timing of the decision all vary according to whether the group is considered backward or advanced and whether it resides in a backward or advanced region. Table 1 does not provide an exhaustive enumeration of separatist movements, of which there have been dozens, if not hundreds, in the postcolonial period. Furthermore, it includes some nonsecessionist groups for comparison. Even so, the table suggests the prevalence of so-called backward regions among secessionists. In part, this may be a function of the coincidence of regional backwardness with geographic distance from the center. Economic backwardness is more common on the periphery. In states where the span of governmental control is limited, peripheral areas might more readily contemplate secession.20 Yet the logic of secession comprehends much more than just the difficulty of the center in exerting control. Distance is but a minor factor in the over-all prevalence of backward regions among secessionists. Indeed, there is more than one rationale for the secession of a backward region. There are four different paths to ethnic secession, which correspond to the four different cells of Table 1.

Backward Groups in Backward Regions By far the largest number of secessionists can be characterized as backward groups living in backward regions. These groups are typically early seceders. They often attempt to secede rather soon after their countries have achieved independence or after rejection of the claims they advance. They conclude rapidly that they have a small stake in preserving the undivided state of which they are a part. In fact, some such groups had earlier doubts: Moros in the Philippines, Nagas in India, Karens in Burma, and Southern Sudanese were among those groups that asked for a prolonged colonial period, a separate independence, or special arrangements to protect them after independence. All of them feared competition with their neighbors within the bounds of a single political arena. Fears of this kind are based not merely on numerical inferiority, but on a sense of weakness vis-à-vis more “efficient,” “aggressive,” “sharp-witted,” “dynamic,” “industrious,” and better educated members of other ethnic groups.21 Sensing competitive


197

incapacities, they at first tend to demand representation in politics and the public service in proportion to their numbers. Inevitably, this demand is unmet, for relative backwardness implies a shortage of candidates eligible for such positions. When the denial of such opportunities is coupled with clear signs that the state is dominated by members of other groups, backward groups in backward regions choose to opt out. Often these regions are deficit areas that receive a subsidy from the center.22 Consequently, the decision to secede is taken despite the economic costs it is likely to entail. This willingness to sacrifice, together with the rapidity with which such movements get going, is evidence of the sense of desperation backward groups feel in assessing their ability to compete in the undivided state. Elite and mass economic interests, however, generally diverge at the moment of decision. Whereas the region as a whole stands to suffer if it opts for secession, educated elites stand to gain from the creation of new opportunities in a smaller, albeit poorer, state. These include those high positions from which these elites, with their generally lower seniority, would be excluded in the undivided state. Secession creates new positions, while reducing the pool of competitors. Advanced segments of backward groups do not resist but generally lead the movement.23 Nonetheless, the frequency, enthusiasm, and violence of separatist movements among backward groups in backward regions can scarcely be put down to selfish elite motives alone. It is true that, whereas secession enables leaders to eliminate the interethnic competition they previously faced, many other people may be adversely affected by the end of revenue subsidies and the severance of the backward economy from the state. Yet the formal divergence of interest is just that; by the time the movement gets underway, calculations of sacrifice and opportunity are invariably overwhelmed by an avalanche of ethnic sentiment that the undivided state is intolerable. It is instructive to examine more concretely how such sentiments develop. Time and again, it is the civil service issue that highlights grievances. Not only do backward groups in backward regions receive a share of government positions which is dramatically smaller than their share of the population, but, in addition, civil servants are imported from more advanced regions to serve in theirs. Kurdish demands in Iraq reflect this dual grievance very well. They recurrently embody proposals for proportional representation in the Iraqi civil service, cabinet, and national assembly and for the exclusion of non-Kurds from government service in Kurdish areas.24 The dual character of the demands indicates that the issue is not merely one of ethnic representation; it has shifted to domination. In the Sudan, for example, Southerners, more than one quarter of the population, received only six of eight hundred civil service openings at independence, and they were slighted in other ways as well. They held only three of forty-three seats on a constitution-drafting committee; their position in other government bodies was equally poor: less than 3 percent of postindependence army commissions, 4 percent of newly gazetted police officers, and so on.25 The Southern elite was small, but it had great expectations that were quickly thwarted. Moreover, British civil servants in the South were usually replaced by Northerners. By all accounts, the new administrators were not attuned to Southern sensibilities. It took little beyond this to convince Southerners that a new colonialism had arrived: imposition of Arabic for certain official purposes in the South, hints of alignment with Arab Egypt, hostility toward Christian missionaries.


198

Concluded two Southern leaders: “the administration, the army, the police, the judiciary and trade in the South [are] all in Arab hands; Arabic is the official national language as well as the medium of instruction; Friday is the day of rest, etc. Could domination be better expressed?”26 Civil service appointments and postings have been prominent accelerators of separatist sentiment among a variety of backward groups in backward regions, ranging from Chad to Nagaland to the Telangana region of Andhra Pradesh. With the French departure from Chad in 1960, Southerners, better educated in French, were able to claim the best civil service positions. Like Northern Sudanese sent to administer the backward South, Southern Chadiens were sent to govern the backward North, which they did with scant regard for Northern local authorities. This, perhaps more than the arrest of Northern politicians, prompted rebellion. Within a few years of independence, there was fighting.27 Similarly, in Nagaland, non-Naga Indian officials replaced the British, and “a change in the spirit of administration, if not yet in the pattern, was immediately felt by the Nagas.”28 By 1951, the Nagas had issued a declaration of independence. And when Andhra Pradesh was created out of a mosaic of Telugu-speaking areas in 1956, wellqualified coastal Andhras moved into Telangana to take civil service positions in Hyderabad. Aspiring Telangana students were outraged by what seemed a theft of their opportunities. On top of this, it was said that there was a shortage of “skilled” Telangana applicants for public service posts, and Telanganans were told that they were “indolent,” that their “Urduized” Telugu was impure, and that their habits were “feudal.”29 In each case, a backward region inhabited by a backward group was, it seemed, “colonized” by administrators from a more advanced region and a more advanced group. In economic terms, of course, the actions that precipitate separatist activity can be viewed as merely the equilibration of factors of production between two unequally developed regions.30 Regions with a surfeit of human resources export them to regions with a deficit. The same is true of capital for investment. In Telangana, to cite one case, cheap but fertile land was purchased by ambitious coastal Andhra farmers. A single state implies a single, unbounded market for labor and capital. And there, precisely, is the rub, for the market may be unbounded, but its populations are encapsulated within ethnic boundaries. There may be no economic barriers between regions, but there are psychological barriers. Telangana farmers thus resented the more efficient, productive coastal farmers who migrated there, even though their activities presumably raised the value of Telangana land. Typically, developments that occur within separatist regions are paralleled by actions at the center that are unfavorable to the backward group. These include abandonment of promises of special concessions—for example, ignoring repeated pledges to consider federalism in the Sudan or failing to enforce arrangements for local job preferences in Telangana. They also include policies that augur homogenization, such as adoption of a single state language in Assam and the Sudan, of a single state religion in Burma, or of pan-Arabist doctrine in the Sudan and Iraq.31 Groups like the Karens, the Nagas, the Mizos, the Southern Sudanese, the Philippine Muslims, and the Kurds, all with a keen sense of weakness, are easily convinced by such policies that their only hope of resisting domination lies in some form of separation. As will be shown, however, not all groups are so readily persuaded to withdraw from competition within the unified state.


199

It may seem paradoxical that a poor region, benefiting from association with more prosperous regions, should want to terminate the arrangement. Yet the desire recurs. Occasionally, the economic costs of separatism are tempered by the prospect of claiming some resource located in or near the secessionist area, such as oil on the fringes of Iraqi Kurdistan. Yet this is rarely decisive. Many groups without such opportunities simply choose to pay whatever price is required. In the secessionist idiom of the Northern Nigeria of 1966, “What does money matter when it is a question of honour?”32 For backward groups in backward regions, secessionist sentiment is weak when political debate still revolves around predictions of whether secession would entail an economic loss and, if so, whether the political gain would be worth the economic sacrifice. Finally, it is necessary to take account of some exceptions. Why, for example, did East Bengal remain in Pakistan for almost twenty-five years before seceding? Why did Northern Nigeria edge toward secession in 1966, only to pull back again? The East Bengalis, particularly the Bengali Muslims, were clearly a backward people in a backward region. At $63, per capita income in East Bengal before secession was just half that in West Pakistan. At independence, only three Bengali Muslims, of a total of about a hundred members of the Indian Civil Service, chose to serve Pakistan.33 Poorly represented in the civil administration, in the army, and in business, the Bengalis were said to be rich only in politicians.34 The Bengalis contended that expenditure per capita was skewed toward West Pakistan, where development investment would presumably bring a higher rate of return. In addition, they could claim something backward regions are usually in no position to claim—that (because of jute production) they made a disproportionate contribution to export earnings. And, as in other cases, they were governed by outsiders—the East Bengal administration was filled with West Pakistanis, and Bengali was only grudgingly recognized as an official language. If backward groups in backward regions have a low threshold for separatism, why were the Bengalis, having similar characteristics, such late seceders? For one thing, the great distance between East and West Pakistan, which was so often said to have rendered the unity of the country precarious, may instead have contributed to its durability. Distance may have made complete domination of the East by the West more difficult and may have limited irritating contact between people in the two wings. More important for present purposes, however, the East Bengalis were a very large group, more than half the total population. Unlike most backward minorities, they could and did, right up to the eve of secession, entertain a hope that their numbers might be translated into sufficient political power at the center to compensate for their competitive disadvantages. This was reason enough to be patient, at least until the majority finally won by the Bengali party, the Awami League, in the 1970 elections was decisively rebuffed by the armed forces. For the Nigerian Hausa, the strategy was the same. Despite their competitive weakness, they embarked on a course of pan-regional politics, because the Northern region gave them formidable numerical resources to do so and also because they had an historic aspiration, antedating the British conquest, to march south to the sea. For some years after independence, Hausa politicians ruled at the center by skillful exploitation of differences among Southerners. When military rule reversed this political ascendancy, there was separatist sentiment expressed by some Hausa, but it was quickly checked.


200

After the Northern countercoup of July 1966, nothing further was heard of it.35 The East Bengalis and the Hausa travelled in different directions, to be sure, but the animating forces were the same. The Bengalis waited a long time, but finally chose to leave Pakistan. The Hausa clearly contemplated leaving Nigeria but reversed course. In both cases, urges to secede were propelled by the same fears that underlay the actions of backward groups in backward regions elsewhere. In each case, however, these were offset by the prospect of exerting considerable, even hegemonic, power in the undivided state. What distinguished the Hausa and the Bengalis from comparable groups in other countries was their large numbers, which afforded them a chance to resist domination and perhaps even impose their own. That is why neither chose an early secession: Regardless of advancement or backwardness, groups with substantial political power in the undivided state will prefer to remain in it, rather than secede.

Advanced Groups in Backward Regions In sharp contrast to the secessionist activity of backward groups in backward regions is the behavior of advanced groups from backward regions. Where backward groups are early seceders, advanced groups are late seceders. Where backward groups have often sought colonial protection, a postponement of independence, or special arrangements once independence seemed inevitable, advanced groups are sometimes in the forefront of anticolonial movements.36 Where backward groups seek proportionality in government employment, advanced groups seek only assurances of nondiscrimination. Where backward groups attempt to keep ethnic strangers from government service in their regions, advanced groups affirm the principle of unrestricted mobility. Where backward groups attempt secession as soon as their competitive fears seem confirmed, advanced groups attempt secession only when all hope of salvaging their position in the country is dashed. Where backward groups in backward regions attempt to separate despite the economic costs of secession, advanced groups in backward regions decide on separation only when the advantages of remaining in the unified state are much reduced and the costs of remaining seem perilously high. Advanced groups in backward regions have a much higher threshold of tolerance for political events inimical to their interests than backward groups do. Initially, most groups try to remain in the undivided state, but how hard they try is a function of how able they feel to compete in it. After the first serious rebuffs, the Southern Sudanese moved rapidly from an equivocally federalist to an openly secessionist position under the aegis of successively more intransigent organizations. It took little to push them out. By contrast, the Ibo went through serious collective violence at Jos in 1945 and at Kano in 1953, their nationalism unimpaired. It took two massacres in 1966, separated by a Northern coup in which many Ibo officers and men lost their lives, before the Ibo embarked on Biafra. Thus did the Nigerian nationalist Ibo become secessionists and the Northern secessionists become preservers of a single Nigeria. The Ceylon Tamils have also been patient. They endured the riots of 1956 and 1958, Sinhalaonly legislation, and discrimination against them in government employment, all without demanding anything more than a mild federalism. Then, in 1972, came a new constitution


201

that ignored their demands and conferred state patronage on Buddhism, the religion of most Sinhalese. This was followed by sharp discrimination against Tamil university applicants and by the anti-Tamil riots of 1977. Only in 1976 did Tamil leaders unequivocally declare for a separate state, and in 1978 there was an upsurge of separatist terrorism. Both Ibo and Tamil claims took a long time to mature. Because advanced groups in backward regions secede only as a last resort, when pushed to the limits of their endurance, many such groups, even when severely frustrated, do not reach the point of choosing separatism. The Lozi in Zambia are such a group, aggrieved but not inclined seriously toward secession. The Lozi homeland, Barotseland, was administered separately by the British. At independence it was agreed that Barotseland would have a special status within Zambia.37 This arrangement was soon abrogated by the Zambian government, which treated Barotseland as just another province, restricting the powers of the Lozi monarch and abolishing the Barotse legislature. Following this, Barotseland turned decisively away from the Zambian nationalist party, UNIP, and toward the opposition. Each time the central government has acted to limit provincial power, secession has been on the lips of the Lozi aristocracy. But it has gone no further. Secession has had no real support from educated Lozi elites and has produced no coherent movement. Despite provocation, important elements of the Lozi community still prefer a unified Zambia. The position of advanced groups in backward regions makes it clear why their threshold of tolerance is higher than that of backward groups. Advanced groups in backward regions are generally population exporters.38 Barotseland has poor soil. Jaffna, the heartland of the Ceylon Tamils, is dry and unproductive. Iboland, having suffered soil erosion, is also infertile. In each case, group survival has depended upon the search for opportunities outside the region, upon what is called push migration. Barotseland exported Lozi labor to South African gold mines and white-collar workers to the Zambian copper-belt. Lozi also achieved education far out of proportion to their share of the population; they were the largest group of students at Lusaka’s premier secondary school on the eve of independence. This paved the way to opportunities all over Zambia.39 The same applies to the Ceylon Tamils, who took advantage of educational opportunities and migrated to the South of Sri Lanka as traders, bureaucrats, and professionals.40 Ibo, too, settled all over Nigeria; Ibo clerks, traders, and laborers were to be found in every urban area.41 Each of these backward regions thus came to depend on remittances from the sons it had exported to other regions of the country. This explains why the Ibo became the apostles of pan-Nigerian nationalism and the implacable opponents of compartmentalizing the country. “The Ibo, as well as the Ibibio, had strong personal economic reasons tor wanting Nigeria to be a nation with freedom of movement and enterprise.”42 The educated Lozi elite, especially but not only those outside Barotseland, took an equally strong position in favor of a united Zambia.43 And the Ceylon Tamils preferred “fruitful participation in national affairs instead of being cramped and cribbed in the arid and overcrowded Jaffna Peninsula.”44 One undivided nationwide field of opportunity seemed in each case to be at the heart of ethnic interest. Whereas for backward groups, unable to compete outside their home regions, the question is whether they will govern themselves or will be governed by carpetbaggers, for advanced groups the issue is whether they will be accepted outside their own regions.


202

For an advanced group, widely distributed throughout the country, secession would have the clearest disadvantages. It would dry up vital extraregional income sources and trigger a return of talented but unemployed group members to the homeland. Alternatively, it would leave large segments of the secessionist ethnic group outside the homeland, where they would be vulnerable to discrimination and attack and where their own income and their remittances to the home region would also be jeopardized. Advanced groups do not feel unable to compete—indeed, others often sense in them excessive confidence. Unlike backward groups, therefore, advanced groups are unwilling to disregard the formidable economic costs of secession in order to free themselves from disagreeable competitive relationships. Hence their extreme wariness of abandoning the national system for a more parochial secessionist region, with its greatly restricted opportunities. Backward groups are, of course, not troubled by these inhibiting considerations. One reason for their ability to make such quick judgments in favor of secession is that they need worry less often about the presence of large numbers of their kinsmen outside their home region. There are exceptions, to be sure. The majority of Karens, for example, do not live in the core Karen area, and it was not easy for Karen secessionists to stake out a contiguous territory that would embrace an acceptable number of Karens.45 But the usual situation of backward groups is different. Often lacking education and marketable skills, they are less likely to migrate in large numbers out of their home regions.46 Migration out of their region is inherent in the situation of advanced groups from backward regions, as it is not for backward groups. It is their diaspora, then, and the nationwide field of opportunity that together inhibit secessionist impulses among advanced groups from backward regions. Yet the Ibo did fight a war of secession, the Kasai Baluha did set up their own state, and the Ceylon Tamils have threatened to do the same. What forces overcame their inhibitions? If the dispersion of group members and the advantages of a single, unbounded, nationwide field of opportunity impede the growth of seccessionist sentiment among population-exporting groups, then clearly the reversal of these conditions can provide a real fillip to separatism. When the national system begins to break down because of regional parochialism or because of discrimination against advanced groups, the advantages of “one Nigeria” or “one Sri Lanka” can readily be called into question. When a population-exporting region experiences an in-gathering of its scattered exiles, inhibitions on secessionist impulses can be swept aside. This is all the more so because push migrants from backward regions do not return home en masse unless something dramatically unfavorable has happened to them. The two unfavorable things that happen most often are discrimination that curtails their opportunities and violence directed against them. The two sometimes go together, and they are the most common precipitants of secession among advanced groups in backward regions. Advanced groups from population-exporting regions are disproportionately victims of ethnic violence. This explains the paradox of their position: reluctant separatists yet not infrequently pushed to the point of seceding. The most severe episodes of such violence produce massive back-migration that fosters secession. In Zaire, in 1959, when Lulua killed Baluba in Luluabourg in Central Kasai, some 50,000 Baluba fled back to South Kasai. Gradually, Baluba from all over Central Kasai followed suit. By 1963, the population of South Kasai had quadrupled. The result of this flight, which signified an


203

end to Baluba opportunities outside their own region, was the secession of South Kasai.47 This was the Ibo case in microcosm. The Nigerian violence of 1966 spurred a similar eastward movement of Ibo. Many Ibo fled after the May riots, then later returned to the North. But the September-October killings were more organized and extensive. These riots generated a flood of refugees, some of them maimed. Their arrival in the East inflamed sentiment there. Perhaps a million Ibo returned to the East, convinced that it was dangerous to be an Ibo elsewhere in Nigeria. This situation was the culmination of a long process of whittling away Ibo opportunities. It began, in formal terms, with the “Northernization” of administration and business pursued by the Northern Regional Government in the 1950s.48 The result was discrimination against Ibo government servants and businessmen in employment, contracts, and licenses. The process accelerated in the 1960s, with attacks on alleged Ibo nepotism and concerted struggles to remove Ibo from high government and university positions. The victims of violence who fled eastward were joined by those who felt Ibo prospects in other regions were no longer salvageable. The Ibo, it was said, had built the country but would not be permitted to reap the rewards.49 Thus far, the Ceylon Tamils have been spared the massive violence of an episode comparable to the killings in Nigeria in 1966. But they have periodically been victimized in widespread riots. At such times, refugees from the South have carried back credible tales of lack of protection for Tamils in Sinhalese areas. Beyond this, the Tamil position in the country has time and again failed to receive the official recognition the Tamils demand. The relegation of the Tamil language to a distinctly secondary place in official business resulted in a decline in opportunities for Tamil government servants without a knowledge of Sinhala. The 1972 constitution reinforced the position of Sinhala, accorded a “foremost place” to Sinhalese Buddhism, and denied Tamil claims for regional autonomy.50 The constitution was a decisive symbolic rebuff to the Tamils. Then came a policy of “standardizing marks,” a system of weightage in grades to offset the superior performance of Tamil students in academic examinations. The result was a dramatic decline in Tamil representation in higher education, a decline that had begun earlier. Reductions of 30–40 percent in Tamil enrollment, depending on the field, were experienced in a period of one to three years.51 In the decade between 1963 and 1973, the percentage of Ceylon Tamils with university education fell from 2.2 to 0.6, below the Sinhalese level.52 The effect of all of this on government employment has been serious; from 1956 onward, the Tamil position in the public service has declined. Still, Tamils initially held their own in professional and technical categories.53 Standardization of marks, however, had a profound impact on Tamil representation in the professions of engineering, dentistry, agriculture, and medicine. The Tamils have found themselves in a position similar to that of the Ibo, albeit to a lesser degree. The repeated failure of the majority to acknowledge the Tamil position in the country, the steady contraction of opportunities for Tamils in the South, and occasional violence have all contributed to a growing Tamil willingness to forgo opportunities in an undivided state if those opportunities could be exchanged for expanded opportunities in a smaller, sovereign Tamil state. For most Tamils, this


204

willingness remains equivocal, partly because many Tamils remain in the South. Like other advanced groups in backward regions, the Tamils are still reluctant secessionists.

Advanced Groups in Advanced Regions As indicated previously, the vast majority of secessionist regions is economically backward. Advanced regions are far less inclined to separatism. But just as backward and advanced groups in backward regions have different reasons for choosing a separatist course, so, too, do the paths traversed by groups inhabiting advanced regions differ from each other. The calculations of advanced groups in advanced regions are easy to fathom. In the nature of things, these groups are likely to have a regional economic grievance. Advanced regions usually generate more income and contribute more revenue to the treasury of the undivided state than they receive in benefits. They believe that they are subsidizing poorer regions. The Basque and Catalan cases in Spain are extreme examples of this. These are industrialized provinces, with per capita incomes far above the national average—in the Basque case, more than twice that average. In the late 1960s, Catalonia paid 31 percent of all of Spain’s taxes, but received only 13 percent of all expenditures. The Basque region paid 13 percent of all taxes but obtained only 5 percent of all expenditures.54 Myths have grown up in the Basque country that the hard-working Basques are supporting less productive peoples and regions of Spain.55 A Basque protest song characterizes Spain as “a cow with its muzzle in the Basque country and its udder in Madrid.”56 From this standpoint, separatism would permit productive regions like the Basque country to retain their revenues and to control and limit migrants from other regions who are attracted to advanced industrial regions because of economic opportunities there. If this were all there were to it, there would be many more separatist advanced regions than there are. There are, however, countervailing considerations that stem the growth of secessionist activity among advanced groups in advanced regions. To begin with, such groups are likely to export surplus capital and population outside their region. Their prosperity generates capital for investment that does not respect regional boundaries. Their education creates a talent pool in search of opportunities. Like the Ibo, but less from necessity than from opportunity, the Yoruba had sent their sons all over Nigeria, and particularly to the North, where they were engaged in business and in government service. When the Ibo created Biafra, the Yoruba did not follow suit. There are several reasons for this,57 but surely one of the most prominent is the fact that the Yoruba were well positioned, by dint of qualification and seniority, to move into opportunities in Nigeria vacated by the Ibo. And this they did. In Uganda, the Baganda were vastly overrepresented in the civil service during the 1960s.58 They also were overrepresented in business and the professions, and they had a long history of taking up opportunities all over the country. This favorable position no doubt had much to do with overcoming earlier Baganda reluctance to joining an independent Uganda. Ultimately, the Baganda ruler did threaten secession, but only after the prime minister, A.Milton Obote, had reneged on the independence agreement, removed the ruler as head of state, and


205

forced through a new constitution.59 For the Yoruba and the Baganda, the attractions of exerting influence and reaping rewards in a large, undivided state were stronger than the temptations of a more homogeneous, contracted homeland. The effect of investment and employment not tied to the home region is to create among advanced groups from advanced regions outward-looking interests that retard their enthusiasm for secession. There is something beyond this that is not present in the situation of advanced groups in backward regions, such as the Ibo. The economic development of advanced regions almost inevitably leads to claims of revenue imbalance of the sort described earlier, but these complaints may mask the enormous economic advantages to such regions that inhere in the undivided state. If the advanced region produces for the domestic market of the undivided state, it is not certain that regional prosperity will survive separation. Some 90 percent of the production of the Basque provinces, for example, is purchased within Spain under a protectionist economic policy in aid of Basque products that would not be competitive on the international market.60 Once the Ibo returned home, the economic interest of the Eastern Region in the undivided state was practically at an end. But this would not be true for the Basques—or for other, similarly situated advanced groups in advanced regions—even if all group members were to return at once to the home region. The prosperity of advanced groups in advanced regions typically depends not merely on the contributions of migrant sons located in other regions, but on a web of interregional economic relations that may include dependence on other regions for materials and markets, sometimes specially protected markets.61 The apparent economic advantages of secession must thus contend with the circumstance that the favorable economic position of advanced regions may not be wholly attributable to factors of production under the exclusive control of those regions. Like advanced groups in backward regions, advanced groups in advanced regions will secede only if the economic costs of secession are low, but the reduction of such costs is far less likely in the case of advanced groups in advanced regions. This circumstance is reflected in the Basque ambivalence toward secession. The Basque movement is strong but far from unanimous on its goals. Businessmen and others with far-flung interests, or with doubts about the ability of the region to survive an end to protection, have tended to oppose a separate state.62 Noting that Basque industry has always thought in terms of the broader Spanish economy, Stanley Payne has opined that Basque separatism is “shrill and fanatical” partly “because of its minority position” in the Basque country.63 Nevertheless, the Basques have experienced some special conditions conducive to separatism. The Basques have relatively fewer group members outside their home region than most other similarly situated advanced groups do, though exact figures are not available. In the past, many who left the Basque country became Castilianized. The Basques have also faced an enormous influx of immigrants from other areas of Spain. These conditions are likely to foster separatism regardless of the backward or advanced character of the region or group.64 Moreover, these conditions are especially acute among the Basques. So, too, was the repression of the Franco regime, a regime that had remarkably few Basques in its public service and that carried out its repression largely through the medium of the Guardia Civil, a military body composed of ethnic strangers to the Basque country.


206

Thus, despite the widespread distribution of their interests within Spain, the Basques had special reasons to regard the Spanish state as offering more penalties than opportunities. What the Basque case shows is that the growth of a powerful secessionist movement is not impossible among advanced groups in advanced regions but that it takes some extraordinary conditions to bring it about. Most of the time, the lure of interests and opportunities throughout the undivided state is enough to ward off the possibility. The point is well illustrated by what might at first appear to be the exceptional case of Eritrea. With about 8 percent of the total population of Ethiopia, Eritreans have had, by some estimates, as much as one fourth of the opportunities in higher education and government service all over the country. Their literacy rate is estimated to be considerably higher than the Ethiopian average.65 Beginning in the 1930s, under Italian rule, Eritrea experienced industrial development that more than compensated over-all for the region’s soil erosion and unreliable rainfall, leaving it with a per capita income higher than the Ethiopian average. Moreover, as Ethiopia’s only access to the sea, Eritrea benefited from the transit trade and from priority in government investment.66 Yet, practically from the moment of federation with Ethiopia in 1952, there was Eritrean resistance, culminating in a full-fledged secessionist war by the 1970s. Is this a case, then, of an advanced group in an advanced region willing to forgo the advantages of the undivided state, including numerous opportunities outside the home region, on scarcely a moment’s reflection? To answer this question, it is necessary to restore some of the complexity that our simplified framework has deliberately omitted. Eritrea is a heterogeneous region, composed of nearly equal numbers of Christians and Muslims. The advanced group that has had opportunities in education and employment, and has migrated out of the region, has been disproportionately Christian.67 For example, 19 of 138 senior central government officials serving between 1941 and 1966 were Eritreans—a total second only to Emperor Haile Selassie’s own Shoan Amharans—but, of the 19, only 3 were Muslims.68 The secessionist movement, although not totally lacking in Christian support, has been overwhelmingly Muslim. Around the time of federation, many Christians supported full integration with Ethiopia—witness the growth of a Unionist Party composed of Christians. When these Muslim-Christian qualifications are introduced, the profile of the Eritrean secessionists no longer resembles that of an advanced group from an advanced region. Rather, it is largely a movement of a backward group from an advanced region. Groups so positioned have, as I shall demonstrate shortly, little reason to equivocate on secession once they detect signs of domination. (One strong sign that we have observed elsewhere was the frequent appointment of outsiders—here, Shoans, rather than Eritreans—to key positions in the region.) Advanced groups from advanced regions, however, are more often inclined to participate actively in the undivided state—and even, if possible, to dominate it—than they are to withdraw from it.

Backward Groups in Advanced Regions There is a different reason for the infrequency of secessionist claims made by backward


207

groups in advanced regions. Such groups are quite likely separatists, but they are rarely in a numerically predominant position in such a region. Economically advanced regions tend to be the home of advanced ethnic groups who have benefited from the economic institutions that bring prosperity to the region. Over and over again, fortuitous location in or near a center of investment has given local ethnic groups opportunities for education and employment denied to those less well situated. Like the Basques, the Yoruba, Kikuyu, and Baganda are all advanced groups indigenous to economically advanced regions. But this is not invariably the case. Sometimes opportunities of this kind are taken up by migrants to an area, typically by advanced groups from backward, population-exporting regions. This has been largely the case in the Sind province of Pakistan, the prosperous urban centers of which are controlled by Urdu-speaking migrants from north India, rather than by Sindhis. Where this occurs, the indigenous population becomes a backward group in an advanced region. Katanga (later renamed Shaba) in Zaire was one such case. Mineral-rich Katanga was sparsely populated…with the result that by the 1920s labor recruiters began going farther afield, notably to what became Kasai Province. The new mining towns, Elisabethville, Jadotville, Kolwezi, and others, began filling up with “strangers” from outside Katanga, predominantly Luba from Kasai, who were particularly receptive to European influences and social change…. In both commercial and clerical jobs in Katangan towns, the Luba/Kasai were markedly more numerous than the Katangans.69 By the late 1950s, Lunda and other indigenous groups were greatly outnumbered by migrants in the towns, especially by Kasai Baluba. In Elisabethville, for example, more than half of those employed came from outside Katanga, and Kasai Baluba outnumbered Lunda by more than four to one, a fact quickly reflected in election results.70 Political organization in Katanga responded to this situation. Moise Tshombe’s party, the Conakat, described itself as a movement of “authentic Katangans,” which was another way of saying it was organized by indigenous Lunda and Bayeke of Southern Katanga and directed against the Kasai Baluba. Conakat had originally had some support from the Baluba of Northern Katanga, which was a group different from the Kasai Baluba, but the support proved short lived. The key reason for the split was Conakat’s hostility to the Kasai Baluba, with whom the Katanga Baluba felt at least some affinity. It took very little beyond the double threat of immigrant Kasai Baluba power in Southern Katanga and Katanga Baluba power in the Northern part of the province to persuade Conakat of the desirability of a separate state. As Zaire neared independence, the Baluba of Northern Katanga seemed to have considerable influence in the central government, and this was enough to push Tshombe over the brink.71 Within the first fortnight of independence, the secession of Katanga was under way. But the secession was effective only in the South of Katanga, where the Lunda and Bayeke are concentrated. In the North of the province, a separate movement developed among the Katanga Baluba. The sentiments and claims expressed by backward groups in advanced regions are a hybrid of those enunciated by backward groups in backward regions and by advanced


208

groups in advanced regions. They are compounded of a substantial dose of collective anxiety and a desire to end revenue-expenditure imbalances. Like backward groups in backward regions, those in advanced regions are fearful of competing with advanced groups and are keenly sensitive to threats of domination. Their fears are magnified by the large numbers of advanced group members in their midst. Backward groups in backward regions can escape disagreeable competition by withdrawing from the undivided state. A backward group in an advanced region—such as the Lunda—must also cope with the advanced group within its own regional borders. To do so, it proposes various discriminatory measures. Secession foreshadows yet more severe and xenophobic action.72 As noted earlier, backward groups in backward regions typically attempt secession despite economic costs. For backward groups in advanced regions, however, secession appears to promise economic benefits. The region typically contributes more to the income of the undivided state than it receives in budget allocations. The contribution of Katanga to Zaire’s total income was close to 50 percent just before independence. But Katanga’s share of Zaire’s budgetary expenditure was only 20 percent, more than its per capita share but, obviously, less than its share based on productivity.73 Just as the Basques likened Spain to a cow being fed by the Basques but milked in Madrid, so in Katanga there was a longstanding slogan, “Katanga, milk cow for the whole Congo.”74 Before independence, Tshombe had demanded that “the resources of each province be properly its own.”75 These are claims characteristic of advanced regions, even expressed in the same bovine imagery. Against the anticipated benefits of secession, backward groups in advanced regions do not need to balance certain costs that trouble the calculations of advanced groups in advanced regions. Unlike advanced groups, backward groups generally do not have widespread interests throughout the undivided state. Unless other regions of the state are industrialized, backward groups are less likely to export population or capital.76 Tariff protection, for the products of an advanced region may be another matter, but three factors are apt to limit or cancel its inhibiting effect on secession. First, there is a difference between industrialized and developing countries. In the less developed countries, the prosperity of advanced regions is less likely to be based on production of finished goods for a protected domestic market than on exports of minerals or primary agricultural products. The latter are very often taxed, rather than subsidized. As tariffs are a disincentive to secession, so centrally imposed taxes are an incentive. Second, this effect is particularly apt to be operative in the case of advanced regions dominated by backward groups, who are likely to be located in areas where primary agricultural products are cultivated or minerals are extracted. Third, even if the economy of the advanced region depends on tariff protection or other benefits derived from membership in the undivided state, the loss of these benefits that accompanies secession, though it hurts the region as a whole, may still be offset by gains that inure to the backward group that controls the region and opts for secession. A regional loss may still be an ethnic gain. Altogether the incentives are heavily weighted toward separation in such a case. Backward groups would like simultaneously to have a free hand to deal with the advanced groups in their midst and to retain the revenues of economic enterprises located in their regions. But, as previously indicated, it is rare that a backward group finds itself


209

in political control of an advanced region. Katanga, once the specter haunting all of Africa, turns out to be an exceptional case.

The Differential Disposition to Secede Table 2 summarizes much of the discussion so far. It makes clear just how much can be deduced from group and regional position. Backward groups tend to measure disadvantage in terms of deviation from some concept of proportionality in relation to population. Advanced groups gauge deprivation by discrimination, utilizing a standard of proportionality in relation to merit. Advanced regions tend to complain of revenueexpenditure imbalances. Backward regions may also complain of inadequate expenditure if they receive from the center less than their per capita share, albeit more than their contribution to revenue. Backward regions that are the home of advanced groups, however, tend not to complain of revenue imbalances, probably because they receive remittances from outside the region and certainly because they eschew claims based on numbers. Here, too, there is more than one criterion of proportionality. The four categories of political claims are, as Table 2 shows, a combined function of group and regional characteristics. These claims do not, however, invariably ripen into secession. The columns headed “Precipitants” and “Calculations” indicate the conditions under which the disposition to secede is likely to emerge. Precipitants tend to be events that have the effect of rejecting unequivocally claims put forward by regional groups. In the case of backward groups, as we have seen, precipitants foreshadow political domination. In the case of advanced groups, precipitants tend to reduce the economic advantages of remaining in the undivided state. In short, precipitants may act either to raise the costs or to reduce the benefits of remaining in the state provided, of course, that benefits and costs are understood to embrace nonmaterial as well as material values. Indeed, Table 2 demonstrates that separatism results from varying mixes of sheer economic interest and group apprehension. Economic interest may act either as an accelerator or a brake on separatism. Yet, among the most frequent and precocious secessionists—backward groups in backward regions economic loss or gain plays the smallest role, ethnic anxiety the largest. The occurrence of the specified precipitating events and calculations that follow them are not inexorable. Claims need not be denied. Civil servants from

Table 2: The Disposition to Secede by Group and Regional Position Group and Political claims Precipitants Calculations Timing region and relative frequency Denial of Secede Early, Backward Proportionality in group in civil service, proportionality in civil despite frequent. backward occasionally also in service; changes economic region revenues. involving symbolic costs.


Advanced group in backward region Advanced group in advanced region Backward group in advanced region

210

issues like language and religion; influx of advanced group civil servants. Nondiscrimination; Severe discrimination; Secede only if Late, no revenue issue. repeated violence; economic somewhat migration back to costs are low. frequent. home region. Nondiscrimination; Severe discrimination; Secede only if Late, rare. spend revenue where violence and economic generated. migration back to costs are low. home region if population exporter. Proportionality in Denial of Secede Early, rare. civil service, spend proportionality; regardless of revenue where political claims made economic generated. by immigrant benefits or strangers in the costs. region.

advanced groups need not be posted to backward regions. Advanced groups from population-exporting regions can be protected from discrimination and violence; they need not migrate home. Much depends on the reception accorded group claims. The conditions that promote a disposition to secede, though derived from group and regional position, are subject to intervention and deflection.77 The list of potential candidates for secession is much longer than the list of actual secessionists. Some Basques in Spain want independence; but the Nigerian Yoruba, who might have chosen to secede, chose not to; and the Baganda, who threatened secession, did not follow through. The Ibo fought a war of secession; but the Lozi, not treated like the Ibo, did not secede; and the Tamils of Sri Lanka might still go either way. Backward groups are frequent secessionists, but the Northerners in Ghana, every bit as backward as Northerners in Nigeria78—and far less powerful—have not even mooted secession. Likewise, the backward Batéké in Southeast Gabon, a region rich in uranium and manganese, have evidenced no serious inclination toward a Katanga-like secession.79 Every category of regional group has its negative cases. Moreover, as I have suggested, there are varying thresholds of secession and therefore differential frequency of secession among the various categories of groups. Clearly, backward groups in backward regions are most easily persuaded that it is in their interests to leave. The same is true for backward groups in advanced regions, but there are many fewer such groups in a position to secede. Despite their generally greater reluctance to secede, there are differences among advanced groups. Advanced groups from advanced regions often receive extraregional benefits that are not confined to remittances from migrant sons and therefore not terminated precipitously if back-migration should occur. They are less likely to secede. As the last column in the table indicates, the four paths to


211

secession are not equally well trodden. The much greater frequency of secessionist movements in backward regions has a number of important implications. Many of the regions that choose secession are likely to be those economically least capable of sustaining themselves. This applies particularly to the secession of backward groups in backward regions, which may also be short on administrative capacity and personnel. However, the position of advanced groups in backward regions is at least equivocal. They will have no shortage of administrative talent once their migrant sons return to the region. But this surfeit of talent may quickly become a drain on the budget. The experience of Biafra and the difficulties of Benin in reabsorbing civil servants it had exported to other West African states both attest to this. No doubt many countries once proclaimed “unviable” have survived. It is all too easy to exaggerate the economic problems a secessionist region will face. Yet there is no gainsaying the fact that a great many regions that do manage to secede can be expected to have postsecession economic difficulties. The distinction between early and late seceders—which, as Table 2 makes clear, is largely coterminous with the distinction between backward and advanced groups—also has important consequences. In general, late secessions are more cohesive, better organized, and more often conducted under the auspices of a political party than are early secessions. Early secessions in countries like Chad, the Sudan, and Burma consisted of more than one movement. The secessionist region was highly heterogeneous, and the secessions occurred so soon after independence that no political party had a chance to capture the support of the entire region. Because it was not centrally organized, the warfare was sporadic, and—except in the Sudan—there was no single organization in a position to make peace. In Chad, for example, an amnesty was accepted by members of one ethnic group fighting in one region but ignored by other groups fighting elsewhere. In all the cases, the fighting lingered on for many years: in Burma, it still does. In the late secessions of Biafra and Bangladesh, by contrast, the movements were under much tighter control. The fighting was more intense, widespread, and simultaneous in all areas; and victory for one side or the other was quicker and more decisive.80 In the case of groups likely to become late seceders, if seceders at all, there is more time for the center to work on policies averting secession and, because of the reluctance to secede, more latitude regarding the actual substance of policies that might prove sufficient to avert secession. There is also, however, more time for both sides to prepare for the battle when it comes: to cement foreign alliances, procure sophisticated weapons, and organize the secessionist region and the rump region for war. This extra time, preparation, and organization are likely to ensure that the resolution of the fighting, when it eventually occurs, will be clearcut.

Notes 1. For reasons to be stated shortly, movements for separate territorial identity will be referred to as “secessionist,” whether the demand is for independence or only for autonomy from a larger territorial unit. I shall also use secession and separation, as well as secessionist, interchangeably throughout. For an equally inclusive conception of separatism, see Joane


2.

3.

4.

5.

6.

7. 8.

9.

10 11.

212

Nagel, “The Conditions of Ethnic Separatism: The Kurds in Turkey, Iran, and Iraq,” Ethnicity, Vol. 7, No. 3 (September 1980), pp. 279–97. It will also be apparent that the materials for this article are drawn from Asia and Africa, but occasionally other cases are referred to in order to supplement these materials. This is particularly true of the Basque case, which exemplifies a pattern not generally found in Asia and Africa. Although it makes no difference to the analysis that follows, I assume for the sake of argument that there has been an upsurge of separatism in the last two decades. That assumption squares with my own observation and with that of most close students of the subject. This is not on any logical grounds. In principle, there is no reason why conditions responsible for an upsurge in separatist activity could not also differentiate between positive and negative cases. That is, a variable accounting for an increase in separatism might tend to be present in the positive cases, absent in the negative ones. In practice, however, variables adduced to explain aggregate trends tend to be so widely diffused that they are present in both kinds of cases, as shown in the text. The relationship of the idea of self-determination to secession is admirably traced by Walker Connor, “Self-Determination: The New Phase,” World Politics, Vol. 20, No. 1 (October 1967), 30–53. See also Tamotsu Shibutani and Kian M.Kwan, Ethnic Stratification (New York: Macmillan, 1965), pp 444–45. For some of the philosophical tools of national selfdetermination, see Anne M.Cohler, Rousseau and Nationalism (New York: Basic Books, 1970). Connor, “Self-Determination: The New Phase,” p. 30. Harold Isaacs, Idols of the Tribe: Group Identity and Political Change (New York: Harper & Row, 1975), p. 13, puts the point graphically: The holy grail of self-determination in anti-colonial politics became the poison potion of group conflict, secession, rebellion, and repression in the postcolonial era. Cf.Connor, “Self-Determination: The New Phase,” p. 46: Walker Connor, “Ethnology and the Peace of South Asia,” World Politics, Vol. 22, No. I (October 1969), 70; Isaacs, Idols of the Tribe, pp. 175–78. Anthony H.Birch, “Minority Nationalist Movements and Theories of Political Integration,” World Politics, Vol. 30, No. 3 (April 1978), 325–44. Ibid., p. 335. Birch suggests that the existence of the American nuclear umbrella in Europe makes it less necessary than previously to secure protection against external attack by remaining part of a large, strong stale. The Netherlands is protected just as France and Germany are. The Common Market, he argues, likewise makes it possible for groups like the Corsicans or the Bretons to salvage their prosperity even if they opt for independence. An independent Corsica, for example, could become a voting member of the European Economic community and receive all of its benefits. Both of these conditions constitute an alteration of the earlier balance of advantages. The particular changes identified by Birch, of course, apply only to Europe. In Asia and Africa, presumably, it remains advantageous in external security terms for groups to belong to large states. Likewise, though the economic advantages of large states are subject to debate, no recent change in the balance of those advantages can be detected for Asia and Africa. Even within Europe, however, there are wide variations in the enthusiasm of potential separatist groups for the prospect of separatism. Or, in a large federation like India, by forces beyond the affected state or province. “No one in Baluchistan wants to break away [from Pakistan]. All the Baluchis want is not to lose their identity.” commented a Baluch spokesman. “Who in his right mind would want to join Afghanistan? We’d be worse off there than we are in Pakistan.” Washington Post,


12. 13. 14.

15. 16.

17.

18.

19. 20.

213

February 8, 1976, p. A17. This is also the theme of Khalid B.Sayeed, “Pathan Regionalism,” South Atlantic Quarterly, Vol. 63, No. 4 (Autumn 1964), 478–506. Far Eastern Economic Review (Hong Kong), August 17, 1979, p. 28. Ibid., September 14, 1979, p. 30. See Keith Kyle, “The Southern Problem in the Sudan,” The World Today, Vol. 22, No. 12 (December 1966), pp. 512–20. An article in a journal published by the Southern Sudan Association in Britain during the Sudanese civil war illustrates the point: “What, then, are we fighting for? We are fighting for freedom: freedom to unite with the North; freedom to federate with the North; freedom to reject the North; freedom for the people of the South Sudan to determine their own future without interference from the Arabs or any other people.” Jacob J.Akol, “What We Are, and Are Non Fighting For,” The Grass Curtain (London), Vol. 2, No. 2 (October 1971), p. 26. Africa: The Politics of Independence (New York: Vintage Books, 1961), p. 88. I am well aware that this terminology is less than optimal, but alternatives (e.g., traditionalmodern, disadvantaged-advantaged) seem even less apt. Of course, the term backward is not used in a pejorative way, and its frequently pejorative connotations among observers are offset for our purposes by the fact that the groups concerned very often use the term in an objective sense, referring to themselves as backward and to their antagonists as advanced. There are some variations from country to country in the pattern of group stereotypes though not in persuasiveness or in the general tendency to dichotomize. For example, most advanced groups are regarded as progressive and backward groups as traditional, but in the Philippines the “traditional” label is applied to the Ilocano, who score high on virtually all objective and subjective indices of group advancement. Rodolfo A.Bulatao, Ethnic Attitudes in Five Philippine Cities (Quezon City: University of the Philippines Social Research Laboratory, 1973), p. 58. It would take us far afield here to examine the origin, content, and correlates of such stereotypes. For a recent study, see Marilynn B.Brewer and Donald T. Campbell, Ethnocentrism and Intergroup Attitudes: East African Evidence (New York Halsted Press, 1976). One complexity of which I shall not take adequate account concerns differences of opinion within given ethnic groups on the advisability of secession. Sometimes secessionist sentiment is virtually unanimous, but very often there are debates on whether to secede. See, e.g. S.J.Dudley, “Western Nigeria and the Nigerian Crisis,” in S.K.Panter-Brick, ed., Nigerian Politics and Military Rule: Prelude to the Civil War (London: Athlone Press, 1970), pp. 106–08, identifying al least five Yoruba opinion strains ca 1966–67. More often than not, I shall ignore such differences, dealing instead with central tendencies or merely with the outcomes of such debates. Jeffrey G.Williamson, “Regional Inequality and the Process of National Development,” Economic Development and Cultural Change, Vol. 13, No 4, Pt. 2 (July 1965), 14, 17. See Charles Anderson, Fred R.von der Mehden, and Crawford Young, Issues of Political Development (2d ed.; Englewood Cliffs: Prentice-Hall, 1974), p. 75. When the Uganda governments refused to accede to demands for a separate district, the Rwenzururu movement in Western Uganda became secessionist very quickly, for reasons that are probably related to span of control. “The Ruwenzori mountain areas are extremely inaccessible; effective administration had never become established in the higher attitudes and, in a sense, anybody could set up an independent government there without facing the consequences for some time,” Martin R.Doornbos, “Protest Movements in Western Uganda: Some Parallels and Contrasts,” unpublished paper, 1966, p. 12. But the emergence Of the movement in the first place had little to do with these geographic conditions.


21.

22.

23.

24.

25.

26.

27.

214

Distance, of course, is a condition that can cut both ways. While great distance may make secession easier—or at least make its suppression more difficult—distance may also reduce the intrusiveness of central government penetration of peripheral areas. See, e.g., Ba Maw, Breakthrough in Burma: Memoirs of a Revolution 1939–1946 (New Haven: Yale University Press, 1968), p. 187; Fred R.von der Mehden, Religion and Nationalism in Southeast Asia (Madison: University of Wisconsin Press, 1963), p. 193; Bulatao, Ethnic Attitudes in Five Philippine Cities, pp. 57–62; Hugh Gray, “The Demand for a Separate Telengana State in India.” Asian Survey, Vol. 11, No. 5 (May 1971), 463–74, esp. 464. In wealthier states, however, even this subsidy may not be enough to bring per capita spending in poor regions up to levels proportionate to their share of the state’s population. In such a case, a demand for per capita proportionate spending is likely to be received most unsympathetically by the center. See, e.g., Charles M.Benjamin, “The Kurdish Non-State Nation,” unpublished paper presented at the annual meeting of the International Studies Association, February 19–22, 1975, p. 6. See, e.g., von der Mehden, Religion and Nationalism in Southeast Asia, p. 171; Rasheeduddin Khan, “Political Participation and Political Change in Andhra Pradesh (India)” (Hyderabad: Osmania University Department of Political Science, mimeographed paper, June 969), p. 33. In some cases, even elites that were ahead of ethnically-differentiated competitors saw separatism as a way of reducing the competition. The agitation for a separate Pakistan in the 1930s and 1940s was disproportionately led by Muslims in what was then called the United Provinces. As a whole, Indian Muslims were backward, and they feared domination by educationally more advanced Hindus. But, in the United Provinces, Muslims were ahead of Hindus in government employment, the professions, and the modern private sector. Still, U. P.Muslim elites feared their minority position in an undivided India, and they demanded a separate state to protect their position. Paul R.Brass, “Muslim Separatism in United Provinces: Social Context and Political Strategy Before Partition,” Economic and Political Weekly (Bombay), Vol. 5, Nos. 3, 4, and 5 (January 1970), 167–86. Lorenzo Kent Kimball, The Changing Pattern of Political Power in Iraq, 1958 to 1971 (New York: Robert Speller & Sons, 1972), pp. 141–42; Abdul H.Raoof, “Kurdish Ethnic Nationalism and Political Development in Republican Iraq,” unpublished paper presented at the annual meeting of the Middle East Studies Association, Denver, November 11–13, 1971, p. 10. So conscious are the Kurds of their backwardness that these demands sometimes make provision for exceptions when no qualified Kurds can be found for particular positions. Mohamed Omer Beshir, The Southern Sudan: Background to Conflict (London: C. Hurst & Co., 1968), p. 72; Richard Gray, “The Southern Sudan,” Journal of Contemporary History, Vol. 6, No. 1 (1971), p. 117. Joseph Oduho and William Deng, The Problem of the Southern Sudan (London: Oxford University Press, 1963), p. 14. For reports of similar sentiments, see Kyle, “The Southern Problem in the Sudan,” p. 513:1. William Zartman, Government and Politics in Northern Africa (New York: Praeger, 1963), p. 140; Robert O.Collins and Robert L. Tignor, Egypt and the Sudan (Englewood Cliffs: Prentice-Hall, 1967), pp. 147–64. Robert Pledge, “France at war in Africa,” Africa Report, June 1970. pp. 16–19; John A.Ballard, “Four Equatorial States,” in Gwendolen M.Carter, ed., National Unity and Regionalism in Eight African States (Ithaca: Cornell University Press, 1966). pp. 272–74 William H.Lewis, “Francophone Africa,” Current History, Vol. 60, No 335 (March 1971) 143: René Lemarchand, “Sisyphus In Chad: The MRA as a Development Partnership,”


215

unpublished paper, n.d., ca. 1974, p. 4: Africa Report, November 1969, pp. 10–12. 28. Neville Maxwell, India and the Nagas (London: Minority Rights Group Report No. 17, n.d), p. 9. 29. Hugh Gray, “The Demand for a Separate Telengana State,” Asian Survey, Vol. 11, No. 5, (May 1971), 463–74. 30. See Williamson, “Regional Inequality and the Process of National Development.” 31. C.P.Cook, “India: The Crisis in Assam.” The World Today, Vol. 24, No 10 (October 1968), p. 446: Collins and Tignor, Egypt and the Sudan, p. 159; Donald Eugene Smith, Religion and Politics in Burma (Princeton: Princeton University Press, 1966), p. 230; Zartman, Government and Politics in Northern Africa, p. 140: George E.Kirk, Contemporary Arab Politics (New York: Praeger, 1961), p. 145. 32. Quoted in Walter Schwarz, Nigeria (New York: Praeger, 1968). p. 249. For an explanation, cast in terms of welfare economics, of “why even individuals who will probably lose in terms of tangible rewards through increased political autonomy may nevertheless be willing to invest in its attainment,” see Douglas G.Hartle and Richard M. Hird, “The Demand for Local Political Autonomy: An Individualistic Theory,” Journal of Conflict Resolution, Vol. 15. No. 4 (December 1971), 443–56. 33. Richard D.Lambert, “Factors in Bengali Regionalism in Pakistan,” Far Eastern Survey, Vol. 28. No. 4 (April 1959), p. 54. Hugh Tinker, India and Pakistan: A Political Analysis (rev. ed.; New York: Praeger, 1968), p. 167, reports that none of Pakistan’s share of the ICS was Bengali. 34. Crawford Young, The Politics of Cultural Pluralism (Madison: University of Wisconsin Press, 1976), p. 482. 35. The Biafrans suggested that influential foreigners (presumably the British) discouraged the idea of a Northern secession, Nigerian Crisis 1966 (Enugu: Eastern Regional Government, n.d.), pp 41–42. 49, Ruth First, Power in Africa (Harmondsworth Penguin Books, 1970), p. 320, attributes the suppression of secessionist sentiment to a group of Northern civil servants, British and American diplomats, and Middle Belters in the army “who saw in Northern secession the danger that they would be a perpetual and unbearable minority in the North.” 36. This is not invariably so; before Independence, the Ceylon Tamils, for example, sought additional parliamentary representation to compensate for their numerical weakness. See W.Howard Wriggins, Ceylon: Dilemmas of a New Nation (Princeton: Princeton University Press, 1960). p. 41. 37. Gerald L.Caplin, The Elites of Barotseland. 1878–1969 (London C.Hurst & Co., 1970), Chaps. 6–8; Caplin, “Barotseland: The Secessionist Challenge to Zambia,” Journal of Modern African Studies, Vol. 6, No. 3 (October 1968), 343–60; Margaret Rouse Bates, UNIP in Post Independence Zambia (Ph.D.diss., Harvard University, 1971). 38. This fits economic expectations quite well. See Williamson, “Regional Inequality and the Process of National Development,” pp. 5–6. 39. Caplin, The Elites of Barotseland, pp. 175–76. 40. Wriggins, Ceylon: Dilemmas of a New Nation, p. 234. 41. James A.Coleman, Nigeria: Background to Nationalism (Berkeley & Los Angeles: University of California Press, 1958), pp. 332–34. 42. Ibid., pp. 338–39. 43. Caplin, The Elites of Barotseland, p. 194. 44. Wriggins, Ceylon: Dilemmas of a New Nation, p. 146. 45. Brian Crozier, The Rebels: A Study of Post-War Insurrections (Boston Beacon Press 1960),


216

p. 89. 46. This is not an inflexible rule, or course, but it does hold for large parts of the developing world, especially less industrialized countries. Often the migration such groups undertake is temporary or seasonal, as in the case of agricultural labor. That tendency is altered, however, as industrialization proceeds, creating a need for large, unskilled and semi-skilled labor forces. In Spain, for instance, the poor Southern region of Andalusia exports much unskilled labor to Northern industry. The distribution of the population of backward groups may be a major difference—with implications for secession of backward regions—between developing and developed countries. 47. Thomas Turner, “Congo-Kinshasa,” in Victor A.Olorunsola, ed., The Politics of Cultural Sub-Nationalism in Africa (Garden City: Anchor Books, 1972), pp 217–24. 48. Richard L.Sklar, Nigerian Political Parties: Power in an Emergent African Nation (Princeton: Princeton University Press, 1963), pp. 327–28. 49. K.Whiteman, “Enugu: The Psychology of Secession, 20 July 1966 to 30 May 1967,” in Panter-Brick, ed., Nigerian Politics and Military Rule: Prelude to the Civil War, p. 117; Young, The Politics of Cultural Pluralism, pp. 467–68: Victor A.Olorunsola, “Nigeria,” in Olorunsola, ed., The Politics of Cultural Sub-Nationalism in Africa, pp. 35–36. 50. Robert N.Kearney, “Language and the Rise of Tamil Separatism in Sri Lanka,” Asian Survey, Vol., 18, No. 5 (May 1978), 521–34. See also Urmila Phadnis, “Keeping the Tamils Internal,” Far Eastern Economic Review, 25 March 1972, 21–22: W.A.Wiswa Warnapala, “Sri Lanka in 1972: Tension and Change,” Asian Survey, Vol. 13, No. 2 (February 1973), 217–30. 51. Kearney, “Language and the Rise of Tamil Separatism in Sri Lanka,” p. 531: Walter Schwarz, The Tamils of Sri Lanka (London: Minority Rights Group Report No 25, 1975). pp. 12–13. 52. Report on the Survey of Consumer Finances, Pt. 1 (Colombo: Central Bank of Ceylon. 1974), p. 32. 53. See Donald L.Horowitz, Coup Theories and Officers’ Motives: Sri Lanka in Comparative Perspective (Princeton: Princeton University Press, 1980), Tables 2–1, 2–2. 54. William T.Salisbury, “Some Aspects of the Regional Issue in Contemporary Spanish Affairs,” unpublished paper presented at the annual meeting of the International Studies Association, Toronto, February 26,1976, p. 6, drawing on data developed by Juan Linz. 55. For these resentments, see Kenneth Medhurst, The Basques (London: Minority Rights Group Report No. 9,1972), p. 5; William A.Douglas and Milton da Silva, “Basque Nationalism,” in Oriol Pi-Sunyer, ed., The Limits of Integration (Amherst: University of Massachusetts Department of Anthropology, Research Report no. 9, 1971). p. 149. 56. Douglas and da Silva, “Basque Nationalism,” pp. 149–50. 57. For a discussion of some of these reasons, see Dudley, “Western Nigeria and the Nigerian Crisis,” p. 109. 58. Nelson Kasfir, “Cultural Sub-Nationalism in Uganda,” in Olorunsola, The Politics of Cultural Sub-Nationalism in Africa, pp. 123–28. 59. On the Baganda, see Young, The Politics of Cultural Pluralism, pp. 149–56. 60. See Pedro Gonzalez Blasco, “Modern Nationalism in old Nations as a consequence of Earlier State-Building: The case of Basque-Spain,” in Wendell Bell and Walter Freeman, eds., Ethnicity and Nation-Building (Beverly Hills: Sage Publications, 1974), p. 347. 61. So far as tariff protection is concerned, much is likely to depend on whether the prosperity of the region is based on production of finished goods for the domestic market or on production of primary products or extraction of minerals for export. For elaboration, see Table 2.


217

62. Gonzalez Blasco, “Modern Nationalism in old Nations as a consequence of Earlier StateBuilding: The case of Basque-Spain,” in Wendell Bell and Walter Freeman, eds., Ethnicity and Nation-Building, p. 366. 63. “Catalan and Basque Nationalism,” Journal of Contemporary History, Vol. 6, No. 1 (1971), 15–51, esp. 50. The reference to Spanish industry appears on 38. 64. See Donald L.Horowitz, “Cultural Movements and Ethnic Change,” The Annals of the American Academy of Political and Social Science, Vol. 433 (September 1977), 6–18. 65. John Franklin Campbell, “Background to the Eritrean conflict,” Africa Report, May 1971, pp. 19–20. Campbell’s estimate is three to four times the Ethiopian average. Asmara, the Eritrean capital, had a literacy rate of 50 percent in the early 1970s, higher than any other area of Ethiopia, including Addis Ababa, which had a 43 percent rate Provisional Military Government of Ethiopia, Central Statistical Office, Population and Housing Characteristics of Asmara (Addis Ababa Central Statistical Office, Statistical Bulletin no. 12, December 1974), p 5. 66. Ethiopiawi (pseud.), “The Eritrean-Ethiopian Conflict,” in Astri Suhrke and Lela Garner Noble, eds., Ethnic Conflict in International Relations (New York: Praeger, 1977), p. 131. 67. There are many indirect indications of this. Asmara, by far the largest city of the region, is 85 percent Christian. Literacy in those districts of Asmara with the heaviest Muslim concentration (Akria and Geza Berhano) was below the average for the city, and housing in those districts was also of less than average quality. Provisional Military Government, Central Statistical Office, Population and Housing Characteristics of Asmara, pp. 68–69, 76–78. 68. Christopher Clapham, Haile Salassie’s Government (New York: Praeger, 1969), pp. 75–76, 83. 69. Turner, “Congo-Kinshasa,” in Olorunsola, p. 224. 70. Ibid., p. 226; Jules Gérard-Libois, Katanga Secession, Rebecca Young, trans. (Madison: University of Wisconsin Press, 1966), pp. 12–13, 27–28; René Lemarchand, Political Awakening in the Belgian Congo (Berkeley & Los Angeles: University of California Press, 1964), pp. 235–36, 241; Crawford Young, ‘The Politics of Separatism: Katanga, 1960–63,” in Gwendolen M.Carter, ed., Politics in Africa: 7 Cases (New York: Harcourt, Brace & World, 1966), pp. 172–74; René Lemarchand, “Congo (Leopoldvilie),” in James S.Coleman and Carl G.Rosberg, Jr., eds., Political Parties and National Integration in Tropical Africa (Berkeley & Los Angeles: University of California Press, 1964), p. 581. 71. In the 1960 Katanga provincial elections, Conakat did well. It also did better than Balubakat, the party of the Katanga Baluba in the North of the province, in the national elections, but the Balubakat leader, Jason Sendwe, was nonetheless named by the central government to be High Commissioner for Katanga. This appointment triggered the Katanga secession. Turner, “Congo-Kinshasa,” in Olorunsola, The Politics of Cultural Sub-Nationalism in Africa, p. 227. I leave aside the role of Katanga’s European settlers in supporting the secession. For the settlers’ role, see Gérard-Libois, Katanga Secession. Lemarchand’s judgment, which seems well-supported by the evidence, is that the settlers’ secessionist “dispositions could not have led to the secession of the province unless they were shared and abetted by a substantial segment of the African population.” Political Awakening in the Belgian Congo, p. 233. 72. In Katanga, this meant, concretely, the exclusion of Kasai Baluba from political and administrative positions and the expulsion of many from the province Lemarchand, Political Awakening in the Belgian Congo, p. 239: Gérard-Libois, Katanga Secession, p. 28. 73. Gérard-Libois, Katanga Secession, pp. 3–5.


218

74. Ibid., p. 51. 75. Ibid., p. 41. 76. See n. 46, above. If anything, a backward group in an advanced region may have the opposite problem: a population so compact it does not cover the whole region. If so, the secession will be of a limited area, as the Katanga secession was and as the Rwenzururu movement of Western Uganda was confined to the areas inhabited by Baamba and Bakonjo, Doornbos, “Protest Movements in Western Uganda: Some Parallels and Contrasts;” Kasfir, “Cultural Sub-Nationalism in Uganda,” 99. In both cases, these happened to be areas within which major mineral resources were also located. Eritrean Muslims are also confined largely to lowland areas. 77. It would be a mistake, however, to minimize the policy dilemmas created by the coexistence in a single state of various kinds of regional groups. Occasionally, the claims are in direct opposition to each other, so that when can salve the apprehensions of one group will simultaneously precipitate secessionist action by another. 78. See Philip J.Foster, “Ethnicity and the Schools in Ghana.” Comparative Education Review, Vol. 6, No. 2 (October 1962), pp. 127–35. 79. Brian Weinstein, Gabon: Nation-Building on the Ogooué (Cambridge: M.I.T. Press, 1966). pp. 220–25. The Batéké case, however, may fall within a caveat stated earlier regarding Pakistani, Baluch and Pathan reluctance to do anything that might link them with Afghanistan. There is a similar reluctance among Batéké to do anything that might result in their annexation by neighboring Congo (Brazzaville). 80. All else being equal, there will also be more subgroup amalgamation among advanced groups than among backward groups; see Donald L.Horowitz, “Ethnic Identity,” in Nathan Glazer and Daniel P.Moynihan, editors, Ethnicity: Theory and Experience (Cambridge: Harvard University Press, 1975), pp. 111–40. On these grounds, too, late seceders will be more cohesive, their fighting forces less likely to fight with each other than with the forces of the rump state.


219


SECTION THREE: Liberalisation and Democratisation


Commentary One of the great modern works of social science has been Barrington Moore’s The Social Origins of Dictatorship and Democracy.1 The articles in this section variously support, re-consider and complement Moore’s initial analysis. J.Stephens’s ‘Democratic transition and breakdown in Western Europe, 1870–1939’ examines all the main countries of Western Europe in order to test whether Moore’s arguments—which are very carefully specified—in fact gain empirical support. One variable of Moore’s does seem to be extraordinarily powerful: when landholding is concentrated and labour repressive, it is both unlikely that democracy will emerge and likely that any breakthrough to democracy will be short lived. In contrast, wide diffusion of land-holding, as in Scandinavia, is related to the consolidation of democratic rule. But Stephens interestingly notes that Moore has slightly too favourable a view of the bourgeoisie: this class has no essential character favouring freedom and justice, and may ally with authoritarian rulers if they feel threatened from below. This is empirically true, as is Stephens’s contention that workers often play a vital role in pushing for democracy. One point that Stephens makes concerns the need for a theory of the state in the late industrialising countries of Europe. A.Janos’s The Politics of Backwardness in Continental Europe, 1780–1945’ supplies such a theory by contrasting core, semiperipheral and peripheral states in Europe as a whole, that is, in Europe both East and West. One of the most striking findings of his article is that the size of peripheral states was proportionately larger than those of the advanced core. The fact that state service was an avenue of social mobility more important than industry or commerce does much to explain the backwardness of such states. It is worth noting additionally that Janos goes some way beyond Stephens in recognizing that the working class no more has an essential character than does the bourgeoisie: whilst workers have much more uniformly favoured democracy, examples cited from Eastern Europe show that they have not uniformly done so. A.Przeworski’s ‘Social Democracy as a Historical Phenomenon’ also concentrates on the classical period of European transitions to democracy, and it does so by examining the characteristic choices that social democrats as to whether to participate and improve capitalism rather than to hold back and seek to transcend it. Przeworski argues convincingly that the social democratic option necessarily rules out genuine transformation—which is, correctly, to say that the much vaunted Swedish Model is a way of running capitalism rather than an alternative to it. D.Rustow’s Transitions to Democracy’ is the earliest attempt to theorise in general terms about transitions to democracy in general, both with Europe and elsewhere, and it remains highly distinguished within a field that is becoming ever more powerful. Rustow begins with sound methodological remarks about the necessity for a genetic account of transformation processes. It is not enough to specify social pre-requisites since these are sometimes ignored; what is necessary is an account of the politics of the transition. To that end, Rustow offers us concepts—the need for national unity, the importance of pacts, the need to combine genuine quarrel with understanding as to its limits—that retain their

usefulness today. T.Karl’s ‘Dilemmas of Democratisation in Latin America’ is best seen as a continuation of Rustow’s concerns, or even as an application of his concepts. This empirically informed piece of theory places especial emphasis on the importance of pacts, noting that no successful transition, in the absence of reconstructions at the end of wars, has yet taken place without an accommodation being found for the ruling élite. Karl also brings the methodological debates as to how to study democratisation up to date, and she shows exemplary judgement in finding middle ground between structural variables and political processes. 1. Boston: Beacon Press, 1966.


224

49 Democratic Transition and Breakdown in Western Europe, 1870–1939: A Test of the Moore Thesis*1 John D.Stephens

* Source: American Journal of Sociology, 1989, vol. 94, pp. 1019–1077.

Barrington Moore’s Social Origins of Dictatorship and Democracy is widely regarded as a contemporary classic, yet there have been few attempts to evaluate the validity of his argument on a large number of comparable cases. This article makes such an attempt with all Western European countries experiencing democratic rule between 1870 and 1939. It seeks (1) to explain what structural and historical features distinguish the breakdown cases from those that remained democratic, and (2) to trace the process of class coalition formation in the transition to democracy and the subsequent breakdown. Moore’s thesis does fit, with some modification. All four breakdown cases were characterized historically by an authoritarian coalition of labor-repressive landlords, the state, and the bourgeoisie that contributed to the breakdown of democratic rule in the 1920s and 1930s. In contrast, in none of the democratic survivors did such a coalition materialize. This difference can be traced largely to the strength of the agrarian elite in the late 19th century. However, in contrast to Moore’s characterization of the conservative authoritarian path, the ruling coalition, except in Germany, did not play a modernizing role, and only in Austria can the bourgeoisie be described as a “dependent” partner in the coalition. Moreover, because it stops too early, Moore’s analysis greatly underplays the role of the organized working class in the transition to democracy and attributes far too positive a role to the bourgeoisie. Without a doubt Barrington Moore’s Social Origins of Dictatorship and Democracy has become a contemporary classic and is one of the most widely read and cited books in social science. Yet there have been only a few studies that attempt to apply his arguments to other cases, and these have been limited to case studies or the comparison of a very small number of cases. What has been absent in the discussion of Moore’s work has been any attempt to test the validity of his conclusions about the social origins of various political regimes on a large number of comparable cases to provide a “test” of the thesis, to use the language of multivariate data analysis. My purpose here is to provide such a “test,” or better said, a further specification of Moore’s thesis on the social origins of democracy and modern authoritarianism, limiting the analysis to the historical development of democracy and the collapse of democratic regimes in Western Europe between the end of the Franco-Prussian War and the outbreak of World War II. I will extend the Moore thesis in three ways. First, he limits himself not

Democratic Transition and Breakdown in Western Europe, 1870–1939

225

only to the larger countries but even among those to three cases in Western Europe (Germany, France, and England), whereas my analysis will include all countries in Western Europe that experienced some period of democratic rule, defined as the existence of universal male2 suffrage, cabinet responsibility to parliament (or a president elected by universal franchise), and the prevention of voter intimidation (e.g., freedom of organization, secret ballot, etc.). Moore excludes Italy and Spain entirely for reasons of economy but at several points indicates that his analysis of fascism fits these cases, with one modification. For the small countries, he argues that the “decisive causes of their politics lie outside their own boundaries” (Moore 1966, p. xiii). In contrast, Rokkan (1970) and Katzenstein (1985) contend not only that this is an exaggeration but also that the small Western European countries have special characteristics that make them an essential point of comparison in the study of the political dynamics of all of the countries in the region. Second, Moore ends his discussion with the events of the mid-19th century, well before democracy was introduced anywhere in Europe. Here the focus will be on the events of the following three-quarters of a century, the events that Moore ultimately claims to explain. Third, I will also consider some alternative explanations of the social conditions leading to democracy proposed by Lipset (1960), Therborn (1977), and Rueschemeyer (1980).

The Moore Thesis In The Social Origins of Dictatorship and Democracy, Barrington Moore attempts to “explain the varied political roles played by the landed upper classes and the peasantry in the transformation from agrarian societies…to modern industrial ones. Somewhat more specifically, it is an attempt to discover the range of historical conditions under which either or both of these rural groups have become important forces behind the emergence of Western parliamentary versions of democracy, and dictatorships of the right or the left, that is, fascist and communist regimes” (1966, p. viii). On the basis of historical case studies of six countries (England, France, the United States, Japan, India, and China) and extensive research on two more (Germany and Russia), Moore identifies three distinct sets of conditions that contribute to the development of communism, fascism, or parliamentary democracy. The condition for the development of a peasant revolution leading to communism is the existence of a weak bourgeoisie, a powerful agrarian elite, a highly centralized state, and high peasant revolutionary potential owing to increased traditional forms of exploitation in noncommercialized agriculture, the existence of solidaristic peasant communities, and weak ties to the (often absentee) landlords. This path will not concern us here since communism did not develop in any of the cases covered in this study. In clarifying Moore’s theoretical argument on the contrasting paths to modern capitalist dictatorship and capitalist democracy, it is simplest to outline the path to authoritarianism because there are multiple paths to democracy. Once a country passes an initial stage of industrialization and has avoided peasant revolution, it will develop in a democratic direction if it lacks any of the essential characteristics leading to


226

authoritarianism. The critical condition for the development of fascism is the development of a coalition of large landholders, the crown (the monarch, bureaucracy, and military—i.e., the state), and a politically dependent bourgeoisie of medium strength. The following factors lead to the development of this coalition: 1. The landed upper classes must be strong, or, more precisely, they must be the politically dominant force into the modern era (i.e., late 19th century), and must retain a significant amount of that power in a “democratic interlude.” 2. The maintenance of peasant agriculture under landlords oriented to the market but employing political rather than market control of labor—labor-repressive agriculture, as Moore calls it—into the modern era is a second essential feature of the path to fascism.3 The method of labor control leads the landlords to seek an alliance with those in control of the means of coercion, the state, and it accounts for the strong antidemocratic impulse of the aristocracy (Moore 1966, p. 435). 3. The country has to have experienced sufficient industrialization so that the bourgeoisie is a politically significant actor, but it cannot be more politically powerful than the landed classes. Skocpol (1973) points to the difficulty involved in measuring the strength of “bourgeois impulses,” but as I have formulated it here, which admittedly takes some license with Moore’s work, this is less problematic. 4. The bourgeoisie is kept in a politically dependent position as industrialization is aided, and to some extent directed, by the state through protection, state credits to industrialists, state development of infrastructure, promotion of modern skills, and even state development of enterprises later handed over to private entrepreneurs. This is the core element of “revolution from above,” and it could occur only after English development demonstrated the possibility of capitalist industrialization. Militarism and thus armaments production seal the bourgeoisie into the statelandlord dominated coalition and its reactionary and imperialistic politics. 5. Skocpol (1973) takes Moore to task on a number of points, among them, his neglect of state strength/capacity as a variable. In truth, Moore alludes a number of times to this (see esp. pp. 439–41, 444). A state with a sufficient capacity to repress peasant (and worker) protest, something absent in England (because of, among other things, the reliance on the navy for defense), is an essential element of the authoritarian class coalition. The dependence of the bourgeoisie on the state in the German case was conditioned by the top-down nature of industrialization, which was made possible by the existence of a strong bureaucratic state. However, Skocpol’s view and Moore’s are not identical: in particular, in his brief discussion of the postNapoleonic period of reaction in England, Moore indicates that the development of sufficient repressive capacity would not have been a difficulty; the crucial feature of the English case was the interests of the landed upper class (i.e., no need to repress the peasants) not the capacity of the state, a view clearly not shared by Skocpol (see also Skocpol 1979). 6. Finally, there must be no revolutionary break with the past. Thus, peasant revolutionary potential must be low (for the obverse reasons pointed to in the case of peasant revolution), or else the whole process, in particular the power of the landlords, would have been broken earlier.


227

It should be noted here that while the coalition of the state, labor-repressive landlords, and a dependent bourgeoisie seems to be an essential feature of the authoritarian path, the six factors listed above may simply contribute to the outcome, that is, may make it more probable, without being essential. In developing this list, I have focused on the elements that come closest to being necessary conditions; other contributing conditions could have been added. For instance, in the synthetic chapters, Moore (p. 417) initially argues that the persistence of royal absolutism into modern times is a feature of nondemocratic paths, but he later indicates that it is only a contributing feature, recognizing that Italy was a constitutional monarchy. One might add that Spain was also, but that Sweden and Denmark, like Germany, were ruled by limited absolutist governments in which the king could appoint his cabinet without reference to the composition of parliament. Nevertheless, it is probably accurate to say that the persistence of royal absolutism in Germany and Austria did contribute to the authoritarian outcome in those countries. Two final points on Moore’s argument involve the problematic characterization of the end state as “fascism” and the path from traditional predemocratic authoritarian regimes (ruled by Moore’s authoritarian coalition) to the interwar fascist regimes. In his synthetic chapter on fascism, Moore’s treatment of this period is extremely brief, and it is based largely on the German case. His argument follows: Fascism required mass mobilization. What developed initially was not fascism but royal authoritarianism, in which there was some mobilization but most of it against the regime, not by it. In this period, the landed class successfully promoted its authoritarian ideology among the peasantry. This royal authoritarianism was broken by war; democracy was not an internal development. The landed upper class retained a substantial amount of power in this democratic interlude not only in the countryside but also in the bureaucracy, judiciary, and the army (i.e., in the state). It allied with fascism, which was based in the urban middle class and the peasantry, who felt themselves squeezed between labor and capital and who were open to fascism’s extreme nationalist and authoritarian ideology in part because of the previous propaganda efforts of the landed upper class. Now, if one were to suggest that a country had to fit this description for Moore’s thesis to be correct, only Germany would really support his argument. Moreover, labeling the end state “fascism” eliminates Spain and Austria, as these countries cannot be considered fascist according to his definition or Linz’s (1975), the leading authority on typologies of authoritarian regimes. I suggest a less strict and, I think, more useful test of Moore’s thesis. Does the existence of certain preindustrial agrarian class relations (laborrepressive agriculture) in general, and more specifically the presence of the statelandlord-dependent bourgeoisie coalition, distinguish the countries that succumbed to modern capitalist authoritarianism (fascist or otherwise) in the interwar period from those where democracy survived? Can difficulties in the initial transition to democracy and the events leading to the subsequent breakdown be traced to the pattern of class-state relations identified by Moore?

Other Theories in the Development of Democracy In his classic statement of the modernization perspective on the development of


228

democracy, Lipset (1960) argued that there is a strong relationship between socioeconomic development and democracy, supporting his assertions by presenting some simple correlations between various indicators of development and democracy. Lipset argued that industrialization leads to increased wealth, education, communications, and equality, which, in turn, are associated with a more moderate lower and upper class and a larger middle class (which is by nature moderate); and this in turn increased the probability of democratic politics. Subsequent, considerably more elaborate, crossnational statistical studies all confirmed that there was a strong relationship between socioeconomic development and democracy (e.g., see Bollen 1979, 1983; Bollen and Jackman 1985a, 1985b). Generally, the results of these cross-national studies have sharply diverged from comparative historical studies, such as those of Moore and O’Donnell (1973, 1978). Rueschemeyer (1980) argues in his overview and analysis of this controversy that the results of cross-national studies, which invariably show a moderately strong relationship between development and democracy, are too persistent to be ignored. However, using the frame of reference of the comparative historical studies, he provides an alternative explanation of the statistical results; that is, briefly, industrial capitalism creates conditions that facilitate the organization of the working class and the middle strata, making it much more difficult for elites to exclude these groups politically. Here Rueschemeyer finds some common ground with the more Marxist analysis of Therborn (1977). In Therborn’s comparative study of the historical transition to democracy in the advanced capitalist world, he emphasizes the role of the working class as represented by working-class parties and unions. Because of the question they ask, these theories do not explain deviations from the central tendency, such as relatively highly developed countries that are nonetheless authoritarian. Thus, though they may provide an explanation for the general trend toward democratic government in Europe in the 1870–1920 period, they contain no alternative explanation to Moore’s for why some countries’ development culminated in authoritarianism in the interwar period. Recent work by Kurth (1979) provides such an alternative. Developing O’Donnell’s line of thought on bureaucratic authoritarianism in Latin America and Gerschenkron’s (1962) on European industrialization, he argues that the timing of industrialization and the movement through various industrial phases (from light consumer goods to capital goods to consumer durables) relative to other countries influenced the political development of Europe over the last century and a half. Industrialization in early industrializes, such as Britain, France, and Belgium, was propelled by the light consumer goods industries, especially textiles. The relatively small amount of capital required for the development of these industries facilitated industrialization without dependence on the state, which Moore argues was a feature of the authoritarian pattern. German industrialization, because it was late, was dependent on the state, first for protection and later for aid in mobilizing capital in the capital goods phase, which tied the German bourgeoisie to the state and strengthened the authoritarian Junker-crown alliance. A similar situation prevailed in Austria. The entry into the capital goods phase, in which the development of the steel industry played a key role, further reinforced this pattern. The capital required (which was relatively high in this case) was mobilized from the profitable and internationally competitive light consumer goods


229

industry in the early industrializers, whereas this was not possible in the later industrializers, which turned to the state for this task. Kurth argues that it is in the saturation phase of the steel industry that the link between the timing and phasing of industrialization and the development of modern (i.e., 20thcentury) authoritarianism is most clear. Initially, the vast majority of steel output was absorbed by railroad construction. Once the domestic market for rails was saturated, steel producers turned to other countries and overseas colonies—and to armaments. Germany, with few colonies and with overseas markets preempted by the British, had to rely more heavily on armaments. The loss of colonies as a result of World War I exacerbated the situation, and, not surprisingly, the steel industrialists favored rearmament and rejection of the Versailles Treaty and were strong supporters of the German National People’s Party (DNVP) and authoritarian “solutions” throughout the Weimar Republic. As for the “late-late” industrializers of Latin Europe, Kurth suggests that there was a link between the trasformismo/El Turno politics of the late 19th and early 20th centuries and the industrial phase at the time. The uncompetitiveness of both industry and agriculture was reflected in protection of both sectors and in a merging of the interests of land and industry, which was in turn reflected in the political arrangements of trasformismo/El Turno. Both involved, among other things, an accommodation of interests of segments of the economic elite through bargaining and clientelistic ties. Thus, Kurth’s argument suggests a different economic basis for the development of the landowner-bourgeois coalition in Latin Europe from that in Central Europe. From this summary of Kurth’s work, it appears that his argument may be seen as complementary to Moore’s rather than as an alternative to it. But there is one critical difference: Moore’s view suggests that the source of the antidemocratic posture of the bourgeoisie in the authoritarian cases is its dependence on the aristocracy-dominated state, whereas Kurth locates the source in the economic situation of the class or segments of it. Thus, political (and ideological-cultural) subordination of the bourgeoisie to the agrarian elite is not a necessary feature of his argument.4 To foreshadow my conclusions briefly, I will employ the class analytic/class coalitions approach of Moore, Therborn, and Rueschemeyer. Following Therborn, I will underline the role of the organized working class in the final push for democracy but will demonstrate not only that the working class needed allies in other classes in this final push but also that other social classes were, in many cases, more important in earlier extensions of suffrage or struggles for parliamentary government. Building on Moore’s work, one can account for the variation in such class coalitions in the push for democracy and the causes of the interwar breakdowns. The existence of a large landed class created a set of alliance possibilities for other classes that impeded the development of democracy in the first place and, once democracy was established, facilitated its eclipse by authoritarian forces.

Table 1: Agrarian Elite Strength and Political Outcomes Strength of Agrarian Elite in Late 19th Century Weak Strong


Democracy survives in interwar period

Partial eclipse of democracy Authoritarian regime established

230

Sweden Denmark Norway Switzerland Belgium Netherlands France Finland Austria-Hungary Spain Italy Germany

The Transition to Democracy In 1870, no country in Europe was democratic by the criteria laid out above. By 1920, the overwhelming majority were. Two decades later, democratic rule had been eclipsed in a number of these countries. Let us now look more closely at the processes that brought democracy to these countries and at what separated the democratic survivors from the cases of breakdown in the interwar period. Moore’s analysis focuses heavily on the type of agricultural arrangements and labor-force control adopted by the landed aristocracy. Had Moore included the smaller European countries, his focus would certainly have begun with the existence (or absence) of a politically powerful landlord class. This, in turn, is largely a product of the pattern of concentration of landholding itself: In all the small countries, there were too few large estates to support the development of a politically significant class of landholders. This one factor prevents the development of the class coalition that Moore argues is fatal for democracy. And in fact, the correlation between the strength of large landlords and the survival or breakdown of democracy in the interwar period (Table 1) indicates that this one factor provides a powerful explanation for the survival or demise of democracy.5 It should be noted that large landholding may not be “dominant” in a statistical sense. In Germany, the West and South, the major portion of the country in land area, were dominated by small farming, as was the North of Italy (except for the Po Valley). Spain and Austria-Hungary also contained regions in which smallholding was dominant. The critical factor here is that, in all these countries there was a sufficient number of large estates to give rise to the formation of a politically powerful landed elite. In all the small countries, by contrast, small to moderate holding was the overwhelmingly dominant form of land tenure, and no large agrarian elite existed. A few points on the strength of the agrarian elites indicated in Table 1 are in order before we proceed. Historically, the French agrarian elite was very powerful, but the revolution broke its power. By the late 19th century, the French countryside was dominated by small peasants, and the landed upper classes were no longer the powerful political actor they had been a century earlier. Thus, the revolutionary break from the past


231

that Moore hypothesises as a necessary feature for democratic development was essential in the French case. However, as Katzenstein (1985) points out, most of the small states in Europe did not experience a revolutionary break but nonetheless developed in a democratic direction. Again, Moore’s analysis is flawed by the exclusion of the small states. The virtually perfect correlation between country size and landlord strength is no accident. As Tilly (1975, pp. 40–44) points out, military success was one factor that distinguished the successful state builders from the unsuccessful ones, and success in war was greatly facilitated by “strong coalitions between the central power and major segments of the landed elite.” The small states avoided being gobbled up only by reason of geography (Scandinavia), the operation of the interstate system from the Treaty of Westphalia onward,6 or both (Switzerland). Britain stands out as a deviant case in terms of landholding, and resort to Moore’s emphasis on the type of commercialized agriculture as an explanatory factor is necessary to bring this case into line. And indeed it is also accurate to classify the three authoritarian cases not discussed by Moore (Italy, Spain, and Austria-Hungary) as cases of dominance of “labor-repressive” agriculture. Still, while the correlation presented here is suggestive of the causes of breakdown, we must examine the individual cases to uncover what social forces produced democracy and what forces and dynamics appear to explain the relationship between landed class strength and breakdown. By the eve of World War I, Switzerland (1880s), Norway (1898), and France (1875–84 for the consolidation of parliamentary government based on male suffrage and rights of assembly, etc., or 1913 for the secret ballot reform) had become democratic, and in 1915 Denmark joined this group.7 These are all nations of smallholders, urban petty bourgeoisie, and with a significant though not dominant industrial sector (and therefore significant working and capitalist classes) at the time of democratization. In Switzerland, democracy was achieved through pressure from small farmers and the urban petty bourgeoisie. In France, the various Republican factions of the late 1860s and 1870s, which provided the final push to democracy, were supported by the working class, the petty bourgeoisie, segments of the peasantry (depending on local economic organization, clerical presence, and revolutionary traditions), and segments of the bourgeoisie, especially in the provinces. The events of the late Second Empire clearly built on earlier democratic advances, especially 1848, which, though rolled back, continued to influence the course of events. In these developments, the bourgeois influence was weaker, and rebellions of the largely artisanal working class played a much larger role (Anderson 1977; Aminzade 1981, in press; Elwitt 1975; Valenzuela 1979). In the two Scandinavian countries, the working class organized in unions, and political parties played some role in the drive for democracy. In the case of Norway, the working class contributed to the final push for universal suffrage (embodied in various laws passed between 1898 and 1913), though earlier suffrage extensions were largely the work of the peasantry with the help of sections of the urban middle class (Rokkan 1966; Derry 1973). In Denmark, the working class/small and medium farmer/urban middle segments coalition as represented by the Social Democratic-Venstre coalition pressed through the 1901 introduction of parliamentary government. The driving force behind the 1915 introduction of universal suffrage was the Social Democrats and the Radikale Venstre, representing the working class, small farmers, and segments of the middle class (Miller 1968; Dybahl 1969). It is


232

important to note here that in Denmark and Norway (as well as Sweden), the medium farmers were ambivalent about the final suffrage extensions that resulted in the inclusion of large minorities of working-class voters not previously included, and it was among segments of the urban middle classes and the small farmers and tenants that labor found its ally. So, to argue that the effect of landholding patterns on political outcomes was caused simply by the authoritarian posture of large landlords and the democratic posture of smallholders is inaccurate. At the same time, it is important to note that the Scandinavian peasantry was divided, and even the medium and larger farmers, though not supportive of the final push for universal suffrage, generally contributed to the process of democratization by supporting earlier suffrage extensions. In the rest of Europe, but particularly among the antagonists in World War I, the social dislocations caused by the war contributed to the breakthrough of democracy. The war and its outcome changed the balance of power in society, strengthening the working class and weakening the upper classes. The ruling class was discredited, particularly in the defeated countries. Labor support was necessary—at home for the production effort and on the front for the first mass-mobilization, mass-conscription war of this scale and duration. And, finally, the war economy and mass conscription strengthened the hand of labor in the economy, enabling it to extract concessions for the coming period of peace.8 One indicator of the change in class power was the swell in labor organization from an average prewar level of 9% of the labor force to a postwar peak of 30% among the antagonists, which experienced the transition to democracy in this period (1918 or 1919). Organization more than doubled in the two nonparticipants (Sweden and the Netherlands), which experienced the same transition at this time (Stephens 1979, p. 115). In all these countries, the working class played a key, if not the key, role in the transition to democracy. But, as Therborn (1977) notes, the working class was not strong enough alone. It needed allies or unusual conjunctures of events to effect the introduction of democracy. As an indicator of this, it could be pointed out that in no case did the working-class parties receive electoral majorities even after the introduction of universal suffrage. In England. Sweden, Belgium, and the Netherlands, it can be argued that the war only accelerated the introduction of democracy. In each country, the prodemocratic coalition—the parties and the underlying alignment of social forces—had formed or was in the process of formation. In most cases, the coalition had been responsible for previous suffrage extension, such as the 1907 reform in Sweden or the 1893 reform in Belgium. In Sweden, Belgium, and the Netherlands, the agrarian elites were too weak to be a significant political force. In Sweden, as in Norway and Denmark, the peasantry was split on the question of universal suffrage. It was the Liberals (who were based in the urban middle classes, in the dissenting religions, and in the small farmers in the North and West) who joined the Social Democrats in the push for suffrage extension. The war stimulated the Conservative capitulation in Sweden, and an interparty compromise, implemented in 1917, followed several decades of political pressure (through strikes, demonstrations, and parliamentary obstruction) by the Social Democrats and the trade unions in cooperation with segments of the middle class.9 In Belgium, the Workers’ party, after decades of struggle, including six general strikes, found support in the Social Christian wing of the Catholic party, which was based in working-class Catholics


233

(Fitzmaurice 1983; Lorwin 1966; Therborn 1977, pp. 12, 25). In the Netherlands, similar divisions in the religious parties and the liberals produced alliance possibilities for the Social Democrats (Daalder 1966, pp. 203–11). It is worth underlining that the accounts of the transition in both low countries make it clear that the growing importance of the working class in society created the pressures that moved these nonsocialist parties toward a more democratic posture. In part, this pressure was transmitted by workers and artisans, already mobilized by self-help societies and trade unions, who joined these parties, and, in part, the pressure was a result of the efforts of these parties to compete with the Social Democrats for the loyalties of unmobilized workers. The British case is singular in so many ways, in both the antecedents of democracy and the process of democratization, that it is virtually impossible to decide which factor(s) was (were) the most important on the basis of comparative analysis. Various analysts have argued that it was the absence of labor-repressive agriculture (Moore), the absence of a bureaucratic state and standing army (Skocpol), or the independence of the bourgeoisie due to the country’s status as an early industrializer (Kurth) that separates Britain from Germany and the other authoritarian cases. To these, one might add an argument based on Rueschemeyer’s interpretation of the relationship between development and democracy and combined with Moore’s arguments on the role of landlords and the elimination of the “peasant question” in Britain. Relative to the level of industrialization, and thus the configuration of the class structure, democracy came late to Britain. By the time of the first suffrage extensions to the working class in the late 1860s, less than one-fifth of the labor force was engaged in agriculture, and over two-fifths were in mining, manufacturing, and construction (Mitchell 1978, pp. 51–64). Almost no other European countries took on such a labor-force profile, and corresponding class structure, until after World War I or, in some cases, after World War II. Thus, when one compares Britain with other large landholding cases (the authoritarian cases) during this same period, one must keep in mind that the latter were much more agricultural and that the landlords had much more economic power and thus, potentially, political power, and the working class was a much less important potential force in the country. The establishment view of suffrage extension in the British case, one that Moore appears to share, argues that the “peculiarities of English history” (however specified) meant that segments of the British upper classes had settled into a pattern of peaceful political competition by the mid-19th century, and this extended to competition for working-class votes, which resulted in the suffrage extensions of 1867 and 1884. My comparative analysis of the transition to democracy suggests that in Britain this process in itself, if it were true, would be a peculiarity. In no other case did middle-class-based (and largely upper-class-led) parties unilaterally extend effective suffrage to substantial sections of the working class (except where suffrage was irrelevant to the actual governing of the country because of the lack of parliamentary government, as in Germany, or because of electoral corruption, as in Spain and Italy). At best, some sections of the middle classes (and in France, some segments of the bourgeoisie) allied with the working-class parties for such suffrage extensions. All cases in which the working class was politically included without substantial pressure from the politically organized working class itself (Switzerland and, to a lesser extent, Norway) were essentially agrarian democracies in which democracy was established by a peasant/urban


234

middle-class coalition before the working class had become a significant political actor. On deeper examination, the establishment view appears to be flawed. As Johnson (1976) argues in his critique of Moore’s view of the British route, these reforms were in large part a response to working-class pressure beginning at least as early as the Chartist movement, whose main demand was universal suffrage, and extending throughout the 19th century. In this regard, the British case bears some resemblance to the French case, as the final transition to democracy was in part a delayed response to earlier workingclass agitation that predated the formation of late 19th-century social democratic parties. Nonetheless it is a peculiarity that the final political initiation of the reforms came from upper-class-led parties without a strong working-class base. Part of the explanation of this lies in the late development of the Labour party itself.10 The Liberals and the Tories were willing to extend the right to vote to workers only because they hoped to benefit from the votes of the newly enfranchised workers. Had a substantial Labour party already commanded the loyalty of workers, the established parties would have certainly been reluctant to make such a move. If this argument is correct, it also suggests that the absence of a significant socialist working-class party in France in the late 1860s and 1870s may have contributed to the willingness of significant sections of the bourgeoisie to support parliamentary government based on universal male suffrage. And finally the reform of 1918, which established male suffrage and eliminated all but minor provisions for multiple voting, was the culmination of the Labour-Liberal cooperation that led to the rise of the Labour party. No one would deny the important role of the working class in this reform. Rather, it is contended that the reform was of minor significance compared with the 1867 and 1884 reforms. Blewett’s (1965) careful study demonstrates that this is a mistake. Although 88% of the adult male population would have qualified to vote in 1911 were it not for complications and limitations in the registration procedures, which were biased against the working class, less than two-thirds were on the voting rolls. The importance of these restrictions can be seen from the fact that this figure rose to 95% after the 1918 reforms (Matthew, McKibbon, and Key 1976, p. 731). Moreover, in 1911, half a million of the eight million voters were plural voters, and needless to say not many of them were working class. The final proof of the importance of the 1918 reform is the analysis of Matthew et al. (1976), which demonstrates that the reform was critical in allowing Labour to displace the Liberals as the second party in an essentially two-party system. This survey leaves us with our breakdown countries11 as cases in which the war may have influenced more than the timing of the introduction of democracy. But before moving on to them, let us take stock of what can be learned from the development of democracy in Europe as I have outlined it. One obvious lesson, stressed by Therborn (1977), is the important role played by the working class, that is, by its organizational representatives, the trade unions and the socialist parties. One can add to that the role of artisan agitation and early craft unions in the French and British cases and the role of workers in the confessional parties in the Netherlands and Belgium in. pressing those parties toward a more democratic posture. The rapid development of industrial capitalism in the second half of the last century stimulated working-class organization that first gradually, and then with the war and its outcome, decisively changed the balance of class power in all these countries; indeed, it changed the balance of class power in the entire


235

core of the world capitalist system. The change in the underlying class structure as indicated by labor-force figures is significant enough: between 1870 and 1910, the nonagricultural work force grew in these countries by one-third to one-half to an average of 61%.12 The change at the level of class formation and class organization was even more significant: in no country in 1870 were the socialists a significant mass-based party, and the trade unions organized a minuscule proportion of the labor force; by the eve of World War I the parties affiliated with the Second International garnered an average of 26% of the vote (despite suffrage restrictions in a number of countries), and the trade unions organized an average of 11 % of the nonagricultural labor force. In the immediate postwar elections, the socialists’ electoral share increased to an average of 32%, while trade union organization grew spectacularly, increasing two-and-a-half-fold. The organized working class was also the most consistently prodemocratic force in the period under consideration: at the onset of World War I, European labor movements, all members of the Second International, had converged on an ideology that placed the achievement of universal suffrage and parliamentary government at the center of their immediate program (Zolberg 1986). This interpretation turns Lipset (1960) and all the cross-national studies that followed on their heads; the working class, not the “educated, literate, moderate, growing middle class” was the driving force behind democracy. But it also turns Barrington Moore on his head, who agrees with “the Marxists” on one point: “No bourgeoisie, no democracy” (Moore 1966, p. 418). Moore also completely ignores the role of the working class.13 However, Therborn’s (1977) focus on the last reforms in the process of democratization leads to an exaggeration of the role of the working class. First, in the two agrarian democracy cases (Switzerland and Norway), the role of the working class was secondary or nonexistent even in the final push to democracy. Second, in other cases, not only did the working class need allies in the final push; in earlier democratic reforms, multiclass alliances were responsible for the success of the reform (France, Britain, Denmark, Sweden, and Belgium). But, as will become apparent in my analysis of the authoritarian cases, none of these other social classes were as consistently prodemocratic, both across countries and through time, as the working class.14 Both the urban middle class and segments of the peasantry provided the mass base for authoritarianism in the breakdown cases. The bourgeoisie, whose role in the introduction of democracy has been emphasized in so many accounts, from Marxist to liberal, played a positive role in only two cases, Britain and France. Moreover, in these two countries, it was only segments of the class that cooperated in the push for democracy and then only after earlier histories of popular agitation for democracy and bourgeois resistance to it. In all the others, the bourgeoisie was one of the centers of resistance to working-class political incorporation. It did make an indirect contribution to the outcome, however. In the cases discussed so far, the bourgeoisie sought entry into the corridors of power, and in all cases, except for Denmark and Sweden, it supported the drive for parliamentary government. Bourgeois political forces established parliamentary government with property, tax, or income qualifications for voting—that is, democracy for the propertied—a true “bourgeois democracy” in contrast to the bourgeois democracy of Leninist Marxism. This system then was opened up by


236

successive organized groups demanding entry into the system: the peasantry, the middle class, and finally, the working class. There is a certain amount of truth to the extremely crude interpretation that each group worked for its own incorporation and was ambivalent about further extensions of democracy. The positive contributions of the bourgeoisie were to push, for the introduction of parliamentary government and then to capitulate to pressures for further reforms rather than risk civil war.

The Breakdown of Democracy As we saw, the working class needed allies; its power alone was insufficient. Here is where the Moore thesis (revised to accommodate the smaller countries) comes in as it outlines the social and historical conditions that created the possibilities for alliances. In the cases of coalitions of the landed upper classes, the state, and the bourgeoisie, no alliance strong enough to overcome their opposition could be constructed. It was only the change in the balance of class power caused by the war that allowed for the democratic breakthrough. But, as Maier (1975) argues in his study of Germany, France, and Italy, this surge in the strength of labor and the political Left was quickly, though not completely, rolled back. A quick glance at union membership and voting statistics indicates that this was a general European pattern. Where this surge of working-class strength was the essential ingredient in the transition to democracy, the working class and its allies (where it had any) were unable to maintain democracy when a new conjuncture presented new problems (depression, worker or peasant militance, etc.) and new alliance possibilities for the upper classes moved the bourgeoisie and the landlords from passive to active opposition to the democratic regime. This still leaves us with something of a black box in terms of the mechanism by which the existence of a relatively strong class of landlords actually influenced the political structures and events of 1870–1939. One might first ask what difference it makes that landlords were an element of the ruling coalition (as opposed to a simple bourgeoisiestate alliance). Moore, as we have seen, gives a straightforward answer to this question: the landlords, who had earlier cemented an alliance with the crown/state, exercised a political and ideological hegemony over the rising bourgeoisie, in which the latter accepted the ideological leadership of the landlords, in part as a result of state support for industrialization. The authoritarian politics of the agrarian elite were transmitted to the bourgeoisie. In tracing the state’s motivation, one might hypothesize that initially the crown/state made the alliance with the landlords because, as Tilly points out, the alliance was militarily strong. This alliance was progressively strengthened as the elements of the state apparatus (the military, bureaucracy, and judiciary) were drawn from agrarian elites directly or the occupants of these positions were absorbed through accretion or both. All three groups then retained these authoritarian politics in the “democratic interlude.” And, to the extent that the haute bourgeoisie was drawn into the authoritarian politics of the agrarian elites, it also came to participate in the social linkages to the state apparatus strengthening the antidemocratic posture of the state. As pointed out earlier, Kurth argues that the bourgeoisie in some countries may have had autonomous reasons for adopting antidemocratic politics, and a similar line of


237

argument has been advanced by a number of historians critical of Moore’s view of German developments (see below). The main function of a strong agrarian elite in this perspective is to create an alliance option for the bourgeoisie to pursue such antidemocratic political impulses, an option not present in the smallholding countries. This still leaves one with a problem for the democratic period because even a highly cohesive upper-class alliance must reach beyond its ranks to influence the political developments in the era of mass politics. It can be said that three basic mechanisms can translate the power of landlords and the more general antidemocratic impulses of both segments of the upper classes into influences on the events of the democratic period (and the mass politics of the period immediately before World War I). The first is conscious instrumentation, such as funding authoritarian parties and movements, using political influence to obstruct democratic procedures, and so on. The second mechanism is what is referred to as structural determination in Marxist theories of the state. If one rejects functional and tautological structural explanations of the Althusserian Marxists, one is left with the more limited but nonetheless very powerful meaning given it by Block (1977). In a capitalist society, he argues, any government must ensure that the basic conditions for capital accumulation are met. The threat of investment slowdown and capital flight is a constraint on any government and forces it to take into account the interests of holders of productive property, even in the absence of overt pressure from those classes, if they are to keep the economy on an even keel. Conversely, stated in terms of my problematic, governments that do not have the confidence of capitalists (and landlords) may find that declining investment, capital flight, and so on add economic difficulties to their other difficulties, resulting in a destabilization of the regime. The third mechanism is through the effect of ideological hegemony, to draw on the Gramscian concept. In the view of the state in advanced capitalist democracies that I have advanced elsewhere (Stephens 1979; see also Korpi 1982), the state is seen as reflecting the balance of class power in civil society.15 Crudely put, the capitalist class derives its power from property and the working class from organization. In advanced capitalist societies, as Gramsci argues, the ruling class rules in large part through a historically developed hegemony or ideological domination. In the state- and nation-building process, the state-building alliance (e.g., in Germany, the Junker-state alliance) produces, in a quite nonconspiratorial way, an ideology that legitimates its rule and its development project (where such a project is present). This alliance can be referred to as the hegemonic fraction of the overall ruling-class coalition (in Germany, the state-Junkerheavy industry alliance). As more social groups are mobilized, as civil society becomes dense, the ruling ideology is diffused to other groups. This attempt is generally successful, especially in the upper middle classes and more affluent middle strata. However, in cases such as Germany and Austria, the labor movement insulated much of the working class from ruling-class hegemony by building, in a very conscious fashion, a counterhegemony through the development of a dense organizational life—the party, trade unions, workers’ education associations, sports clubs, youth organizations, women’s organizations, and the development of alternative mass media and so forth. Gramsci was prescribing what labor movements should do rather than what they, in fact, had achieved. However, while there is reason to believe that even the German


238

movement did not inculcate all or even most rank-and-file members of its movement with a clear conception of an alternative socialist society, it did insulate them from the authoritarian politics of the Imperial German elites. Moreover, it is important to observe that the Catholic church did a similar thing; it organized an institutional culture for its believers. Clearly, the class content of this culture was quite different from that of the working-class institutional culture. At one extreme, when the church was a minority and under attack by the state-building elites, it did insulate its believers from ruling-class hegemony. At the other extreme, where the church itself allied with the state and the large landed class, the Catholic institutional culture became a conduit for ruling-class ideology. To bring this back into the context of this comparative study, it is my argument that the political postures of the urban middle classes and the peasantry were heavily influenced by the ruling-class coalition that led the political development of the country in question. Where Moore’s authoritarian upper-class coalition was well established, it not only affected the content of the ideology propagated by the ruling classes, that is, a particularly hierarchic, rabidly antidemocratic, antiliberal set of values, but it also affected the extent to which the ruling ideology was accepted by the urban middle classes and, especially, the peasants. This way of employing the concept of ideological hegemony can be fruitfully combined with traditional social scientific analyses of political mobilization in Europe (see esp. Lipset and Rokkan 1967; Rokkan 1970) to give a more class analytic content to the latter. Such a combination yields the following insights in the case of the peasants. Lipset and Rokkan (1967, esp. pp. 44–46) point out that, in Protestant smallholding countries (or regions of countries, e.g., Swiss cantons), the peasants themselves were the agents of their own mobilization, and the political form was agrarian parties. In Protestant largeholding countries, the mobilizing agents were the landed upper classes; thus the political weight of the peasants strengthened that political bloc. In Catholic countries (or areas of countries, e.g., southern Germany), the mobilizing agents were Catholic parties. For a full specification of the ideological orientation of these Catholic parties, it is necessary to bring into the analysis both the posture of the state vis-à-vis the Catholic church in the Catholic countries and the size of the Catholic community in predominantly Protestant countries (see Table 2). As one can see from the table, the central tendency is clear. The ideological posture of the parties varies by the landholding structure—thus, the role of large landholders in shaping the parties. Above all, in all countries with a significant landed elite, that class was a key force behind the party that mobilized significant sections of the peasantry. There are, of course, exceptions because of historic relations among church, landlords, and peasants in particular regions (especially in Spain and France), but the scheme presented in the table gives one a baseline that holds in most cases. A similar scheme could be presented for the urban middle class. One caveat should be issued before closing. It is not my intention to attribute, directly or indirectly, all antidemocratic and reactionary impulses in peasant and middle-class politics to the ideological hegemony of the Moorian coalition. This would be clearly wrong, as the examples of the Lapua movement in Finland, Rexism in Belgium, and Action Française or Poujadism in France demonstrate. Moreover, all mass support for fascism or other forms of authoritarianism in the four breakdown cases cannot be traced


239

to that source. Rather, I want to argue that the existence of a strong agrarian elite and allied bourgeoisie significantly increased the appeal of such reactionary ideologies in these other classes.

Table 2: Mobilizing Agent of the European Peasantry Strength of Agrarian Elite in Late 19th Century Weak Strong Catholic: State allied with Belgium (Rightist but moderate Austria (Rightist church Catholic party) Catholic party) ?* Swiss Catholic cantons (Rightist but Spain (Rightist moderate Catholic party) Catholic party)† Not allied with Northern Italy (Centrist Catholic party) Southern Italy† church Protestant: Scandinavia Eastern Germany Small or no Catholic minority Britain (Rightist Swiss Protestant cantons (Centrist Protestant parties) agrarian parties) Significant Catholic South and West Germany† minority Netherlands (Centrist Catholic parties and Protestant parties of the Right and center) * The classification of Switzerland on this dimension is problematic since the central state clearly had no religious alignment but the individual cantonal governments did. The Spanish case is problematic as the alliance of the crown and the state forces shifted during the 19th century from opposition to the church at the time of the Carlist wars to alliance at the end of the century. † These are deviant or problematic cases. The Protestant farmers of northwest Germany were mobilized, in part, by the conservatives and the Agrarian League in Imperial Germany and the DNVP in the Weimar Republic, all of which were dominated by the East Elbian large landlords, yet particularistic and regional parties also gained support from these farmers. In Southern Italy, landlords controlled the peasant vote and threw their support to the anticlerical liberals in the clientelistic logrolling process characteristic of the period. In Spain, the party referred to is the CEDA, but much of the Carlist peasantry supported Basque regional parties, and the southern agricultural workers and tenants supported the anarchists and socialists. The conspicuous absence of France from the table is because of the great regional variation in peasant politics owing to variations in local modes of production, role of the clergy, and legacies of the Revolution. With this last mechanism in mind, the class alliance option argument can be restated. The existence of a strong class of large landholders with close ties to the state not only


240

changed the alliance options of the bourgeoisie. Together, these three groups exercised an ideological influence over segments of the middle class and the peasantry that also pushed these segments in an authoritarian direction or at least prevented them from allying with the working class in the push for democracy, thus reinforcing the viability of the authoritarian option for the bourgeoisie. In clarifying the status of these three mechanisms, it is useful to draw on Stinchcombe’s (1968) distinction between historical and constant causes. A historical cause is one that happens at a point (or, in this case, a period) of time in the past, and then the pattern created reproduces itself without the recurrence of the original cause. So, for instance, the Kulturkampf, Naval League, and Agrarian League propaganda campaigns (see below) did not have to be carried on constantly for them to have an effect on the political attitudes of the peasantry a generation later. The instrumentation and structural mechanisms are closer (though not identical) to a constant cause, which is a set of social relationships, activities, and so on that are constant from year to year and produce a constant effect. In the present analysis, I am concerned with institutional change rather than maintenance, but otherwise the argument is the same. The importance of this distinction for this analysis is that the mechanism of ideological hegemony need not have a close relationship with the current economic and political strength of large landlords or with the cohesion of the authoritarian coalition. Naturally, persistence of landlord power or the cohesion of the coalition will serve to maintain the ideological legacy of the past, but that legacy will not decline in a one-to-one relationship with the decline in landlord power or coalition cohesion. Italy From the point of view of the historical development of the state- and nationbuilding coalition, Italy would appear to be a very poor fit with the Moore thesis. As Lipset and Rokkan (1967) point out, the Italian state allied with the urban upper class in its drive to unify the country. This usually indicates a weak landed upper class, since European states generally chose to ally with landed interests if these were sufficiently strong. Yet Italy did contain significant largeholding regions in the South and the Po Valley, and the large estates in these areas were sufficiently numerous to sustain a very significant landed class. Italy deviates from the modal pattern because unification was spearheaded from the Piedmont, a smallholding region, as was most of the rest of the industrially advanced North. The South resisted incorporation into the new Italian state. But even after the incorporation of the South, it is clear that the landed upper class did not assume the political leadership of the country; thus Italy does not fit Moore’s pattern of a landholderdominated state in alliance with a dependent bourgeoisie. Nevertheless, an alliance, or at least an accommodation, among the landed class, the bourgeoisie, and the state did develop in the period between unification and the war. The first step in this direction was the development of trasformismo, a political barter system in which the main political and economic interests (i.e., land and industry) were accommodated through clientelistic exchanges, first under Depretis and then Giolitti (Seton-Watson 1967, pp. 51–52, 91–97, 246–48). Second, the government attempted to encourage economic development through protection of industry, protecting agriculture


241

as compensation. And finally, the government embarked on an armaments program that tied segments of industry to the government and a militaristic foreign policy. Lyttelton (1973, pp. 11–12) observes, “The political effects of the 1887 tariffs were extremely serious. The decision to create a national steel industry was defended by invoking the needs of national defence, and the new iron, steel and shipbuilding interests were heavily dependent on a lavish programme of battleship construction. The link between an uncompetitive heavy industry dependent on state contracts and the pursuit of an imperialist power policy, first forged between 1882 and 1887, was ominous for the future. The alliance between Northern industrial interests and Southern latifondisti bought into being by the tariff was a powerful obstacle to democratic development.” The parallels to the German tariffs of a decade earlier and subsequent developments in that country are obvious: the tariff barter brought heavy industry and agriculture into an alliance similar to the rye-iron alliance in Germany, and the armaments program made heavy industry dependent on the state and an imperialistic policy. It is true, of course, that the landed classes never assumed a role of political leadership in Italy, and thus the case differs from the typical coalition pattern outlined by Moore. Still, the 1887 tariffs were not an isolated instance. At the turn of the century Pelloux, the prime minister, made an explicit attempt to establish a heavy industry-agrarian coalition in support of reactionary and repressive policies. Though this failed, part of the project of his successor, Giolitti, the dominant politician from the 1890s to the war, was to create links between the state and agricultural and industrial interests through protection, defense expenditure, labor peace, and economic nationalism. In the Italian case, the working class played an important but not leading role in the introduction of democracy. Giolitti extended suffrage to all adult males in 1912 in an attempt to gain support for his government’s Libyan venture.16 Certainly one of the target groups for his cooptative efforts in this case was the Socialists (and, more generally, the working class), as cooptation of the Socialists was a long-term project of Giolitti. However, the 1913 election was managed by Giolitti and his allies through corrupt practices, and it is actually the 1919 election that marks the introduction of democracy in Italy. The mobilization of workers and peasants during the war and particularly in its wake did play an important role in this transition. This is reflected in the outcome of the 1919 election: the principal parties of these classes, the Socialists and the Catholic Populari, were the big victors, polling a slim majority of the vote together, 32% and 21%, respectively (Rokkan and Meyriat 1969, p. 228). In accounting for the breakdown of Italian democracy, conscious instrumentation was in fact very important. The postwar strike wave, factory occupations, peasant organizing, and victories by the Socialists in local council elections alarmed the bourgeoisie and the Po Valley landlords. Both groups began to fund the Fascists on a massive scale as the Fascists made violent attacks on peasant organizations, trade unions, Socialist party offices, and local councils controlled by the Left, the main focus of their activity. Increasingly, the Po Valley landholders not only provided money but actually participated in the movement, whereas the bourgeoisie contributed money but not men. Moreover, they used their influence to prevent any alternative to Fascist rule. As SetonWatson (1967, p. 598) points out, “By 1922 contributions from banks and industrial firms, particularly those of Milan, were flowing into the treasury of the Fascist party, and


242

their representatives in Parliament were using all their influence to block an antifascist coalition.” Given that virtually all accounts of the development of Italian fascism emphasize the penetration of the Po Valley as a crucial turning point in the strengthening and the transformation of the character of fascism (e.g., see Seton-Watson 1967, pp. 505–664; Lyttelton 1973, esp. chap. 3), it should be pointed out here that the size of landholdings in the Po Valley varied and that initially the Fascists got support from modest farmers, who were dependent on wage labor, as well as from large landholders. Once the Fascists had ousted the socialist and Catholic unions, they were in a position to control the local labor market and dispense jobs and thus began to get support from agricultural workers, tenants, and small farmers. Nevertheless, as Cardoza’s (1982) detailed study of the province of Bologna demonstrates, the large, commercially oriented landlords were the most pivotal group in influencing the direction of events in the area. Their support for authoritarianism was clearly motivated by the threat to their control of labor represented by the tremendous thrust of union organization and socialist victories in municipal elections after the war. Forces internal to the state, particularly the security forces, also contributed to the Fascist victory (Lyttelton 1973, pp. 38–40; Tasca 1938, pp. 97–123). Most army officers were sympathetic to the extreme nationalist organizations, and, at crucial points, such as the Fiume invasion and the march on Rome, the government was reluctant to order the army to act against the radical nationalists for fear that they would not obey. The police generally tolerated, often facilitated, and sometimes even participated in the violent attacks of the Fascists on the Socialists, peasant organizations, and trade unions. Without such help from the security forces, the tremendous growth of the movement in the critical winter-spring of 1920–21 would have been impossible. The hegemony argument is less important in this case, as the Fascists were never a mass electoral movement. Moreover, the South was so underdeveloped and its civil society was so weak that no group exercised hegemony in the region; it was integrated into the country’s politics through clientelistic/patronage-type relations and overwhelmingly supported the more conservative factions of the liberals (who, I hasten to add, were not a prodemocratic force). To pose a counterfactual for this case, one might ask what the political situation in Italy would have looked like if there had been no large landholding class, if the countryside were populated by small peasants. Comparative work on party support argues that these Catholic peasants would have supported the Populari, an essentially prodemocratic force, thus strengthening that party and greatly-facilitating the formation of an antifascist parliamentary coalition in 1921–22. Deprived of its Po Valley support, fascism would have remained an urban, and much weaker, phenomenon. Germany The paradigmatic case for Moore’s chapter on the path to modern authoritarianism was Germany, even though there is no chapter on that country in the book. It is not surprising, then, that the German case appears to fit Moore’s analysis quite well. Of course, there are social scientists who do not share Moore’s materialist orientation and object to his


243

treatment of this case among others (see Wiener 1976). But it is probably fair to say that few would contest his assertion that historically an alliance developed between the East Elbian landed upper class, the Junkers, and the Prussian state and that the political leadership of unified Germany after 1870 depended on this alliance for its key political support. In the wake of the depression of 1873, the heavy-industry segment of the bourgeoisie, particularly the coal and steel interests, joined the coalition, with the tariffs of 1879 and later the naval armaments program consolidating the coalition. There are two challenges to Moore’s view of German development. The first is expressed by Skocpol in her contention that he underplays the roles of state autonomy and state strength, a point on which I will comment on several occasions below.17 The second is expressed in recent works by Blackbourn and Eley (1984; also see Eley 1983) and Calleo (1978), both of which attack what they see as the dominant view in current historical thinking on German political development.18 While the views of Blackbourn and Eley and Calleo differ on many points, they agree that the argument of the dominant view, which locates the cause of German authoritarianism and imperialism in the political dominance of the Junkers and the politically dependent—indeed “supine”—posture of the bourgeoisie, is flawed. Blackbourn and Eley argue that the bourgeoisie was clearly the socially and economically dominant class in Imperial Germany and that it was also politically very influential, if fragmented, and certainly did not accept the political leadership of the landed classes. Calleo (1978, pp. 129 ff.) goes further with regard to the relative political roles of the Junkers and the bourgeoisie, asserting that while the Junkers had a veto power in Imperial German policy, big business was more influential in the overall formation of political and economic policy. One key point in Blackbourn and Eley’s analysis (Blackbourn and Eley 1984, pp. 18– 19; Eley 1984, pp. 75–90) is strongly supported by the evidence presented here. They question the historical accuracy of the equation of the bourgeoisie with liberal politics and of liberal politics with support for democracy. They argue that not only in Germany, but also elsewhere in Europe, the liberals not only found much of their leadership and support in other classes, they also did not enjoy the full support of the bourgeoisie. More important, the liberals did not necessarily support the introduction of democracy, which Eley defines, following Therborn, the same way I do. It was the working class and middle strata that accomplished this task, they contend. They argue that the mistaken view that the Imperial German bourgeoisie had accepted the political leadership of the Junkers is based on the fact that the bourgeoisie did not support the introduction of political democracy and on the unwarranted assumption that it would have been in their economic interest to do so. However, it appears that all three authors agree that the German bourgeoisie as well as the middle classes were more antidemocratic than their counterparts elsewhere, and that that was at least part of the reason for the weakness of democratic impulses and the persistence of the authoritarian features of the Imperial German regime. At a party level, Blackbourn (1984, p. 284) notes that not only the National Liberals but also the Left liberals and the Catholic Zentrum opposed the introduction of equal, direct, and secret voting for the Prussian Landtag.19 The analysis of the origins of German imperialism and authoritarianism by the authors differs somewhat. Calleo emphasizes above all the geopolitical position of Germany but also its late industrialization and the accompanying


244

high degree of industrial concentration and its lack of a colonial empire. In an argument similar to Kurth’s, he contends that this led the German bourgeoisie to support imperialism in an effort to secure foreign markets and military contracts. Here he leans heavily on the careful historical scholarship on the Naval Laws of 1898 and 1900 of Kehr ([1930] 1973), who shows that the state (more specifically, the Kaiser and Admiral Tirpitz) initiated the naval build-up. The program was enthusiastically supported by big business, which heavily funded the Naval League and other imperialistic associations aimed at influencing middle-class opinion. The political representatives of the Junkers, however, initially opposed the naval build-up. But he does not dismiss the role of the Junkers in sustaining German authoritarianism entirely, as he indicates that their presence tipped the balance of power in Germany toward the more reactionary, antilabor, and protectionist heavy industry rather than the free-trade oriented and more liberal light industry. Eley (1984, pp. 135, 147, 153–54) argues that the origins of Imperial Germany’s authoritarianism lie in the combination of an aristocratic enclave in the state, the threat of a powerful socialist labor movement, and important economic and religious divisions in the bourgeoisie that prevented the presentation of a united political front. In addition, the high degree of concentration in the Germany economy gave large employers a capacity for repression on the industrial front, which manifested itself in a largely successful antiunion drive in the late Imperial period. This made compromise with labor a less necessary and attractive alternative (Eley 1984, p. 107). Eley (1984, pp. 153–54) nicely sums up his view: The option of the German bourgeoisie’s leading factions for a politics of accommodation with the landowning interests after 1871 was fully compatible with the pursuit of bourgeois interests…. The bourgeoisie entered the agrarian alliance not from a lack of “political self-confidence,” but as the best means of securing certain political goals. The indifference to further “parliamentarization” came less from any “pre-industrial tradition” of authoritarianism, than from a rational calculation of political interest in a situation where greater parliamentary reform necessarily worked to the advantage of the left. Likewise, it made perfect sense for German capitalists to refuse the “just” demands of the working class, once a given level of private economic power and monopoly organization bequeathed it the ability to do so. If we accept the arguments of Calleo and Blackbourn and Eley, at least those in which they agree with one another, what damage does that do to the Moore thesis? It is worth underlining what they share with Moore. They agree that German industrialization was state aided and directed, that the aristocracy held an important, if declining, position in the bureaucracy and army, and that heavy industry opted for an alliance with the landowning class in the wake of the 1873 depression. This rye-iron coalition was reaffirmed by the tariff policies of the Bülow government after the turn of the century. In 1906, even the Left liberals joined with the National Liberals and conservatives in an electoral coalition, the Bülow bloc, in support of the government’s colonial policies. What is contested is that the bourgeoisie was a “politically dependent” or even “supine”


245

partner in the coalition. Thus, on this account, Germany would appear more similar to Italy than the traditional view has admitted, though the aristocracy held a privileged position in the state, something not true of Italy. The main theoretical significance of this challenge to Moore’s analysis of German development is to question the view, as I have, that the bourgeoisie is the “natural” carrier of democratic politics. It does not, however, undermine his contention that agrarian class relations left an important legacy that helped undermine the push for democracy in Germany. Be it Calleo’s tipping the balance of power in favor of heavy industry or Eley’s aristocratic presence in the state, the Junkers still figure heavily in these accounts of the origins of Imperial German authoritarianism. Moreover, the additional explanations offered by the authors do not fare as well under comparative examination. The late industrialization—high concentration—strong Socialist labor movement argument does not hold in the Swedish case. Sweden had all these characteristics, and, indeed, Swedish business carried on an extremely militant and organized fight against unionization after the turn of the century, and it supported the Conservative party, which opposed democratic reform. However, the Swedish bourgeoisie did not have the option of allying with an agrarian upper class with an electoral base in the countryside. Thus, it was relatively isolated, and resistance to democratic reform was a less realistic option. Likewise, the equation of Germany’s imperialistic policies with the thrust toward authoritarianism is not unproblematic, because, as Calleo himself points out, other nations, notably Britain, engaged in imperialistic policies without the strong thrust toward domestic authoritarianism. This is not to deny that these were contributing factors—that late development, concentration, the threat of a strong labor movement, and so on did contribute to authoritarianism in Germany. Rather, I want to argue that they cannot displace the existence of a politically significant landlord class and the landlord-state-bourgeois coalition as critical features of the authoritarian path. The role of the agrarian upper classes and of the authoritarian coalition more generally in the breakdown of Weimar democracy is also a point of contention. For instance, these factors play a role but not a central one in Bracher’s (1970) important work on the breakdown of the Weimar Republic.20 It seems rather obvious that to complete Moore’s argument about the road to modern authoritarianism, it is necessary to specify how agrarian class relations and the coalition helped push Germany toward authoritarianism. In this section, I intend to provide such a specification as well as to highlight the prodemocratic role of the main working-class party, the Social Democrats. As was pointed out, on the eve of World War I, only the Social Democrats were supporters of parliamentary government and full suffrage reform at all levels of government. Thus, it seems eminently plausible to argue that the transition to democracy was a direct result of the war. The defeat in World War I, the discrediting of the ruling class, and the temporary power vacuum on the Right that this created changed the balance of class power in Germany. Labor organization surged to 30% of the labor force, and the democratic parties, which now included the Zentrum and the Left-liberal German Democratic Party (DDP), received 77% of the vote in the National Assembly elections of January 1919. Unfortunately, the Right recovered quickly; these three parties polled only 42% in the June 1920 elections and never again reached a majority during the Weimar


246

Republic (Rokkan and Meyriat 1969, pp. 157–58).21 Moreover, labor organization slid sharply backward in this same period. Without the defeat, it seems quite likely that Germany would not have become a democracy for decades, until something created a decisive shift in the balance of class forces. In the wake of the controversy over Abraham’s (1986) work and Turner’s (1985) concerted attempt to absolve big business from responsibility for the Nazis’ rise to power, the assertion that conscious instrumentation played an important role in the German case is bound to provoke a dispute.22 it must be said at the outset that the controversy focuses on the role of big business, and that the culpability of other elites, notably the Junkers and the closely allied top officers corps, is not a point of contention; thus, the most important point for Moore’s thesis is conceded. Indeed, Turner is eager to shift the blame to these elites, and few would contest the importance of their role in the conspiracies in the circles close to Hindenburg in the last days of the Republic, which led to the installation of Hitler as chancellor. These two groups, along with segments of heavy industry, were the backbone of elite support for and leadership in the German National People’s Party (DNVP), which was always monarchist and authoritarian and which cooperated increasingly closely with the Nazis after 1928. To mediate the controversy on the role of big business, one must clearly state what the question is. First, one might ask whether big business favored, and actively and consciously contributed to, the installation of a National Socialist government. This is Turner’s question, and he argues that, to the final days of the Weimar Republic, the politically active segments of German business continued to be disturbed by the socioeconomic radicalism of the Nazis and thus did not support the installation of a government led by the Nazis, though many leading businessmen favored a DNVP-led coalition with the Nazis. Second, one might ask whether the leading segments of business subjectively favored an authoritarian outcome or, third, in favor of it or not, whether they heavily contributed to it. I contend that the evidence presented by Turner, as well as that presented by Neebe (1981), indicates that both questions must be answered in the affirmative, though I hasten to point out that only an affirmative answer to the third question is important to my overall argument. As indicated above, most of heavy industry, primarily coal and steel, supported the DNVP throughout the Weimar Republic, a historical extension of its support for the conservatives and the rye-iron coalition of Imperial Germany. Other segments of business, export-oriented industry, finance, and so on tended to support the DVP (German People’s Party) or, in the case of the most liberal segments, the DDP. Under Streseman’s leadership, the DVP supported accommodation with the Social Democrats and a reluctant acceptance of the Versailles treaty. The former entailed acceptance of Sozialpolitik, government intervention in wage regulation, and other prolabor measures. With the onset of the depression and the death of Streseman, this segment of business gradually moved to the Right. While it initially favored the continuation of the Grand Coalition, it increasingly came to see Sozialpolitik and other prolabor measures as too costly, constituting, in the view of leading businessmen, the main barrier to renewed investment and thus recovery. Business at first had high hopes for the Brüning government but became disillusioned with it because it still relied on Social Democratic toleration and thus failed to move in a dramatically antilabor direction. By September


247

1931, almost all segments of business had become very critical of Brüning’s policies, though the top leadership of the important Reichsverband der Deutschen Industrie (RDI) continued to support the chancellor to the end of his period in power (Neebe 1981, pp. 102, 109). This drive to exclude the Social Democrats, combined with the rise of the Nazi vote, left business with two options: choosing either a parliamentary government with Nazi support or an extraparliamentary cabinet of the Right, either of which would move the government in an authoritarian direction. Turner (1985, p. 272; also see Neebe 1981, pp. 127–39) argues that the capitalists (including the RDI leadership) finally found their chancellor in Papen, whose government not only began to roll back the Weimar labor legislation and Sozialpolitik but also suspended the Prussian Landtag and drew up plans to revise the constitution in a decidedly more authoritarian direction. In all these developments, business used its money, political contacts, and media to influence events in the desired direction, and while hardly all-powerful, it is not plausible to argue that it was without significant influence. There can be little question that, objectively, the Papen government was a groundbreaker for the Nazis. Papen himself was a key actor in the formation of the first Nazi-led government and a minister in that government. Thus, there is little question that objectively business intervention in the political process contributed to the breakdown of democracy. What did leading businessmen subjectively support? Not Nazism, Turner argues convincingly. He indicates in passages too numerous to cite (e.g., see Turner 1985, pp. 252, 273–76) that they advocated a freeing of the cabinet from dependence on parties and a return to the “sort of mixture of freedom and authority that had prevailed in the Empire.” Business support for the Papen government was motivated not only by the economic policies of the government but also by the political “reforms” on the agenda (Neebe 1981, pp. 135–38). Thus, the evidence indicates that business favored a turn to more authoritarian forms of government. The argument that the effect of upper-class opposition to the sitting government—and ultimately the democratic regime—was mediated by structural mechanisms (lack of investor confidence, etc.) is strongest for the German case for two reasons. First, virtually all scholars concede that the economic situation was a very important contributor to the outcome in Germany, whereas the importance of this factor has been contested (at least by its omission) in the accounts of the other breakdowns. Second, because the attitudes of leading German capitalists toward the Müller and Brüning governments have been so thoroughly documented, we know that it is plausible to argue that lack of “business confidence” in these two governments did make an independent contribution to the dismal economic situation.23 The argument that the authoritarian posture of the middle classes and the peasantry is, in part, a product of upper-class ideological hegemony is important in the German case. Here one must explain why so many people were open to voting for the Nazis (37% in 1932) or, adding in the DNVP (6% in July 1932), for authoritarian parties in general. To facilitate the analysis of the rise of the Nazi vote as well as a comparison of the German and Austrian cases, it is helpful to examine the social and historical bases of the parties with the Lipset-Rokkan scheme (Lipset and Rokkan 1967). The DNVP was the political creature of the conservative monarchist bloc, which represented an alliance of the state (or “nation-building elite,” N, in Lipset and Rokkan’s terminology), the state church (C,


248

the Lutheran church in this case), and the landed upper class (L) (see Table 3). The liberal bloc, the DVP and the DDP, represented the bourgeoisie and other elements of the urban upper and upper-middle classes (U) and assumed a secular posture (S). The Catholic (R) bloc, Zentrum, and the Bavarian People’s Party, BVP, represented the largely south and west German Catholic population. The working class primarily supported the Social Democrats and secondarily the Communists. An examination of election results in this period indicates that the Nazis received votes from everyone who was not absorbed in the socialist/working class or Catholic countercultures (see Table 3). I contend that the authoritarian and militaristic ideology of the ruling groups of Imperial Germany contributed to the susceptibility of every other sector of the population to the reactionary appeals of Naziism. Under the influence of the increasingly desperate economic conditions of the Depression, these social groups turned from the traditional conservative authoritarianism of their old parties to the radical racist authoritarianism of the Nazis. My crude analysis here is confirmed by detailed empirical studies of two sorts. First, in their detailed examination of ecological data on voting patterns, Hamilton

Table 3: Elections in the Weimar Republic (%) January June May December May September November July March 1919* 1920 1924 1924 1928 1930 1932 1932 1933† National Socialists: NSDAP‡ – – 7 3 3 18 37 33 44 Conservatives (NCL): DNVP§ 10 15 20 21 14 7 6 9 8 Liberals (PSU) 23 22 15 16 14 9 2 3 2 DVPI° 4 14 9 11 9 5 1 2 1 DDP# 19 8 6 6 5 4 1 1 1 Catholics (PR) 20 18 16 18 15 15 16 15 15 Zentrum 20 14 13 14 12 12 13 12 12 BVP** – 4 3 4 3 3 3 3 3 Labor 46 42 34 35 41 38 37 37 30 SPD†† 38 22 21 26 30 25 22 20 18 USPD‡‡ 8 18 – – – – – – – KPD§§ – 2 13 9 11 13 15 17 12 Other 1 3 8 7 13 14 2 3 1 Source:—Rokkan and Meyriat (1969, pp. 157–59). (1980) and Childers (1983), though disagreeing on many points, concur that the Nazis failed to penetrate the socialist and Catholic countercultures. Second, Allen’s (1984) fascinating and well-researched community study of the Nazi seizure of power in one German town, based on newspaper reports, interviews with participants, and Nazi party documents, not only confirms the shift in voting patterns referred to above. It also


249

provides a clear picture of how German civil society in the period was organized, with a highly cohesive working-class, social democratic subculture promoting a democratic and socialist ideology, and a somewhat less cohesive, but nonetheless densely organized, middle- and upper-class culture interpenetrated by a large number of rightist, nationalistic, and militaristic groups promoting variants of that ideological current.24 My argument here admits that there is a grain of truth in the cultural interpretations of the collapse of German democracy. In its crudest journalistic form, this view contends that the Germans succumbed to authoritarianism because they were authoritarian. A slightly dressed up academic version of this is that the Weimar Republic crumbled because its culture was “illiberal.” Authors such as Bracher (1970) and Lepsius (1978) present a much more sophisticated variant of the argument by breaking the German society and polity into four blocs, contending that it was in the National Liberal and Conservative Monarchist blocs in which the voters succumbed to the radical authoritarian appeals of the Nazis and in which the leaders were willing to cooperate with them. This closely follows my own argument, which is not surprising since I have borrowed freely from them. The weakness of these points of view is that they either leave the strength of the authoritarian camp unexplained, as Lepsius does in his short essay, or explain it in historical terms without clearly connecting the content of ideology to the material and political interest of the economic and political leaders of the camps promoting them. Bracher (1970, pp. 27–28) hints at a connection between the promotion of the statistnationalist-völkisch ideology, the related Lebensraum philosophies, and the carrying out of the Kulturkampf against the culture of the Catholic southern Germans, on the one hand, and the interests of the political elite of Imperial Germany, on the other. But the mesh of this ideology with the interests and development project of the economic and political elite is nowhere made explicit. In Bread and Democracy in Germany, which is sufficiently close to Moore’s analysis of Germany to be termed the unwritten chapter in Social Origins on the German case, Gerschenkron (1943, esp. pp. 53–55) makes an explicit link between Junker class interests and their conscious and, in his view, successful attempt to spread proto-Nazi ideologies among small peasants in order to maintain and increase support for the Conservative party, the Agrarian League, and their agricultural program. Critics of the Moore-Gerschenkron view, such as Blackbourn, Eley, and Calleo, do not dispute the contention that this attempt to influence the social consciousness and political allegiance of the Protestant peasantry was successful. They contend that it had little effect on the urban middle classes. But they, particularly Calleo, offer an alternative source and conduit for ultranationalistic (though not backward-looking and agrarian) ideological influences on these strata. Following Kehr, it can be argued that the bourgeoisie’s support for imperialism had an independent economic basis and this led the bourgeoisie actively to promote supportive ideologies through the Naval League and other nationalist organizations.25 All three authors emphasize the role played by imperialism in the attempt of the liberals to secure a mass base in response to the rise of the Social Democrats. On the basis of the electoral and community studies cited earlier, I contend that these are specific instances of the general phenomenon of the penetration of the ideologies of the Imperial German elites (the state, Junkers, and the bourgeoisie—


250

Moore’s authoritarian coalition) to all social groups, in varying degrees, outside the working-class and Catholic blocs, thus strengthening authoritarian currents in these groups. Forces inside the state also played a role in the German case. With the strong overrepresentation of the landed aristocracy in the upper ranks of the Reichswehr, it is not surprising that the army shared the Junkers’ authoritarian monarchist politics, and though they never supported Hitler, their neutrality was essential for the Nazis’ seizure of power (Carsten 1973). In the eleventh hour, the period of presidential rule from 1930 to the seizure of power, the contribution of elements of this group went far beyond neutrality, as the circle around Hindenburg was completely dominated by the men drawn from the Junker-military-bureaucratic elite. The project of those closest to Hindenburg was to use the Nazis to install authoritarian presidential rule. This game eventually led to the Nazi entry into the government, with control of the chancellorship. The German judiciary consistently treated equivalent offenses by political groups on the Left and the Right quite differently. The light sentence received by Hitler for the Beer Hall putsch is a case in point here.26 This alignment of forces inside the state is also in part a product of the state- and nation-building coalition and the resulting imperialistic and militaristic ideologies propagated by state elites. The counterfactual in this case is important because potential critics might correctly point out that the Protestant peasants of northern Germany were one of the first groups the Nazis won over in the post-1928 period. Thus, one might ask, Why should we expect the movement to be much weaker if Germany were smallholding? Setting aside that it would be impossible to imagine the whole economic and political trajectory of Germany without the Junkers, we can turn to Finland for an answer. Under the impact of different but equally tumultuous events (revolution, civil war, depression), sections of the peasantry did turn to the radical right Lapua movement in the 1930s, and the movement did manage to have the Communist party suppressed (Alapuro and Allardt 1978; Alapuro 1980). But it never managed to effect the suspension of parliamentary politics. In accounting for its failure to reach this goal, Alapuro and Allardt cite several situational and structural factors. The most important of the structural factors, they contend, was the absence of a strong landed upper class. As a result, the Finnish peasantry had mobilized autonomously (as in all Protestant smallholding countries) and formed its own political party decades before the rise of the Lapua movement. Thus, when the Lapua movement attempted to translate its substantial support among peasants into electoral support at the polls, it failed to make more than marginal inroads into the support of the Agrarian Union. Austria Of the three European authoritarian cases not studied by Moore, the Austrian appears to be closest to the German. Certainly, the class-state constellations in the 18th and 19th centuries are quite similar. The basic state alliance was also among the crown, the army (which was very closely identified with the Habsburg monarchy), the bureaucracy, and the landed nobility. The ethnic divisions within the Empire and later Austria-Hungary split the nobility and often set the non-German nationalities, particularly the Magyar


251

magnates, against the monarchy, but as Taylor (1976) emphasizes, when they had to choose between defending their class privileges and advancing claims for national autonomy, they virtually always chose the former. The system of labor control in the countryside can be accurately labeled “labor-repressive”: though Joseph II abolished true serfdom, the Robot, obligatory labor service, which remained in effect until 1848, was actually the more critical provision in limiting labor mobility (Gross 1973, pp. 247, 255). And even in the post-Robot period, the great lords remained the predominant power in the countryside. Indeed, they gained at the expense of the minor noble landholders. The ethnic divisions within the Habsburg monarchy point to one contrast between Austria and Germany that has a bearing on my argument. For obvious reasons, the conscious promotion of a legitimating ideology in connection with an effort to increase national identification was out of the question for the ruling groups in the Habsburg monarchy, particularly in German Austria, the area of main concern here.27 Thus, while it can be plausibly asserted that the ruling groups propagated an ideology legitimating their rule and that the content was of necessity authoritarian, one has no obvious “smoking gun” as one has in the case of the German Kulturkampf and the campaigns of the Agrarian League and Naval League. The bourgeoisie was dependent on the Habsburg state. The state attempted to promote industrialization through high protective tariffs; subsidies, loans, and tax exemptions to businesses; armaments purchases; building of infrastructure; development of necessary skills and education in the population; and development and ownership of selected undertakings (Gross 1973, pp. 243 ff.). The abolition of the Robot strengthened the links between the landed nobility and the bourgeoisie since many of the nobles used the monetary compensation provided for the abolition to enter business on a significant scale (Gross 1973, pp. 255–56; Taylor 1976, p. 73). In the case of the German-Austrian bourgeoisie, this dependence on the state was reinforced by their (probably correct) view that the end of the Empire was inimical to their interests, even though it is arguable that the objective effect of their subordination was to impede the development of capitalism (see below). Gross (1973, p. 251) contends that the dependence of the German-Austrian bourgeoisie expressed itself in the “persistent adherence to quasi-feudal social values.” In a word, it accepted the ideological leadership of the monarchy and the landed nobility. The transition to democracy in Austria followed a path similar to that in Germany, with defeat in the war creating a temporary change in the balance of class power. The differences were that, on the one hand, the Socialist party and the working class were stronger, electorally and organizationally—the Socialist movement being the very prototype of the successful application of a Gramscian strategy. On the other hand, the Socialists were the only clearly prodemocratic party, with the possible exception of the Landbund. The various parties of the German Nationalist bloc (polling together around one-fifth of the vote), which were supported by a large segment of the bourgeoisie and the Protestant middle class, participated in governments in the late 1920s that turned a blind eye to, and sometimes even cooperated with, the fascist Heimwehr. Indeed, during German Nationalist leader Schober’s chancellorship in 1929–30, his office acted as a conduit for monthly contributions of 250,000 schillings from Austrian banks to the Heimwehr (Simon 1978, p. 96). Moreover, by 1932 virtually all the supporters of the German Nationalists had defected to the Nazis. The Christian Socials (polling around


252

40%) not only protected the Heimwehr while in office; the leadership of the party conspired to end democracy in 1933, and ultimately the party fused with the Heimwehr. On the question of instrumentation, the Austrian case is again similar to the German. There is no doubt that industry contributed to the Fascists; I have just mentioned the role of Austrian banks. Industrial interests nurtured and financially supported the Heimwehr from the outset, specifically as a force of coercion aimed at the Social Democrats and the trade unions (Gulick 1948, pp. 128–30). For example, the country’s largest mining and metallurgical firm, the Alpine-Montan Gesellschaft, contributed significant amounts of money to the Heimwehr (Rabinbach 1983, p. 55). However, perhaps more important for the fate of Austrian democracy was the support in the upper classes for non-Fascist authoritarian forces in the Catholic and German National blocs (Gulick 1948, pp. 693, 778, 858). The Austrian case differs from the German in that the landed upper classes themselves played little active role in the interwar events, though what contributions they did make were supportive of authoritarian forces (Gulick 1948, pp. 8–9). Part of the reason for this is that, compared with other regions in Austria-Hungary, German Austria contained a disproportionate amount of the mountainous areas, which were predominantly smallholding. Thus, the ideological legacy of past and present dominant elites is, if anything, more important for explaining the outcome in Austria than in Germany. On the basis of local election results, it has been estimated that well over 50% of Austrians would have voted for the Christian Socials (30%-35%) or the Nazis (over 20%) if a national election had been held in 1933 (Simon 1978, p. 110). The greater electoral strength of the authoritarian Right parties in Austria compared with Germany is explained by the authoritarian (or, at the very best, ambivalent) position of the Catholic camp. This, in turn, is explained by the historical alliance of the Habsburg state and the landed oligarchs with the Catholic church. This N-R (Roman Catholic)-L alliance in the LipsetRokkan scheme was the historical social structural basis for the Christian Social party. It represented a fusion of the organized Catholic institutional culture with this elite alliance, with the latter group defining the class character of the Catholic camp’s political ideology. Thus, the prodemocratic camp in Austria was weaker than in Germany owing to the differences in the historic alliances of the church.28 For precisely the same set of reasons, the appeal of Nazism was considerably less in Austria than in Germany, and the authoritarian Catholic corporative state of the Dollfuss-Schusschnigg period much less repressive, with considerable autonomy for trade unions and other forms of nonparty social organizations. There is one weakness in the argument linking the ideology of the dominant political coalition in Austrian development to the behavior of the parties in the interwar period. The development of the Christian Social movement substantially predated its fusion with upper-class interests (Boyer 1981). While it is true that the party shed much of its early radical romantic anticapitalism and became solidly conservative after the development of the alliance, its opposition to lower-class interests and universal suffrage and its antiSemitic authoritarianism date back to its early years. Thus, if its authoritarian posture is connected to the ideology of the ruling elites, the connection is not simple or obvious. A stronger case can be made for connecting the class-state coalitions of the 19th century to the outcomes of the interwar period in Austria by pointing to the link between


253

the development coalition and the weakness, and antidemocratic posture, of Austrian liberalism. Austrian liberalism was based in the German-Austrian bourgeoisie and uppermiddle classes and thus reflected the conservative position of those groups, which was directly linked to their perception of their own self-interest in preserving the empire and the Habsburg monarchy, as mentioned earlier. Liberalism in Austria meant anticlericalism, rationalism, support of property rights, and little more. It lacked the democratic tendencies of English liberalism, not to speak of the strong support for democracy characteristic of Scandinavian liberalism.29 Thus, the dependence of the bourgeoisie on the Habsburg state helps to explain the weakness of democratic currents outside the working class in the first Austrian Republic. The Czech case presents a nice counterfactual on this point. For obvious reasons, the bourgeoisie in Bohemia and Moravia did not see the preservation of the Habsburg state in the same light as their German-Austrian counterparts and thus developed a more aggressive and stronger liberalism. This is one factor that helps explain why democracy survived in Czechoslovakia until the German invasion, in contrast to Austria. The case of Czechoslovakia also puts into perspective the arguments that the decisive factor sealing the fate of Austrian democracy was foreign pressure from its authoritarian neighbors. While it cannot be denied that this was a significant factor, Czechoslovakia was similarly surrounded by authoritarian regimes and subject to pressures from them but did not succumb. Moreover, this argument is weakened because the trajectory toward the authoritarian seizure of power in Austria began with the events of 1927, well before the rise of Nazi influence in Germany, and the critical step toward the installation of an authoritarian regime, the suspension of parliament, occurred only a few months after Hitler was appointed chancellor and before the Nazis had consolidated power. In Moore’s work, the authoritarian coalition appears as a modernizing coalition promoting industrialization and economic growth. Taking Germany as an explicit point of comparison, Gross (1973, pp. 249–51) argues that state intervention in Austria failed to further industrial development adequately, in sharp contrast to the German case. He argues that it was precisely the nature of the ruling coalition that prevented the Habsburg state from intervening effectively to promote industrialization. “Since the main supports of the Empire and its dynastic regime were the aristocracy (and the hierarchical principle in general), the Church, the army and bureaucracy, as well as the interests of the ruling social-ethnical classes in certain provinces—a policy of rapid modernization and of promoting meritocratic social trends was out of the question. Moreover, although most Austrian governments in the second half of the [19th] century included promotion of industry in their policy-mix, each according to its fashion, none could pursue rapid industrialization as a national objective. This, first of all, because no ‘national’ policy, in the strict sense, was to be envisioned” (Gross 1973, p. 250). Some of the factors mentioned by Gross would apply equally to Prussia-Germany, and it is plausible that they did impede the development of capitalism there, but they were more than counterbalanced by the strong and effective support of the Junker-military-bureaucratic leadership, in close cooperation with private industry, for industrialization and economic modernization more generally. The nationalities question, Gross argues, prevented the Habsburg monarchy from taking on such a project, and it impeded greater independent pressure from the German Austrian bourgeoisie, as mentioned previously.


254

Spain There is no doubt that landholding in absolutist Spain was extremely concentrated: the nobility and the church held over two-thirds of the land in the country (Carr 1982, p. 39). But, as in the Italian case, the initial development of the ruling political coalition in the 19th century does not seem to make Spain a very good case for Moore’s explanation of the rise of modern authoritarianism. The conflicts of the first decades of that century pitted the urban middle and upper classes, intellectuals, and the army—supporting a program of liberal reform—against the church, nobility, and monarchy. An important part of the liberal program was an attack on corporate property rights—entailed property, seignorial rights, church-owned lands, and communal property—thus directly attacking the interests of the nobility as well as the church. In the course of the century, however, the coalition of forces that gradually assembled around the Moderado Liberal (later dynastic Conservative) party did resemble the Moorian coalition.30 By the time of the first Carlist wars in the 1830s, the crown was aligned with the Moderados, and increasingly most elements in the army supported this party, though the Progresista Liberals continued to enjoy some army support until late in the century. In their periods in power in the 1830s and 1850s, the Progresistas carried through the liberal property-rights reforms of the original Liberal program, which resulted in a very significant redistribution of land, as church properties were sold by the government and the end of entailment led many poorer nobles to sell their land. Since this land was sold on the open market rather than used as part of a land-reform scheme, it was those who had the money to buy it who benefited: the affluent nobility, the upper peasantry, the local political bosses in the rural towns, and the bourgeoisie. The larger landowners, both the traditional nobles and the new landlords, also increasingly gravitated to the Moderados-Conservatives. By the time of the Restoration (after 1874), sections of the Catholic Right, in the form of the Catholic Union, had broken with Carlism and joined the dynastic Conservatives. With the Catholic Union came the episcopate and “the more recalcitrant of the Catholic aristocracy” (Carr 1982, p. 355). These social supports of the Conservatives increasingly became a single social class with a common social and political outlook. Carr (1982, pp. 284, 431–32; also see Linz 1967, p. 204) sketches the situation at mid-century and then at the end of the century: [The generals] were absorbed into the aristocracy by a continuous process of new creation [of titles]…. The amalgam of speculators, industrialists, landowners, together with the prosperous lawyers and ennobled generals who were its political voice par excellence, constituted what democrats were beginning to call a ruling oligarchy—estimated at five hundred families. [The aristocracy] was conspicuous in its support of Catholic values. Since its whole history in the nineteenth century had been one of accretion, through the incorporation of successful soldiers, hauts bourgeois, and politicians, it tended to impose these values on the upper ranges of society as it imposed its way of life.


255

The dynastic Conservatives’ opposition within the El Turno Pacifico system was the dynastic Liberals, the heirs of the Progresistas. The Liberals were based in the provincial urban upper-middle classes and, on paper, supported liberal democratic ideals. When they were in office in the last decades of the 19th century, they put these ideals into law, making Spain, on paper, a democracy: universal suffrage, freedom of association, freedom of the press, and so on. But the reality of the situation was quite different: in the El Turno system, there was a de facto agreement that the two dynastic parties would alternate in power, and this was accomplished by fixing the electoral outcome. The parties outside the system, the Republicans and the Socialists, had no chance to gain power despite the existence of universal suffrage and the numerical significance of the classes they represented. Thus, the political situation under El Turno was very similar to that under trasformismo in Italy. The parallel does not end there. Like Italy, Spain, in 1891, introduced tariffs on both agricultural and industrial imports. Like Italy, in a conscious effort to create a national industry, it embarked on naval building programs in 1888 and again in 1908. Thus, it is probable that all these developments had the same effect as in Italy: to bring the state, the bourgeoisie, and the landed classes into closer alliance. Spain in the closing decades of the last century shows some other interesting contrasts and similarities to Germany and Italy. In both Spain and Germany, the upper officers corps were closely linked with the upper class(es), but in quite different ways. In Germany, many top officers were traditional aristocrats. In Spain, the officers were often of humble origin, and an army career could be a way to move into the political elite (owing to the role of the military in Spanish politics) and then the economic elite, sometimes ending in ennoblement. A second contrast to Germany was that the landlord class was a mix of traditional aristocrats, ennobled haute bourgeoisie and military men, and untitled property holders. As in Italy, it cannot be said that the bourgeoisie was dependent on a landholder-dominated state. The state’s role in development was yet more passive than in the Italian case. Indeed, it would be stretching the facts beyond belief to call any of the Spanish regimes of the 19th or early 20th centuries “modernizing” in Moore’s sense: this is hardly a case of “revolution from above.” Thus, like the Austrian case, the Spanish case argues for an important revision of the Moore thesis. A rulingclass alliance of labor-repressive landlords, sections of the bourgeoisie, and the state did materialize in the second half of the 19th century, but it hardly had the modernizing project of its German counterpart. The process of economic development in the period 1875–1920, and thus the increasing size of the urban lower-middle and working classes, put pressure on the El Turno system, as it increased the strength of the unions and Socialist and Republican parties, as well as the Anarchists. The economic boom produced by the war in particular greatly strengthened labor unions. Indeed, it became more and more difficult to fix elections in the urban areas, and the votes of the urban electorate became known as “votos verdad.” A combination of public revulsion at the excesses of El Turno and widespread indifference to the overthrow of the parliamentary system contributed to the success of the coup of Primo de Rivera in 1923, ushering in a period that the dictator himself said would be a “brief parenthesis” to clean up the corruption of El Turno (Carr 1982, p. 564). The institution of the Republic, and thus democracy, came as a result of


256

the victory of the Republicans and Socialists in the larger cities in the municipal elections of 1931 and the subsequent proclamation of the Revolutionary Committee, composed of Republican and Socialist conspirators. The passive role of the army was critical; it did nothing to stop the republican conspirators. The transition to democracy in Spain is rather like the same process in Italy, as the working-class forces were the beneficiary of the introduction of democracy more than the initiator of it.31 The principal political representative of the working class, the Socialists (PSOE), played a secondary role in the transition to democracy, as they unified behind the Revolutionary Committee only in the eleventh hour. The Socialists garnered a somewhat smaller portion of the vote than in Italy, approximately 20%-25% (Linz 1967, p. 225; Linz 1978, pp. 146–47). And even more than in Italy the immediate reaction of the upper classes (i.e., with respect to the instrumentation argument) to the political mobilization and trade-union organization of the working class and landless agricultural workers was a central, if not the central, dynamic in the breakdown of the regime.32 Not only did large estates dominate the countryside more than in the other three cases discussed here; it was also a very heavily agricultural country. Thus it is not surprising that the agrarian question was the focal point of class conflict, and that the hostility of the large landholders to state intervention in local agrarian labor markets, which strengthened the hand of labor, and to the modest agrarian reforms of the first Republican government was such an important factor in the breakdown (Malefakis 1970). By the same token, given the low level of capitalist development of the country and thus the weakness of civil society, ideological hegemony of the authoritarian (upper-class) forces did not play such a crucial role in the Spanish case. There is a sense in which the Republic was a premature progressive development that was quickly rolled back because the democracy itself and the reforms it introduced exceeded the limits the balance of class power in society would tolerate. The Republic was ushered in by developments in urban Spain and did not reflect the power balance in Spanish society as a whole. Thus, it was quite similar to a number of revolutions in 19thcentury Spain that emanated from the urban areas and resulted in the installation of a progressive government and the promulgation of a liberal constitution or the like and then were quickly, though not completely, rolled back. The Republic was unusual only in that its accomplishments were completely destroyed, and this ushered in five decades of dictatorship. The Spanish case is exceptional in that the most crucial element of the state, the army, played not simply a passive role in the authoritarian seizure of power but rather was the instrument of seizure. Again, this was a continuation of a 19th-century pattern of Spanish politics rather than a new development.

Conclusion and Discussion The central task of this article has been to evaluate Barrington Moore’s (1966) thesis on the role of agrarian class relations and historic state-class coalitions in the paths to capitalist authoritarianism and democracy through a comparative-historical analysis of the transition to democracy in Western Europe in the period 1870–1920 and the


257

subsequent breakdown of democracy in the interwar period. Other theories of the transition to democracy were considered, and, before I proceed to a summary of the evidence on Moore’s thesis, some reflection on these theories is in order. The overview of the transition to democracy confirmed Therborn’s (1977) contention that the working class, represented by socialist parties and trade unions, was the single most important force in the majority of countries in the final push for universal male suffrage and responsible government, though in several of the smallholding countries the small peasants or the urban middle class played the major role. This contradicts the modernization view, as advanced by Lipset (1960), which argues that economic development and democracy are connected primarily through the expansion of education, the growth of the middle class, and so on. Rather, it was the growth of the working class and its capacity for self-organization that were most critical for the final breakthrough of democracy. The rapid industrialization experienced by Western Europe in the five decades before World War I increased the size and, with varying time lags, the degree of organization of the working class and thus changed the balance of class power in civil society to the advantage of democratic forces. But, by focusing only on the final step of the process, Therborn rather exaggerates the role of the politically organized working class. The working class needed allies in the final push, and earlier suffrage extensions that incorporated substantial sections of the lower classes, rural and urban, were often led by other social groups, usually the urban middle class or small peasantry, with the working

Table 4: Social and Historical Factors Leading to Authoritarianism GermanyAustria Italy Spain Britain France Small Countries Landed upper class yes yes yes yes yes no no politically very significant Historically yes yes yes yes no no no engaged in laborrepressive agriculture more Bourgeoisie strong ? yes more yes more more enough to be powerful powerful powerful powerful politically very than by 1900 significant, but not landed more powerful than interests landed class Bourgeoisie ? yes no no no no no dependent partner in coalition “Strong” state yes yes ? ? no ? no Revolutionary no no no no yes yes no break from the past


258

Note.—All evaluations of strength of the various forces, etc., are for the last half of the 19th century. class playing only a supporting role. Moreover, the contribution of the working class to these reforms came in the form of artisanal agitation as often as through the action of the Second International parties emphasized in Therbon’s work. The important role of working-class segments of confessional and liberal parties is also somewhat underplayed in his analysis. And finally, where the working class had few allies (e.g., only the Catholic parties in Germany and Italy) or none (Austria), democracy was fragile and did not survive the interwar period. The bourgeoisie, which appears as the natural carrier of democracy in the accounts of orthodox Marxists and, to a certain extent, of Moore hardly lived up to this role. It almost universally opposed the final extensions of suffrage to the working class. Its contribution was to establish parliamentary government, and it did not do even this in all cases. The bourgeoisie’s resistance to the initial political incorporation of the working class and its support for working-class exclusion in some countries in the interwar period were clearly connected to working-class support for socialist parties. This factor introduces a subtle wrinkle in the argument linking the strength of the working class to introduction of democracy. Certainly a significant portion, if still a minority, of the working class in Italy, Germany, and Spain supported the communists, who were authoritarian in this period, or the maximalist socialists, who were ambivalent about democracy, and there is little doubt that these parties or factions of parties contributed to the breakdown of democracy. However, the example of Austrian social democracy, which was powerful, clearly committed to socialism, and democratic, indicates that it was not only, or even mostly, these parties’ and factions’ attitudes toward democracy but rather their demands for socialism that provoked an upper-class reaction This seems to suggest almost the reverse of Therborn’s thesis: where the working class was well organized and committed to a moderate-to-radical socialist party, it hindered the development of democracy. Yet, the examples of Norway, where the Social Democrats were as strong as in Austria and more radical in the late 1920s, and France, where the communists were as strong as or stronger than they were in Spain, Germany, or Italy, argue against such a generalization. The relationship between working-class strength and democracy may be summarized in the following way. A diachronic analysis within each of the Western European countries reveals that the growth of working-class organizational strength led to increased pressure for the introduction of democracy; a synchronic analysis across countries reveals that these pressures led to the development of stable democratic regimes where the working class found allies in other social groups. The overall picture comes closest to the view of Rueschemeyer (1980), who argues that economic development strengthens both the middle and working classes, thus leading to the strengthening of democratic forces. This view accounts for the essential elements of the process of transition in the countries that experienced an internally generated transition to democracy and in which democracy survived the interwar period. However, as the breakdown cases demonstrate, the middle classes are not invariably democratic forces. The middle classes and the peasantry played quite different roles in


259

different countries. In some, such as the Scandinavian countries, they supported suffrage extension and allied with the working class. In others, such as Germany and Austria, they formed the mass base for authoritarian movements that ended democracy, hardly living up to the heroic democratic role assigned to them in modernization theory. Explanation of the differences among countries is where the Moore thesis comes in. Agrarian class relations appeared to be the critical feature distinguishing those cases in which democracy broke down and those in which it survived: in all four countries in which authoritarian regimes replaced democracies, a politically powerful body of large landholders had survived into the 20th century, and historically these landholders were engaged in what Moore calls labor-repressive agriculture (see Table 4). None of the other countries fit this description: in the single other largeholding case, England, the large landholders did not employ labor-repressive techniques of labor control, and in the rest of the countries the countryside was dominated by smallholders. In the detailed analysis of the four authoritarian cases, I attempted to show that an alliance, or at least an accommodation, did develop among the state, labor-repressive landlords, and the bourgeoisie. However, in no case did the authoritarian coalition develop exactly along the lines outlined in Social Origins. The weakest point in Moore’s analysis is his characterization of the bourgeoisie as the dependent partner in the coalition (see Table 4). In Italy, the bourgeoisie, not the landholders, was the politically dominant segment of the upper classes. Recent historical scholarship on Imperial Germany questions the view that the bourgeoisie was politically dependent. Even in Spain, where the landed class was dominant in sheer economic terms, it cannot be said that the bourgeoisie accepted the political leadership of the landed classes. Only Austria seems to fit Moore’s characterization, and there, the political and economic dependence of the German-Austrian bourgeoisie was, in the final analysis, cemented by its position in the multiethnic state, and this dependence was not shared by its Bohemian and Moravian counterparts. It appears that Moore’s analysis on this point suffers from the unwarranted assumption that the capitalist class has natural prodemocratic tendencies emanating from its economic interests Another problematic point in Moore’s analysis, and even more so in Skocpol’s, is the argument that state “strength” is a necessary feature of the authoritarian path. While there is no question that forces inside the state, particularly the coercive forces, contributed to the rise of authoritarian regimes in the interwar period, it is questionable whether one could describe the state in Spain and Italy as “strong” if one means something more than that it had the sheer repressive capacity to introduce authoritarian rule, a point to which I will return shortly. In order to evaluate fully the importance of the six factors listed in Table 4, it is necessary to examine the cases in which democracy survived. They are strikingly different. For reasons discussed by Moore (and cited in the text), the British and French cases are different from the typical authoritarian path. The small European democracies are diametrically opposed to Moore’s paradigmatic case. It is worth emphasizing how much of this difference can be traced to the historic patterns of landholding and agrarian class relations. First, as was pointed out earlier, size, landholding patterns, and military strength were historically interrelated. With no large landholding class, labor-repressive agriculture was impossible (point 2 in the table); the bourgeoisie was ipso facto stronger than the landed class (point 3); and, for the same reason, it was not the dependent partner


260

in a coalition. Because there was no agrarian oligarchy, a revolutionary break from the past was not necessary. Not only is state repressive capacity connected to the size/greatpower status/ military strength complex, it also may be more directly related to agrarian class relations, if the Swedish case is typical. There, Tilton (1974, p. 568) argues that insufficient repressive capacity in general, and the absence of a standing professional army in particular, did play a role in the calculations of the conservatives at the point of their final capitulation in 1917. The weakness of the repressive apparatus, in turn, can be attributed in part to the influence of the peasants in the late 19th and early 20th centuries, as they used their influence in the lower house to block appropriations for defense, since these were connected to taxes that would fall on their backs (Rustow 1955; Verney 1957). Turning now to how the legacy of agrarian class relations and the authoritarian coalition more generally affected the events of the interwar period, I proposed the distinction between instrumental effects, that is, active intervention in the historical process by the actors in question, structural constraints on government policy, and hegemonic effects, that is, legacies (and re-creations) of the ideological hegemony of the ruling coalitions, past and present, which are particularly important in explaining the behavior of the middle classes, urban and rural. In Italy and Spain, active intervention by landlords and capitalists in support of authoritarian outcomes was found to be of great importance. It cannot be overemphasized how critical the role of agrarian elites’ attempts to maintain the control of rural labor were in these two cases. For Germany, it was argued that all three factors mediated the effect of the historic developmental coalition on the interwar events. The weakest case of the four for the Moore thesis is Austria(-Hungary). Though it was most similar to Prussia-Germany in terms of the development of the state-class alliances in the 18th and 19th centuries, the effect of those alliances on the breakdown of democracy in the interwar period was less clear than in the other three countries. The strongest argument, and one that I believe will stand the test of further study, is that the dependence of the German-Austrian bourgeoisie on the Habsburg state explains the weakness and conservative character of Austrian liberalism, and this in turn helps explain why the democratic currents outside the working class were so weak in Austria. The contrasting role played by the mechanism of ideological hegemony in the four cases, its importance in Austria and Germany as compared with Spain and Italy, is obviously a function of the level of economic development and thus of the strength of civil society. A second contrast among the four cases has been suggested by Stein Rokkan (personal conversation, 1974): it was in the two late nation builders, Italy and Germany, that the hypernationalism of fascism was ideologically dominant, whereas in Spain and Austria the ideologically dominant current was clerical corporate authoritarianism. Rokkan argued that the overlapping of nation building and mass mobilization created a climate favorable to the development of mass hypernationalism. One can add to this the experience of World War I, which was also a component in the fascist trajectory, as it was directly related to the development of the right-wing paramilitary organizations that fed into fascism. Thus, as I argued in my initial discussion of Moore’s work, fascism was only one form of modern capitalist authoritarianism; it is not equivalent to it, as Moore indicates.


261

This discussion of the differing ways in which the historic authoritarian coalitions influenced interwar events helps explain why the analysis here comes to conclusions diametrically opposite to those in the recent work of Luebbert (1987) covering similar cases and a similar period. Using percentage of the labor force in agricultural labor as the indicator of rural social structure, Luebbert contends that there is no correlation between rural social structure and regime outcome. The agrarian elite strength measure used here, which is a product of the underlying distribution of landholdings, shows something quite different. Luebbert then correctly points out that, in the arguments of Gerschenkron and Moore, the reason why landholder power resulted in an authoritarian outcome in Germany in the interwar period is, in large part, that they still exercised political control over the rural masses in this period. This path was not replicated in the other cases, he contends. Therefore, he concludes, landlord strength cannot explain the breakdown in these other cases. My objection to this was made clear in my initial discussion of Moore; this is simply too severe a test of Moore’s thesis given that he discusses interwar events only tangentially. The mechanism suggested by Moore and Gerschenkron does not exhaust the ways in which late 19th-century coalitions and landlord power expressed themselves in the events of the 1920s and 1930s. To put the results of the analysis in a single statement, I conclude that the agrarian class relations and patterns of state-class alliances of the 19th and early 20th centuries were necessary though not sufficient causes of the breakdown of democracy in interwar Europe. The existence of a large landed class changed the alliance options for other classes in both the late 19th century and in the 1920s and 1930s and as a consequence changed the political outcomes. It both opened up authoritarian options for the bourgeoisie and, to the extent that the landlord-state-bourgeois alliance affected the politics of the middle class and peasantry, closed off options for the working class. Had the country in question had different landholding patterns and resultant state-class relations, democracy would have survived. As I pointed out in the discussion of Germany, this counterfactual will appear to many to be absurd, since those patterns of agrarian class relations and state-class alliances are an integral element of one’s understanding of the essential features of German (Italian, Spanish, Austrian) economic and political development. In a sense, this is precisely my point. The analysis indicated that Kurth’s explanation of the paths to democracy and authoritarianism has to be considered supplementary to Moore’s and not an alternative to it. The timing and phases of industrialization reinforced tendencies created by legacies of agrarian class relations but could not override them. On the positive side, the thesis does elucidate some features of the four authoritarian cases and highlight differences between them and Britain and France. Late industrialization in the former four cases was associated with tariff policies and state intervention in the economy that facilitated the formation or strengthening of Moore’s landlord-state-bourgeoisie coalition, and in every case this was further reinforced by an armaments policy that made heavy industry very dependent on the state and imperialistic policies. Thus, in a fashion that dovetails with the arguments of Calleo and Blackbourn and Eley, Kurth’s argument supplies an economic explanation for the posture of the bourgeoisie in these countries, an alternative to Moore’s dependence on the landlords argument, which did not fare well in the present analysis.


262

However, Kurth’s argument must be considered supplementary to Moore’s because it holds only in countries characterized by the presence of a significant landlord class, as an examination of the small democracies reveals. Most of the small democracies were late industrializers, and most industrialized with tariff protection for industry and agriculture. At least some of them were characterized by some of the features typical of German industrial structure, which in Eley’s account form one link between late industrialization and authoritarianism. Here again, the Swedish case is instructive. In a manner consistent with the Kurth and Eley theses, by World War I, virtually the entire Swedish bourgeoisie had shifted into the Conservative party, the most antidemocratic of the Swedish parties. Indeed, in the spring of 1914, after a bitter conflict over constitutional government, the king appointed a royal cabinet that contained many of the leading figures of Swedish industry (Söderpalm 1969, p. 24). But the Conservatives’ support was limited and they had no allies in other parties; thus capitulation to the democratic forces was almost the only alternative. The political independence of the Swedish peasantry is clearly a key reason for the isolation of the Conservatives. Kurth’s argument on the importance of the saturation phase of steel production and imperialism might be saved by introducing a geopolitical condition along the lines of Calleo’s thesis on the German case. That is, it might be argued that imperialism was a viable alternative only for countries in a position to aspire to great-power status, and thus the argument simply does not apply to small countries. Without denying the importance of armaments production and imperialism in explaining the trajectory of the authoritarian cases, it can simply be pointed out again that the British case shows that imperialism does not imply a push toward domestic authoritarianism. The one point on which the analysis of the three cases other than Germany appears to fit the Moore thesis poorly is that it is questionable whether the authoritarian coalition in any of them was modernizing, even in a strictly economic sense. This was most clear in the case of Spain, but economic historians also contend that Austrian and Italian development was impeded rather than promoted by the way the state intervened in the economy. It might be argued that this assessment is a product of the neoclassical biases of many economic historians. To a certain extent, this is the case: generally the tariff policies of all these countries, including Germany, are assumed by the economic historians to have had a negative effect on economic development. As Senghaas (1985) demonstrates in his comparative-historical study of European industrialization, the reality is that free-trade-oriented industrialization was very much the exception rather than the rule. Most countries went through a period of protection, which was necessary for successful industrialization, as it allowed the domestic market to develop and encouraged forward and backward linkages in the economy and thus self-sustaining growth. Still, even if one rejects an orthodox condemnation of all state intervention as being counterproductive, there is reason to believe that the type of intervention typical of regimes in which large landlords played a significant role impeded economic development. This includes but goes well beyond both the usual “feudal barriers to capitalist development” line of argument found in both Marxist (e.g., Brenner 1977) and neoclassical economic literature and the reasons connected to idiosyncratic features of the particular country.33 Senghaas’s comparative study again offers an insight on this point. He contends that agricultural modernization was a necessary feature of all successful


263

developers because of the macroeconomic function of agriculture as a market for manufactured goods and because of industry’s contribution to rising agricultural productivity (Senghaas 1985, pp. 46–52). Land concentration, and thus income concentration, impeded the former. Senghaas goes so far as to argue that the response to the British industrial challenge was predetermined, as it was in those countries with autonomous processes of agricultural modernization and a relatively equal distribution of assets in agriculture that met the challenge with successful industrialization of their own. With this in mind, Germany becomes something of an anomaly, since it was characterized by a powerful landed aristocracy, but it experienced rapid and selfsustaining industrialization in the 19th century. A possible answer to this problem is that Germany did contain substantial areas of medium and small peasant agriculture in the South and West, which could serve the internal market function and as a source for linkages between the agricultural and industrial sectors, two of the functions of this group in Senghaas’s account. And, indeed, it is true that German economic development followed an east-west gradient, with the West being the most developed and the East most backward (with some exceptions, e.g., Saxony and Silesia). Nevertheless, there is little doubt that the Prussian and Imperial German bureaucratic state in which the Junkers had great influence did intervene effectively in developing the national economy. Thus, paradoxically, if the Junkers had dominated the entire rural economy of the country, the national economic development project, which they supported, might not have succeeded. This point also leads us to a revision of Skocpol’s argument that a strong state was a necessary component of the authoritarian coalition analyzed by Moore. My analysis of Spain (and to a lesser extent this could be said of Italy) suggests that a strong state was necessary only if the coalition was attempting to carry out a “revolution from above,” a project of rapid economic development.34 It was not a necessary feature of the path to modern authoritarianism in interwar Europe. Clearly, the state had to have sufficient repressive capacity to install authoritarian rule. Moreover, the analysis here showed that the state played an autonomous role in the events of the 1920s and 1930s: the actions of state forces, particularly the coercive forces, contributed to the authoritarian outcome in all four cases. Thus, though autonomous state action contributed to the outcome and adequate state repressive capacity was essential to it, the capacity to intervene effectively in the economy, to develop modern skills in the population, and so on was not essential for the installation of modern authoritarianism.

Notes to Table 3 * Elections for the National Assembly † Conducted by the Nazi-Nationalist government with significant harassment of the opposition. ‡ National Socialist German Workers’ Party § German National People’s Party ° German People’s Party # German Democratic Party


264

** Bavarian People’s Party †† German Social Democratic Party ‡‡ German Independent Social Democratic Party §§ German Communist Party

Notes 1. An earlier version of this paper was delivered at the meetings of the American Sociological Association, Chicago, August 1987. I would like to thank the following people for comments on earlier drafts of this paper: Ronald Aminzade, Valerie Bunce, James Cronin, Gøsta Esping-Andersen, Peter Evans, Roger Friedland, Daniel Garst, Jack Goldstone, Peter Hall, Peter Katzenstein, Guillermo O’Donnell, Timothy Power, Charles Ragin, Dietrich Rueschemeyer, Evelyne Huber Stephens, Arthur Stinchcombe, Samuel Valenzuela, and the participants in the seminar The State and Capitalism since 1800, Center for European Studies, Harvard University. Much of the research on this paper was completed while I was a faculty fellow at the Kellogg Institute, University of Notre Dame, and I would also like to thank the institute for its support. Requests for reprints should be sent to John D. Stephens, Department of Political Science, Northwestern University, Evanston, Illinois 60208. 2. I use male suffrage as the critical transition point because it best signifies the point at which the democratic rights of the masses become (or come to be seen as) a threat to the upper classes, as it became possible to put together an electoral majority of dispossessed classes. Initially, female suffrage actually had a conservatizing effect in many cases. See Therborn (1977) on the relationship between the introductions of female and male suffrage. 3. Moore implies that labor-repressive methods of labor control are very likely in cases where large landlords oriented to the market have a need for a large agrarian labor force. It should be emphasized that while slavery and serfdom are the most obvious examples of laborrepressive methods, legal binding of labor to the land or the master is not a necessary feature of labor-repressive agriculture. The postbellum U.S. South is an excellent example of a system that, while certainly labor repressive, did not employ such features. 4. It is clear that Moore argues that the bourgeoisie’s dependence on the state and the agrarian elite causes the class to absorb the values of the landed class, but it is not clear that he means that the bourgeoisie thereby adopts a political posture that is not in its interest, as Blackbourn and Eley (1984) imply, grouping Moore with a number of theorists, such as Dahrendorf (1967), who hold this position. Moore’s work can just as easily be read as indicating that the bourgeoisie’s interests change in such a situation because of state aid, etc. Nevertheless, Blackbourn and Eley are clearly correct in asserting that Moore assumes that, in the absence of a powerful landed class closely allied with the state, the bourgeoisie will “naturally” adopt democratic politics. 5. The classification of landed class strength in Table 1 is based on the comparative and country studies consulted for the analysis of the cases. Dovring’s (1965) data, by far the most comprehensive compilation of data on landholding at the turn of the century, strongly indicate that the political position of large landlords in the society is mainly a product of their economic position, which in turn can be read off from landholding patterns. His data are not conclusive, however, because they do not control for land quality (distorting, e.g., a comparison of England and Spain) and because the data for some countries include crownor state-owned lands, greatly skewing the distributions. It should be emphasized here that the comparative and case studies consulted revealed virtually no dissent from the classification


265

presented in the table. 6. The Congress of Vienna and subsequent power-balance politics were especially important for the preservation of Belgium and the Netherlands. It should be noted that the association of agrarian elite strength with country size/great-power status is a historical generalization about Europe that does not hold in regions where external war was not such an important feature of state formation (e.g., Latin America). 7. It should be noted that the following discussion of the transition to democracy is based not only on the final reform establishing democracy as I have defined it but also on the previous steps that resulted in suffrage extension to the majority of the working class or the establishment of cabinet responsibility to parliament (e.g., the 1884 suffrage extension in Britain or 1901 establishment of parliamentary government in Denmark). 8. The effect of this change in power relations already made itself felt on the material welfare of the mass of noncombatant working people in Britain. Winter (1986) shows conclusively that life expectancy rose and infant mortality declined dramatically during the war and that that was a result, primarily, of the improved material situation of the working class in the war economy and, secondarily, of health and welfare policy innovations. 9. On Sweden, see Rustow (1955), Verney (1957), Hadenius et al. (1970), Castles (1973), and Tilton (1974). Tilton and Castles both explicitly apply the Moore thesis to the Swedish case. Both point to the weakness of the landed aristocracy and the strength and independence of the peasantry in accounting for the democratic outcome in Sweden. Tumin (1978) makes a similar point in his application of the Moore thesis to the Dutch case. 10. The reason for the late development of a significant working-class party in relation to the level of industrialization, a characteristic that England shares with France, may be that both were earlier industrializes. The nature of industrialization in these countries (see the discussion of Kurth above), which created small dispersed work units, may be one reason for this. Another obvious candidate is ideological diffusion among the late industrializes. For instance, earlier labor agitators in the Scandinavian countries and the Netherlands were heavily influenced by German social democracy and later the Second International. Thus, the initial organizers of the working class were socialists and, vaguely, Marxists, in contrast to the British case. 11. And, in addition, Finland (see below). 12. Sources for figures in this paragraph are: Mitchell (1978) for labor force, Rokkan and Meyriat (1969) for electoral statistics, and Stephens (1979) for union membership. 13. In his defense, it must be said that the roles attributed to the working class and the bourgeoisie by Moore are much more accurate if one accepts his definition of democracy (Moore 1966, p. 414), which does not clearly require that even a majority of the adult male population be entitled to vote. 14. In the interwar period this generalization about the working class is harder to sustain, since the splits in the working class induced by the war and the Russian Revolution created antidemocratic minorities, above all the communist parties, whose political posture clearly contributed to the breakdown of democracy. Moreover, as Linz (1978) points out, the radical posturing of maximalist socialists frightened the middle classes, which contributed to the strengthening of the authoritarian forces; even the moderate social democrats contributed to the outcome by inflexible postures vis-à-vis parties of the center. This said, I think it is fair to state that all the parties of the social democratic Left, which remained by far the largest of the working-class parties in every country, maintained a commitment to democracy. Their mistakes do not make them antidemocratic. 15. The passage in the text is the brutal summary of my view of the state in advanced capitalist


16. 17.

18.

19.

20.

21.

22.

23.

266

societies. See Stephens (1979, pp. 47–48, 77–81) for an elaboration. Not only would this theoretical view have to be altered to fit the conditions of the authoritarian capitalist regimes of pre-World War II Europe, since it is quite clear that the extension of suffrage increases the influence of the working class, but subsequent work (e.g., Evans et al. 1985; Orloff and Skocpol 1984) also convinces me that the state has to be given a stronger role than I attributed to it in that work. See Evans and Stephens (1988) for an updated viewpoint. Seton-Watson (1967, p. 282) argues that it was with the Libyan war that nationalism became antidemocratic and imperialist and began to attract business support. Skocpol’s point on the difficulty that class analysis has in dealing with the German case dates back at least to Weber’s critique of Marx and is commonplace in recent critiques of historical materialism, which argue, among other things, that the state in both its bureaucratic and military manifestations has played a much larger role in historical development than most Marxist analyses attribute to it (e.g., see Mann 1986, in press; Giddens 1981). This dominant view is expressed most succinctly in the works of Hans-Ulrich Wehler (see esp. Wehler 1985). Earlier works singled out for criticism, all emphasizing the role of the Junkers, include Veblen ([1915] 1966), Schumpeter (1968), Gerschenkron (1943), and Moore. For critical but sympathetic discussions of Blackbourn and Eley, see Evans (1985) and Garst (1987). Also see Sheehan (1978, pp. 264–70) on the position of various liberal parties and factions on the suffrage question in the last decade of the Wilhelmine period. Not even the Left liberals favored the full reform of local government, and, while they officially supported the introduction of responsible cabinet government, they “continually backed away from supporting an unambiguous call for parliamentary sovereignty” (Sheehan 1978, p. 270). The focus on process and the shunting aside of broad structural determinants are characteristic of single-country studies, not only of Germany but also of the other breakdown cases. For example, see any of the essays on Europe in Linz and Stepan (1978). The historical case studies uncover direct instrumentation, but the method does not press one to focus on what structural factors differentiate the breakdown cases from the democratic survivors, as does a comparative study such as this one. It also matters what cases are being compared for what purposes. For instance, the one comparative essay in the Linz and Stepan volumes, the introductory essay by Linz, compares the breakdown cases in order to uncover commonalities among them. In part as a result of the method, Linz’s essay focuses on the process of breakdown, not the distinctive features that differentiate the breakdown and survivor cases. The figures given in the text ignore the votes for the Independent German Social Democratic Party (USPD), whose leadership was clearly democratic. But the party split into two factions, one of which reentered the Social Democratic Party (SPD). The other faction joined the Communist party (KPD), whose disastrous line clearly contributed to the rise of Hitler. See Abraham (1986) and Turner (1985). The Turner critiques of Abraham focus on the documentation of the instrumentalist argument about big business support for the Nazis. This is, in fact, not the main argument of the book, which is a class-coalitions approach to the conditions for stable democracy not unlike the one proposed here. Turner’s study sticks very close to the traditional historical case study method, which, as I pointed out in n. 20, is biased toward instrumentalist arguments (e.g., this group did x with y effect). For the structural argument to hold, it is not necessary to inject the subjective attitudes of businessmen as an intervening variable, but to make the argument without this link requires a more complex comparative analysis or counterfactual, which is fortunately not necessary


24. 25.

26. 27. 28.

29.

30. 31. 32.

33. 34.

267

in this case. It is worth pointing out here that a Keynesian reflationary policy, which would have gone against the subjective preferences of business, might well have improved the economy, as it did under the Nazis (Gourevitch 1986), and thus worked to stabilize the democratic regime. Northeim, the town studied in Allen’s book, was overwhelmingly Protestant; thus there was no Catholic subculture to speak of. Note that the ideological content of these three campaigns (the Kulturkampf, the Agrarian League, and the Naval League), while all authoritarian, varied, and they were pressed by different, and sometimes conflicting, segments of the economic and political elite: roughly the state and the liberals, the Junkers and the conservatives, and the state and heavy industry, respectively. The state prosecutor’s plea began with a statement that sounded more like a campaign speech for a national hero than an indictment of a criminal (see Bracher 1970, pp. 119–20). As stated in the previous paragraph, the landed nobilities of the subject nationalities often led the nationalist pressures for autonomy. Thus, the sociological counterparts to the Christian Socials in Germany were the DNVP, BVP, and the Zentrum. In the remainder of the party systems, the counterparts are closer: the PSU bloc contains the parties representing the secular urban upper classes; and the labor bloc, the working-class parties. The support for the last two blocs are very roughly the same in the two countries. The difference lies in the remaining groups, which are split in Germany, with part supporting nationalistic monarchical authoritarianism and part democratic Catholicism, whereas in Austria they are united supporting the authoritarian Catholic corporativism of the Christian Socials. In comparing Scandinavian liberalism with English (or Austrian) liberalism, it should be pointed out that it had a different social base. Lipset and Rokkan classify the English party as a P-D (Dissenting Sects)-U party and the Scandinavian parties as PD. This misses the urban middle-class component of Scandinavian liberalism, but it accurately catches the key difference: the Scandinavian bourgeoisie supported the parties of the Right and not the liberal parties. This is probably the main source of the more active support for democracy of the Scandinavian liberal parties as compared with the English. The following remarks on 19th-century Spain are based on Carr (1982, see esp. pp. 158 ff., 208 ff., 284 ff., 287, and 342 on the social bases of the parties). For the events of the Republic, see Preston (1983) and Malefakis (1970). The struggle over the role of the Catholic church in Spanish society was the other burning issue of the Republic. But it is important to see that it is linked, in part, to the land question. In a Catholic smallholding society like Belgium, the church and Catholic political forces are more moderate. In Spain, where large landholders and the church are allied, it is not difficult to see why the discontent of the masses was often strongly anticlerical. See the quote from Gross on Austria, above, for an example of both lines of reasoning. The assessment of state capacity is problematic. If the state effectively intervenes in the economy, then one can conclude that state capacity was adequate, but if it does not intervene, then it may be because of either lack of state capacity or lack of (perceived) interest of ruling groups or bureaucrats in doing so. In the Austrian case, it would appear that the sheer bureaucratic capacity was there, but that the dominant social groups in the Habsburg regime did not perceive it to be in their interests to pursue such an ambitious development project.


268

References Abraham, David. 1986. The Collapse of the Weimar Republic: Political Economy and Crisis. New York: Holmes & Meier. Alapuro, Risto. 1980. “Mass Support for Fascism in Finland.” Pp. 678–88 in Who Were the Fascists: Social Roots of European Fascism, edited by Stein Ugelvik Larsen, Bernt Hagtvet, and Jan Fetter Myklebust. Bergen: Universitetsforlaget. Alapuro, Risto, and Erik Allardt. 1978. “The Lapua Movement: The Threat of Rightist Takeover in Finland.” Pp. 122–41 in The Breakdown of Democratic Regimes: Europe, edited by Juan Linz and Alfred Stepan. Baltimore: Johns Hopkins University Press. Allen, William S. 1984. The Nazi Seizure of Power: The Experience of a Single German Town 1922–1945, rev. ed. New York: Watts. Aminzade, Ronald. In press The Origins of Democratic Institutions: Political Parties and Class Formation in France. Aminzade, Ronald. 1981. Class, Politics, and Early Industrial Capitalism: A Study of Mid-Nineteenth-Century Toulouse, France. Albany: State University of New York. Anderson, R.D. 1977. France 1870–1914: Politics and Society. London: Routledge & Kegan Paul. Blackbourn, David. 1984. “The Discreet Charm of the Bourgeoisie: Reappraising German History in the Nineteenth Century.” Pp. 159–292 in The Peculiarities of German History: Bourgeois Society and Politics in Nineteenth-Century Germany, by David Blackbourn and Geoff Eley. Oxford: Oxford University Press. Blackbourn, David, and Geoff Eley. 1984. Introduction. Pp. 1–38 in The Peculiarities of German History: Bourgeois Society and Polities in Nineteenth-Century Germany, by David Blackbourn and Geoff Eley. Oxford: Oxford University Press. Blewett, Neal. 1965. “The Franchise in the United Kingdom 1885–1918.” Past and Present, no. 32, pp. 27–56. Block, Fred. 1977. “The Ruling Class Does Not Rule: Notes on the Marxist Theory of the State.” Socialist Revolution 7:6–28. Bollen, Kenneth A. 1979. “Political Democracy and the Timing of Developments American Sociological Review 44 (4): 572–88. Bollen, Kenneth A. 1983. “World System Position, Dependency, and Democracy: The Crossnational Evidence.” American Sociological Review 48 (4): 468–79. Bollen, Kenneth A., and Robert Jachnan. 1985a. “Economic and Noneconomic Determinants of Political Democracy in the 1960s.” In Research in Political Sociology, edited by Richard G.Braungart. Greenwich, Conn.: JAI. Bollen, Kenneth A., and Robert Jachnan. 1985b. “Political Democracy and the Size Distribution of Income.” American Sociological Review 50 (4): 438–57. Boyer, John. 1981. Political Radicalism in Late Imperial Vienna: Origins of the Christian Social Movement, 1848–1897. Chicago: University of Chicago Press. Bracher, Karl Dietrich. 1970. The German Dictatorship: The Origins, Structure, and Effects of National Socialism. New York: Praeger.


269

Brenner, Robert. 1977. “The Origins of Capitalist Development: A Critique of NeoSmithian Marxism.” New Left Review 104 (July/August): 25–92. Calleo, David. 1978. The German Problem Reconsidered: Germany and the World Order, 1870 to the Present. Cambridge: Cambridge University Press. Cardoza, Anthony L. 1982. Agrarian Elites and Italian Fascism: The Province of Bologna, 1901–1926. Princeton: Princeton University Press. Carr, Raymond. 1982. Spain, 1808–1975, 2d ed. Oxford: Oxford University Press Carsten, F.L. 1973. The Reichswehr and Politics 1918–1933. Berkeley: University of California Press. Castles, Francis G. 1973. “Barrington Moore’s Thesis and Swedish Political Development.” Government and Opposition 8 (3): 313–31. Childers, Thomas. 1983. The Nazi Voter. Chapel Hill: University of North Carolina Press. Daalder, Hans. 1966. “The Netherlands: Opposition in a Segmented Society.” Pp. 188– 236 in Political Oppositions in Western Democracies, edited by Robert Dahl. New Haven: Yale University Press. Dahrendorf, Ralf. 1967. Society and Democracy in Germany. Garden City, N.Y.: Doubleday. Derry, T.K. 1973. A History of Modern Norway, 1814–1972. Oxford: Clarendon. Dovring, Folke. 1965. Land and Labour in Europe in the Twentieth Century, 3d rev. ed. The Hague: Nijhoff. Dybdahl, Vagn. 1969. Partier og Erhverv. Aarhus: Universitetsforlaget. Eley, Geoff. 1983. “What Produces Fascism: Preindustrial Traditions of a Crisis of a Capitalist State.” Politics and Society 12 (1): 53–82. Eley, Geoff. 1984. “The British Model and the German Road: Rethinking the Course of German History before 1914.” Pp. 39–158 in The Peculiarities of German History: Bourgeois Society and Polities in Nineteenth-Century Germany, by David Blackbourn and Geoff Eley. Oxford: Oxford University Press. Elwitt, Sanford. 1975. The Making of the Third Republic: Class and Politics in France, 1808–1884. Baton Rouge: Louisiana State University Press. Evans, Peter, Dietrich Rueschemeyer, and Theda Skocpol, eds. 1985. Bringing the State Back In. Cambridge: Cambridge University Press. Evans, Peter, and John D.Stephens. 1988. “Development and the World Economy.” Pp. 139–73 in Handbook of Sociology, edited by Neil Smelser. Newbury Park, Calif.: Sage. Evans, Richard J. 1985. “The Myth of Germany’s Missing Revolution.” New Left Review 149:67–94. Fitzmaurice, John. 1983. The Politics of Belgium Crisis and Compromise in a Plural Society. New York: St.Martin’s. Garst, Daniel. 1987. “Capitalism and Liberal Democratic Political Development: State


270

Structures in Britain and Germany prior to World War I.” Ph.D.dissertation. University of Minnesota Gerschenkron, Alexander. 1943. Bread and Democracy in Germany. Berkeley: University of California Press. Gerschenkron, Alexander. 1962. Economic Backwardness in Historical Perspective. Cambridge: Harvard University Press. Giddens, Anthony. 1981. A Contemporary Critique of Historical Materialism. Berkeley and Los Angeles: University of California Press. Gourevitch, Peter. 1986. Politics in Hard Times: Comparative Responses to International Economic Crises. Ithaca: Cornell University Press. Gross, N.T. 1973. “The Habsburg Monarchy 1750–1914.” Pp. 228–78 in The Emergence of Industrial Societies, pt. 1. Edited by Carlo M.Cipolla. The Fontana Economic History of Europe. Glasgow: Fontana. Gulick, Charles A. 1948. Austria from Habsburg to Hitler. Berkeley: University of California Press . Hadenius, Stig, Hans Wieslander, and Björn Molin. 1970. Sverige efter 1900. Stockholm: Aldus. Hamilton, Richard F. 1980. Who Voted for Hitler? Princeton: Princeton University Press. Johnson, Richard. 1976. “Barrington Moore, Perry Anderson and English Social Development.” Working Papers in Cultural Studies 9:7–28. Katzenstein, Peter. 1985. Small States in World Markets: Industrial Policy in Europe. Ithaca: Cornell University Press. Kehr, Eckart. (1930) 1973. Battleship Building and Party Politics in Germany, 1894– 1901 Chicago: University of Chicago Press. Korpi, Walter. 1982. The Democratic Class Struggle. London: Routledge & Kegan Paul. Kurth, James. 1979. “Industrial Change and Political Change.” Pp. 319–62 in The New Authoritarianism in Latin America, edited by David Collier. Princeton: Princeton University Press. Lepsius, M.Rainer. 1978. “From Fragmented Party Democracy to Government by Emergency Decree and National Socialist Takeover: Germany.” Pp. 34–79 in The Breakdown of Democratic Regimes: Europe, edited by Juan Linz and Alfred Stepan. Baltimore: Johns Hopkins University Press. Linz, Juan J. 1967. “The Party System of Spain: Past and Future.” Pp. 197–282 in Party Systems and Voter Alignments, edited by Seymour Martin Lipset and Stein Rokkan. New York: Free Press. Linz, Juan J. 1975. “Totalitarian and Authoritarian Regimes.” In Handbook of Political Science, vol. 3. Edited by Fred Greenstein and Nelson Polsby. Reading, Mass.: Addison-Wesley. Linz, Juan J. 1978. Crisis, Breakdown and Reequilibration. Baltimore: Johns Hopkins University Press. Linz, Juan, and Alfred Stepan. 1978. The Breakdown of Democratic Regimes. Baltimore:


271

Johns Hopkins University Press. Lipset, Seymour Martin. 1960. Political Man. Garden City, N.Y.: Anchor. Lipset, Seymour Martin, and Stein Rokkan. 1967. “Cleavage Structures, Party Systems, and Voter Alignments: An Introduction.” Pp. 1–64 in Party Systems and Voter Alignments, edited by Seymour Martin Lipset and Stein Rokkan. New York: Free Press. Lorwin, Val. 1966. “Belgium: Religion, Class, and Language in National Politics.” Pp. 147–87 in Political Oppositions in Western Democracies, edited by Robert Dahl. New Haven: Yale University Press. Luebbert, Gregory M 1987. “Social Foundations of Political Order in Interwar Europe.” World Politics 39:449–78. Lyttelton, Adrian. 1973. The Seizure of Power: Fascism in Italy 1919–1929. New York: Scribner’s. Maier, Charles. 1975. Recasting Bourgeois Europe. Princeton: Princeton University Press. Malefakis, Edward. 1970. Agrarian Reform and Peasant Revolution in Spain. New Haven: Yale University Press. Mann, Michael. 1986. The Sources of Social Power. Vol. 1: A History of Power in Agrarian Societies. Cambridge: Cambridge University Press. Mann, Michael. In press. The Sources of Social Power. Vol. 2: A History of Power in Industrial Societies. Cambridge: Cambridge University Press. Matthew, H.C.G., R.I.McKibbon, and J.A.Key. 1976. “The Franchise Factor in the Rise of the Labour Party.” English Historical Review 91 (361):723–52. Miller, Kenneth E. 1968. Government and Politics in Denmark. Boston: Houghton Mifflin. Mitchell, B.R. 1978. European Historical Statistics 1750–1970. New York: Columbia University Press. Moore, Barrington. 1966. The Social Origins of Dictatorship and Democracy. Boston: Beacon. Neebe, Reinhard. 1981. Grossindustrie, Staat. und NSDAP 1930-1933. Göttingen: Vandenhoek & Ruprecht. O’Donnell, Guillermo A. 1973. Modernization and Bureaucratic Authoritarianism. Berkeley, Calif.: Institute of International Studies. O’Donnell, Guillermo A. 1978. “Reflections on the Bureaucratic Authoritarian State.” Latin American Research Review 13 (1):3–38. Orloff, Ann S., and Theda Skocpol. 1984. “Why Not Equal Protection? Explaining the Politics of Public Social Spending in Britain, 1900–1911, and the United States, 1880s1920.” American Sociological Review 49 (6):726–50. Preston, Paul. 1983. The Coming of the Spanish Civil War: Reform, Reaction, and Revolution in the Second Republic. New York: Methuen.


272

Rabinbach, Anson. 1983. The Crisis of Austrian Socialism: From Red Vienna to Civil War, 1927–1934. Chicago: University of Chicago Press. Rokkan, Stein. 1966. “Norway: Numerical Democracy and Corporate Pluralism.” Pp. 70– 115 in Political Oppositions in Western Democracies, edited by Robert Dahl. New Haven: Yale University Press. Rokkan, Stein. 1970. Citizens, Elections, Parties. Oslo: Universitets Förlaget. Rokkan, Stein, and Jean Meyriat. 1969. International Guide to Electoral Statistics. The Hague: Mouton. Rueschemeyer, Dietrich. 1980. “Über Sozialökonomische Entwicklung und Demokratie.” Pp. 211–30 in Weltgesellschaft und Sozialstruktur. Diessenhofen: Ruegger. Rustow, Dankwart. 1955. The Politics of Compromise. Princeton: Princeton University Press. Schumpeter, Joseph. 1968. Imperialism and Social Classes. Cleveland: World. Senghaas, Dieter. 1985. The European Experience: A Historical Critique of Development Theory. Dover, N.H.: Berg. Seton-Watson, Christopher. 1967. Italy from Liberalism to Fascism. London: Methuen. Sheehan, James J. 1978. German Liberalism in the Nineteenth Century. Chicago: University of Chicago Press. Simon, Walter. 1978 “Democracy in the Shadow of Imposed Sovereignty: The First Republic of Austria.” Pp. 80–121 in The Breakdown of Democratic Regimes: Europe, edited by Juan Linz and Alfred Stepan. Baltimore: Johns Hopkins University Press. Skocpol, Theda. 1973. “A Critical View of Barrington Moore’s Social Origins of Dictatorship and Democracy.” Politics and Society 12 (2): 1–34. Skocpol, Theda. 1979. States and Social Revolutions. Cambridge: Cambridge University Press. Söderpalm, Sven Anders. 1969. Storföretagarna and Det Demokratiska Genomorottet. Lund: Gleerups. Stephens, John D. 1979. The Transition from Capitalism to Socialism. Urbana: University of Illinois Press. Stinchcombe, Arthur. 1968. Constructing Social Theories. New York: Harcourt, Brace & World. Tasca, Angelo. 1938. The Rise of Italian Fascism 1918–1922. New York: Fertig. Taylor, A.J.P. 1976. The Habsburg Monarchy, 1809–1918: A History of the Austrian Empire and Austria-Hungary. Chicago: University of Chicago Press. Therborn, Göran. 1977. “The Rule of Capital and the Rise of Democracy.” New Left Review 103:3–41. Tilly, Charles. 1975. “Reflections on the History of European State-Making.” Pp. 3–83 in The Formation of National States in Western Europe, edited by Charles Tilly. Princeton: Princeton University Press. Tilton, Timothy. 1974. “The Social Origins of Liberal Democracy: The Swedish Case.” American Political Science Review 68 (2): 561–71. Tumin, Jonathan. 1978. “Pathways to Democracy: A Critical Revision of Barrington Moore’s Theory of Democratic Emergence and an Application of the Revised Theory


273

to the Case of Netherlands.” Ph.D.dissertation. Harvard University. Turner, Henry A. 1985. German Big Business and the Rise of Hitler. New York: Oxford University Press. Valenzuela, Julio Samuel. 1979. Labor Movement Formation and Politics: The Chilean and French Cases in Comparative Perspective, 1850–1950. Ph.D.dissertation. Columbia University. Veblen, Thorstein. (1915) 1966. Imperial Germany and the Industrial Revolution. Ann Arbor: University of Michigan Press. Verney, Douglas. 1957. Parliamentary Reform in Sweden 1800–1921. Oxford: Clarendon. Wehler, Hans Ulrich. 1985. The German Empire, 1871–1918. Dover, N.H.: Berg. Wiener, Jonathan M. 1976. “Review of Reviews: The Social Origins of Dictatorship and Democracy.” History and Theory 15 (2): 146–75. Winter, J.M. 1986. The Great War and the British People. Cambridge: Harvard University Press. Zolberg, Aristide R. 1986. “How Many Exceptionalisms?” Pp. 397–455 in Working Class Formation: Nineteenth-Century Patterns in Western Europe and the United States, edited by Ira Katznelson and Aristide R.Zolberg. Princeton: Princeton University Press.


274

50 The Politics of Backwardness in Continental Europe, 1780–1945* Andrew C.Janos

* Source: World Politics, 1989, vol. 41, pp. 325–358.

Introduction The following study is one of history in the frame of reference of social science. As such, it has been written with two broad objectives. The first is to identify historical patterns of class formation and state-building in Europe, and to suggest new explanations for the political differentiation of the Continent from the late eighteenth to the middle of the twentieth century. The second is to examine the relevance of a number of hypotheses concerning the politics of international economic inequality, and thus to contribute to the understanding of a central problem in contemporary political economy. The discipline of political economy inquires, above all, into causal relationships between economic scarcity and political authority. The central hypothesis around which the discipline was originally constructed may be stated quite succinctly. If material goods are becoming more scarce, then the stability of authority will decline, or, alternatively, authority will be exercised by ever smaller and more tightly organized minorities over ever larger realms of social concerns. Conversely, as resources become more abundant, it is easier to resolve social conflicts by peaceful means, civil societies are more likely to thrive, and participation in the exercise of authority is more likely to increase. Simple as they are, these propositions are the foundations of the more complex theories of such classical political economists as Karl Marx and Adam Smith. While the fundamental validity and relevance of these generalizations concerning the consequences of scarcity have rarely been challenged, the probable causes of economic scarcity have recently been the subject of heated debates. A number of major revisionist schools have emerged from these debates, each attempting to expand the explanatory paradigm of political economy by shifting the focus of the discipline from national economies, societies, and cultural systems toward larger, supranational entities, above all, toward the study of relationships among the “core,” the “periphery” and “semiperiphery” (together, the “peripheries”) of a “modern world system.”1 This, to be sure, does not imply that the new political economy is ready to ignore the relevance of social and cultural configurations within national boundaries. It merely implies that such configurations are apt to acquire new meaning and weight within the larger context of the global economic, political, and cultural systems.

The Politics of Backwardness in Continental Europe, 1780–1945

275

While the utility and relevance of this new mode of thinking is widely shared by today’s political economists, the dynamics of relative scarcity, the causes of impoverishment of peripheral countries, and the relationship between development and decay have given rise to several contending schools of thought. Perhaps the most popular of these locates the problem of scarcity and abundance in the international economic system, specifically in processes of economic exchange. Within this school we must deal with two competing theories. One, associated with Marxism, rests on the tenets of the labor theory of value, and distinguishes between “lower” and “higher” ranking goods, according to the amount of “direct labor” required to produce them.2 This theory assumes that trade between advanced and backward societies involves mainly the exchange of goods of different rank, and that when such an exchange takes place, surplus is being transferred from the less to the more developed economy. Since surplus is seen as a source of both consumption and of new investment, the theory purports to explain not only the reproduction of existing patterns of inequality, but also the steady increase of disparities between the few societies of the core and the many countries of the periphery. This school’s second theory is associated with conventional economists like Gunnar Myrdal and Albert Hirschman.3 Much like the Marxists, these economists seek explanations for the problems of less developed nations in the international economy, and see the world market as a mechanism that operates to the disadvantage of poorer countries. The result is a steady “polarization”4 of the world economy and a downward “drift”5 of the peripheries. However, unlike the Marxists, such economists explain these processes not in reference to the labor theory of value and exploitation, but in terms of more standard categories of economic analysis. Among these they list unequal rates of return and remuneration, which attract both capital and talent from the peripheries to the core countries; income elasticities of demand that differ for the products of the core and the periphery; and the competitive disadvantage of the new industries of the peripheries versus the established industries of the core. A second school of the new political economy likewise attempts to explain the relative impoverishment (and political authoritarianism) of the peripheries, but seeks to identify their causes in the context of nation-states that make up the international political system. Contrary to the thinking of a long line of classical economists from Johann-Gottlieb Fichte to Friedrich List and Alexander Gerschenkron,6 this new school views the state not as an instrument of accumulation and development, but as a potential agent of resource extraction for nonproductive purposes. Within this school, too, we can identify two different theories. The first relates this nonproductive spending to the pressures of international competition for expensive military technologies. The costs of these technologies, so the argument runs, vary little from one economic sector to another, but the level of development does. Thus, everything else being equal, military expenditures are likely to take a larger share of the national income of the poorer competitors at the expense of their ability to raise their standard of living.7 A second theory meanwhile calls attention to the potential profligacy of elites, and pressures on them in backward societies to divert public funds for private benefit. In either case, we are dealing with a grand paradox: while poverty and stagnation call for more etatism, this etatism in turn may result in still more poverty and relative backwardness.8 Finally, a third school of revisionist thought attributes the economic problems of


276

backward societies to changing human needs, relative deprivation, and the international demonstration effect (IDE) of the ongoing economic progress of the core societies. This theory stipulates that the economic success of the advanced countries creates new lifestyles, and that these are diffused rapidly from one economic sector to another, creating consumer expectations that the economies of peripheral societies cannot fulfill. Indeed, under the pressure of these expectations the weaker economies may crumble, and instead of developing they may stagnate or decay, thereby further increasing material disparities between themselves and the pioneers of innovation in the core region of the global economy. Although terms like international demonstration effect, relative deprivation, or, more pejoratively, cultural imperialism, are of recent vintage, the ideas underlying them go back to Thorstein Veblen, who formulated his theories of consumption and need in the first decades of this century. According to Veblen, technological innovation results in new amenities. The first beneficiaries of these are social elites, who acquire them by virtue of their power and privileged position. They then spread with “great facility” through the lower echelons of society, by the “force of emulative imitation.”9 This element of status appears to be equally relevant In the international context. In more recent times, though, the dynamics of emulation have been enhanced by a number of psychological factors. For one thing, unlike the innovations of earlier centuries, the fruits of the modern technological and scientific revolutions have affected not just the quality, but the very quantity, of precious human life. For another, the products of these revolutions have been mediated through the particular value system of the Occident, which has created a deep sense that in some respects everyone is “in the same boat,”10 that is to say, they all have a common entitlement to life and to its necessities.11 It is in this broader system of understandings that Veblen’s theory acquires full meaning—that consumer expectations will become “mandatory…staples of decency” and that falling short of such staples will become an experience of a “spiritual” character, apt to generate a deep sense of frustration.12 On the following pages we will trace these sources of relative scarcity in the context of the international economic, political, and cultural systems of the nineteenth and early twentieth centuries. To the extent that elements of these systems may be taken as constants, we will have a model, or general theory, of the politics of backwardness in the modern world. To the extent, however, that these categories are variable, the potential for analogy will be limited, and we will require further historical examples before we can formulate a genuine theory of international income disparities applicable in a wide range of chronological and cultural settings.

The Development of Underdevelopment By general agreement among historians and social scientists, the origins of the modern world economy go back to the great economic revolutions of Northwestern Europe between the sixteenth and nineteenth centuries. The social consequences of these revolutions may be summarized under three general rubrics: (1) the capacity to produce surplus and the subsequent expansion of markets and risk-taking entrepreneurship; (2) the


277

increasing complexity of the social division of labor, and the attendant pressures for more rationalized and elaborate systems of administrative regulation; and (3) new forms of social consciousness, especially a sense of personal efficacy, among ever larger segments of societies. For the body politic, these changes implied the opening of new cleavages and the rise of conflicts, not only between bourgeoisie and aristocracy, but also among the new modern elements—the bourgeoisie, the bureaucracy, and the mass of common people “entering” the political process. In Britain, Holland, and the other regions that qualify as the innovative core of the modern world economy, these conflicts were resolved in two stages. In the first, the bourgeoisie emerged victorious over royal bureaucracies and traditional aristocracies. In the second, the political rights of the bourgeoisie were gradually extended to the working classes. Needless to say, these developments did not follow mechanically from the fact of technological innovation. The institutional framework for political participation was provided by the well-established norms of the Western legal tradition, and its success was made possible by the steady expansion of a resource base that permitted the incorporation of the lower classes into the political community without raising the specter of expropriation for the elites. In the first instance, these economic revolutions referred to certain changes in agricultural production. Admittedly, the term “revolution” is somewhat hyperbolic here. For one thing, it took Flemish and British farmers more than a century and a half to raise cereal yield ratios of seed to harvest from 1:4 to 1:6, and yet another century, the eighteenth, to raise their yields to a critical 1:10 ratio.13 For another thing, the improved methods of plowing, fertilizing, and crop rotation spread at a painfully slow rate of three to four miles a year from the Northwest toward the South and the East of the Continent. At this rate, the technologies invented in Holland and the Rhineland during the “long sixteenth century” (1450–1640) reached Bavaria, Saxony, Bohemia, and Central France around 1750, Hungary, Poland, and the Balkans between 1825 and 1850, and Russia still later. And since innovation in the core was not a one-time affair but an ongoing process, regions located some six to eight hundred miles from the epicenter of economic change remained some two hundred years behind the core in terms of yields per acre. Overall, the distribution of agricultural productivity and income across the Continent acquired a neatly regressive geographical pattern that transcended both climatic and cultural boundaries, and it is this pattern that enables us, by 1800, to distinguish among the three economic regions of Europe: a center of innovation in the Northwest (the Low Countries, England, northern France, the Rhineland, and a few other provinces in western Germany), characterized by high yield ratios and a commercialized agriculture; the adjacent regions, bounded in the East and South by the rivers Elbe, Po, and Loire, where agricultural innovations had made some headway, but where farming maintained a distinctive mode of production with limited reliance on wage labor; and finally the regions east and south of these lines, where the innovations of the Northwest had yet to make a dent on the traditional mode of production. Because this geographically regressive pattern of productivity and income disparity would persist up until 194514 (and even later), and because the economic pattern would closely correlate with patterns of authority, we are justified in designating these three sectors as the core, the semiperiphery, and the periphery of the modern European economy. The second revolution, with its roots in the industrial breakthroughs of eighteenth-


278

century England, is more appropriately designated as such, for the changes in the technology of production created dramatic increases in the volume of manufactures. The amount of raw cotton processed by British factories grew from a mere 2.5 million pounds in 1760 to 22 million pounds in 1787, and then to 366 million pounds in 1837, a sixteenfold increase within a period of fifty years. The total amount of iron processed into steel by English factories was 68,000 tons in 1788. This figure rose to 250,000 tons by 1806, and 678,000 by 1830.15 In the wake of these changes, domestic consumption grew rapidly, giving rise to a new style of life. The availability of cheap cotton “made it possible for millions to wear drawers and chemises where before there had been nothing but coarse and dirty undergarments,” while the more prosperous, “impressed by the color and elegance of cotton prints learned to distinguish more and more between seasons and [to] dress for the summer in muslins and calicoes.”16 The consumption boom, of course, extended beyond textiles to “tobacco, soap, candles, spirits, beer, kitchen utensils, glassware, soap, matches, household furniture.”17 There were further changes in housing, nutrition, and hygiene, and in the process notions of luxury and staple were rapidly reevaluated. “Luxuries came to be seen as mere ‘decencies,’ and ‘decencies’ came to be seen as necessities.”18 To be sure, the distribution of these consumer gains remained highly uneven, their principal beneficiaries having been the new “middling” element that had risen from manual labor to professional status or entrepreneurship. Still, “more men and women than ever before in human history enjoyed the experience of acquiring material possessions. Objects which for centuries had been the privileged possessions of the rich came within the space of a few generations to be within the reach of a larger part of society than ever before.”19 The dynamics of these revolutionary changes provide a convenient starting point for explaining the intricate relationship between development and underdevelopment on the Continent. Significantly, new lifestyles and patterns of consumption spread rapidly from industrial England toward the agricultural peripheries. This is amply evident from both trade statistics and from the observations of contemporaries. Within decades, British exports increased from an estimated 5 to 35 percent of British national income,20 a figure that implied commerce so active and universal that in traveling from Prussia to St.Petersburg, from Amsterdam to the farthest point in Sweden, from Dunkirk to the southern extremes of France [the traveler] was served at every turn from English earthenware. The same fine articles adorn[ed] the tables of Spain, Portugal and Italy, and provide[d] the cargoes of ships to the East Indies, the West Indies and America.21 Consequently, in Eastern Europe, as one observer of the Romanian principalities noted in 1828, “perfumes, champagne, glass, silverware, mirrors, matches, furniture spread rapidly among the members of the upper and middle classes.”22 Or, to quote a Hungarian writer of the same year: “With the abundance of cash people are developing new tastes every day. Silverware, so far only seen in the houses of the rich, now appears in many households. The watch, once as rare as a white crow, has now turned into an article of necessity.”23 And again, in the Balkans


279

better-off households now bought some furniture and household utensils…. Tea, coffee, sugar were passing out of the class of luxury goods into more common use. Town-made lamps were replacing homemolded candles. [Even] in peasant homes a few books were to be found…. Although the amount of purchased articles found in [these] homes appeared small by urban standards, nevertheless they represented considerable change.24 Nor was the imitation of Western lifestyles uniquely European. Half a world away, the same patterns of consumption came to Latin America, where in the words of one observer, there was likewise “a breathless desire to be like the French and the British,…to dress like fashionable Europeans,” to pay homage to the “intrinsic worth of the way in which Europeans went about building their houses, educating their young or spending their leisure time.”25 In Valparaiso, the shops sold “German glass and toys, American furniture, English hardware, textiles, carpets, household furniture, chinaware.”26 Or in the words of another historian: “The houses of the better classes were as replete with objects of American and European manufacture as would be true for the citizen of London or Boston.”27 Whether defined as luxuries or necessities, these patterns of consumption were incongruous with existing modes of production, and the further a region was located from the core, the greater this discrepancy would be. While in the innovative core of the world economy consumer tastes had developed as part of an organic process, and were related to an “agriculturally induced increase in home demand,”28 on the peripheries, especially after the beginnings of the industrial revolution, expectations rose ahead of real income, and thus were apt to have adverse economic consequences. First, the “breathless desire” to acquire Western products boosted effective demand for manufactures, and may well have been one of the factors affecting the terms of trade and the relative movement of prices, to the disadvantage of primary producers in the half century before World War I.29 In addition, this taste-induced consumer revolution was bound to lower marginal rates of saving and private investment. This tendency toward disinvestment then became the starting point for the “downward drift” of the peripheries, and the first hint that effective accumulation would have to be accomplished outside the private sector. Again, the more backward the economy, the greater would be these adverse effects. This is not to suggest that all peripheral societies, and within them all individuals, turned overnight into profligate spenders, but those who did apply consumer restraint did so at a considerable social and psychological cost to themselves, and even with stationary or slowly rising incomes felt that they were becoming poorer with every passing year. True, most peripheral economies responded at first to the progressive redefinition of wants and needs by increasing the exports of their own primary products. To the extent that individual countries found a niche in the world economy, they experienced export booms that in turn created euphoric expectations about the opportunities inherent in trading. There was thus a “golden age” of Spanish wool, of Portuguese wine, of Hungarian and Romanian wheat, and of Serbian plums and pigs (just as there were similar golden ages for Brazilian coffee, Cuban sugar, Chilean copper, and Argentinian beef) following in the wake of the early industrial expansion of Britain.30 But closer


280

observation will reveal that these export booms were based on utilization of surplus land or the exploitation of existing mineral sources, rather than on intensification of production and capital investment. Consequently, the golden ages tended to be brief, and the “false euphoria”31 of the boom years gave way quickly to a sense of malaise, selfpity, and gloom that permeates both public and literary life in these countries after the middle of the nineteenth century. While the predicament of the peripheral societies of Europe varied with their geographical distance from the core region, in each one the progress of the Northwest had a discernible impact on the processes of class formation and social mobility. At the top of the social pyramid, the large producers in countries like Spain, Portugal, Russia, Hungary, Romania, and Poland (much like the estancieros of Latin America) could hold on to their property and keep pace with changing lifestyles, but they could not consume and save at the same time. Increased production and income, therefore, could only be attained by tightening the working conditions of the labor force, and by depressing real wages. Thus, while at first, during the great euphoria of the “golden” years, many latifundiary aristocrats embraced, or at least did not oppose, such liberal reforms as the abolition of serfdom, by the second half of the nineteenth century, most of them had reemerged in politics as the hard core of a conservative pressure group lobbying persistently for greater “social discipline,” protection from the vagaries of the market, and the restoration, in one guise or another, of the status quo ante with respect to labor relations. Intriguingly, therefore, the mode of production and the labor economy of the continental large estate froze into the pattern that had existed in the first half of the nineteenth century and retained this mode until the middle of the twentieth century. In the semiperipheral regions this mode entailed sharecropping, seasonal labor, and various modifications of wage labor, in which peasants rented a piece of land in exchange for working a number of days per year on the owner’s land.32 In the South and East then, we find an increasing tendency toward payments in kind and a “mixture of freedom and serfdom that was the legacy of the past.”33 The Romanian Constantin DobrogeanuGherea appropriately termed this condition “neserfdom.”34 The “solution,“ in any event, ”was to exploit the peasant more intensely.” In the end, “good or bad landowners were an irrelevance, for losses made them all exploiters,”35 being themselves imprisoned within a structure that pushed them toward repressive social and economic policies. Still more desperate, however, was the condition of the middle strata, above all of the agrarian middle class. In the absence of large tracts of land, the lesser landowners could maintain a decent style of life only at the expense of mortgaging, and eventually losing, their estates. Again, not all of them did so, but those who did not turned themselves into déclassés in comparison with the middle classes of the core countries. Max Weber describes this progression in one of his early works.36 Such a fate befell the Polish szlachta, the Hungarian dzsentri, the lesser boyars of Romania, the middle dvoryanstvo of Russia, the petty nobles of Portugal and Spain; their demise has been the staple of literary works from the Iberian peninsula to the Russian plain. Quite often the troubles of these classes have been attributed to their heroic-military traditions, which supposedly prevented them from adapting to the rationalism of the market mechanism. This, however, is contradicted by the fact that the majority of the non-noble middle classes— the artisans and merchants of the pre-industrial age—perished together with them, and


281

their sons became, side by side with the landed gentry, part of the same free-floating intelligentsia that a Romanian contemporary contemptuously described as a “proletariat of the pen.”37 These terms refer to the educated segment of society, but what provided this group of people with a common identity was not so much their education as their common desire to find status and security as a political class in the service of the state. Much as in the countries of the non-European world, “for these groups without space or status…a government job was not only an alternative, it was a virtual imperative.”38 In many countries, however, such jobs were not available for the simple reason that no nation-state existed as yet. In these countries the bankrupt gentry and the failing native merchant class became, under the banner of Western liberalism, the principal protagonists of political reform and national independence. Thus while the history of the modern Western state may well be described as one of the rising middle classes in quest of larger, national markets, the history of the peripheral states is one of the declining middle classes trying to escape the vagaries of the market and hoping to find safe haven in political, rather than economic, entrepreneurship. As the old middle classes of the peripheries were transforming themselves into political entrepreneurs, the less desirable and more risky economic functions were gradually taken over by immigrants and members of low-status ethnic communities. In this process we see the reverse side of Veblen’s hypothesis, for if high-status expectations undermine the propensity for saving, low-status expectations encourage it. Thus from country to country we can witness not only the decline of the old, but also the rise of a new “pariah” middle class.39 In the semiperipheral regions, this form of entrepreneurship was largely restricted to money management, but as we proceed on the map toward the South, and especially toward the East, we find that more and more economic functions— including the management of large estates—are concentrated in the hands of ethnic entrepreneurs. In Russia, Poland, Hungary, and Romania, these entrepreneurs were largely drawn from Jewish immigrant communities, whose members replaced classes of German entrepreneurs long settled in these countries. In the Balkans, the cabal of modern entrepreneurs, the carsija, was mostly Armenian and Greek, the latter mainly immigrants from the districts of Girokaster and Korea, who, incidentally, were replacing an earlier generation of Phanariot Greeks. In Bulgaria, by the end of the nineteenth century only a minority of merchants and artisans were natives, the majority were Armenians, Turks, and Sephardic Jews.40 In Serbia, more than two-fifths of the urban bourgeoisie was of Albano-Greek, or Tsintsar, origin.41 In Iberia, this form of entrepreneurship was somewhat more limited, but even in Spain we can discern the influx of Levantine entrepreneurs, while in Portugal, English, Hansa, and French traders were rapidly replacing native merchants in the beginning of the nineteenth century.42 State and Society on the Periphery The impact of the Occident on the periphery was not purely material: it was cultural and political as well. People there were not only eager to eat, dwell, and dress, but also to write, paint, and think like Westerners. The Hungarian István Széchenyi, the Romanian Constantin Golescu, the Serbian Vladimir Jovanovic, all of them great figures of an


282

epoch of national renovation, gave rapturous accounts of their travels to Western countries and contrasted these countries’ virtues with the backwardness of their homelands and the indolence of their countrymen. Széchenyi complained bitterly about “dust and mud everywhere” on the great “Hungarian fallow,”43 while the Pole Jan Dombrowski in an oft-cited bon mot vowed to die for his fatherland any day, as long as he could spend his remaining days in Paris rather than in the backwater of his own sprawling estates. In the Balkans, one prominent observer noted, “the nations genuinely accepted the proposition that theirs was not a civilized society.”44 And so it went, a seeming paradox in an age of fervent nationalism. In reality, this nationalism also mandated that the societies of the periphery emulate the symbols and accomplishments of a few Western nations that had succeeded in universalizing the validity of their own unique historical experience. Nothing testifies more eloquently to the dominance of Western culture than the eagerness with which Western political institutions were imitated during century between the Napoleonic age and World War I. First in the long line of these imitators were the newly independent nations of Latin America, followed by the Iberian countries, and then by the countries of Eastern Europe, just emerging from under Ottoman and Habsburg tutelage. Ironically, the institutions of these states were sometimes more liberal and democratic than the models themselves. The Portuguese constitution of 1822, the Bulgarian constitution of 1879, or the Serbian constitution of 1889 are cases in point, with their provisions for universal suffrage and vast powers vested in the elective branch of government.45 In the nineteenth century, Russia alone among the European countries had no parliamentary government. But even there concessions were made to the principle of popular sovereignty in 1905 under the pressure of revolutionary events. In the long run, however, far more consequential than the introduction of parliamentary government was the building of modern bureaucracies based on impersonal rules, professional standards, formulated, ultimately, security of employment. Much like parliamentarism, the establishment of modern bureaucracies was spurred by the desire of elites to conform with “civilized” international standards. But the process of bureaucratization was further propelled by class interest, above all by the desire of the educated classes to find dignified salaried employment, and thereby to avoid the roughand-tumble of economic life. Thus the systems of public administration not only changed their character, but expanded at an alarming pace, or at any rate at a pace much faster than in England and in the Northwest of the Continent. The definition of a bureaucrat varies, and hence so do estimates of the actual size of public employment. According to one source, however, it exceeded 5 percent of the labor force in every peripheral country examined—compared to 2.4 percent for Germany and 1.5 percent for England—with 25 to 39.4 percent of the budgets being earmarked for civilian salaries.46 The truly troublesome aspect of the process of bureaucratization was not that there were many more administrators than were justified by existing levels of social complexity, but that their expectations, shaped by foreign examples, were incongruous with the underlying economic base. Inevitably, in response to these pressures, the new nation-states became instruments of revenue raising as well as of income transfer from the societies at large to the new bourgeoisie of state officials. These demands on scarce resources were compounded by the fiscal exigencies of


283

national defense. While available statistics on national income and expenditures per capita are less than perfect, they do allow us to discern the pattern of overconsumption or overspending in the peripheral states. As Table 1 shows, in 1880, the peripheral states of Europe spent on average 73.4 percent of the amount per capita spent by the core states; in 1890, 69.4 percent; in 1910, 44.7 percent. At the same time their per capita incomes may be estimated at about 30 percent of average income in the core societies. These percentages are derived from figures unadjusted to the purchasing power of the American dollar. If we use the available, and admittedly crude and fragmentary information on price indices, as we have done in Table 2, then the disparities between core and periphery (as well as between the small and large countries of the semiperiphery) will be still more pronounced. Insofar as relative spending in the peripheral societies shows a declining trend, this does not so much reflect a lessening appetite for public spending as the gradual exhaustion of revenue bases and the shrinking sources of foreign credit. (It also reflects the beginnings of welfare-oriented, redistributive state spending in some of the core societies, the budgets of which serve as the baseline for our comparisons.)

Table 1: Income Per Capita and State Expenditures, 1880–1910 (International Units and U.S.Dollars) Income per Capita State Expenditures per Capita a (U.S.$) (IU’s) Ca. % of 1880 % of 1890% of 1910 % of 1910– Core Core Core Core 13 Core Countries 1,211 6.4 5.8 9.4 U.S.b Canada 1,182 7.8 7.5 12.2 Britain 1,072 11.9 11.2 17.1 Netherlands 975 11.3 14.8 14.1 – – 38.2 Australia 742 New Zealand 866 – – 44.5 1,008 100.0 9.4 100.0 9.8 100.0 22.6 100.0 Average Semiperiphery: Small Countries Switzerland 764 2.8 4.4 8.3 Denmark 688 6.9 7.7 13.7 Norway 522 6.9 6.2 13.2 Sweden 567 4.5 5.2 11.3 Average 635 63.0 5.3 56.4 5.9 60.2 11.6 51.3 Semiperiphery: Large Countries 352 9.8 15.6 24.7 Austria-Hungary France 786 17.5 16.6 21.1 German 829 16.2 22.2 33.9 Italy 410 9.7 12.1 14.2


Average European Periphery Bulgaria Greece Portugal Russia Romania Serbia Spain Average Latin Periphery Argentina Brazil Chile Costa Rica Cuba Uruguay Average

284

594

59.0 13.3 141.5 16.6 169.4 23.4 103.6

317 383 310 306 298 271 511 342

9.2 10.6 8.0 3.5 5.3 2.5 9.3 33.9 6.9 73.4

5.3 7.9 11.6 3.6 6.1 4.1 8.7 6.8 69.4

7.7 10.0 14.3 8.3 12.9 7.7 10.1 10.1 44.7

350C

10.6 – 4.4 6.5 – 16.0 34.7 14.4 153.2

23.8 5.7 27.6 12.7 – 21.3 18.2 185.7

24.9 4.7 21.3 8.6 18.8 23.8 17.0 75.2

Sources: Income per capita, Colin Clark, The Conditions of Economic Progress (London: Sources for Table 1 (cont.) Macmillan, 1940), 40, 87, 98, 109, 129, 131–32, 136, 139, 141, 144, 146, 148, 151. State expenditure per capita, Arthur S.Banks, Cross-Polity Time-Series Data (Cambridge, MA: MIT Press, 1974), 99–136.

a. Income per capita in IUs defined as “the amounts of goods and services which one dollar would purchase in the U.S., 1925–34.” All IU figures refer to the working population (labor force, employed or unemployed). Expenditures in U.S. dollars are unadjusted to purchasing parities. b. Figures for the U.S. and Germany adjusted to reflect budgets of states and Länder respectively. Adjustment on the basis of information and averages in Thomas F.Borcherding Budgets and Bureaucrats: The Source of Government Growth (Durham, NC: Duke University Press, 1977), 27: Vierteljahrhefte zur Statistik des deutschen Reichs, 1902, No. 2, 257; 1914, No. 2, 123. c. Estimated to be “in the category of 300–400 units,” Clark, 53. Faced with this obvious gap between the needs of their states and the structure of the underlying economies, the political classes first attempted to deal with the problem by foreign borrowing. Thus the debts of these states, virtually nonexistent in the 1860s, began to accumulate in the 1870s, to reach extraordinary proportions in the following decades.47 Borrowing, however, could only postpone the need for more tax revenue, and when the governments finally set about raising taxes, they did so by exceedingly harsh means. In Eastern Europe, as well as in Iberia, tax collection drives sometimes took the form of military campaigns, in the course of which troops would occupy entire villages, search the houses of delinquent taxpayers, and proceed to remove anything of value. Even short of such methods, revenue raising remained a brutal business that generally


285

relied on arbitrary assessments or levies on such staples as salt, kerosene, liquor, and tobacco—a system of indirect taxation invented by ministers of finance in peripheral countries, for example, Vyshnegradsky and Witte in Russia, or Wekerle in Hungary. Where the design was effective, it would transfer income from the public at large to the political classes. Where, as in the Balkans, governments eschewed such highly regressive tax systems, or where, as in Romania and Russia, the system of collection remained ineffective, the revenues would be collected anyway by individual bureaucrats, who used the power of their office to extort bribes that would keep them in a style they regarded as commensurate with their social status. In either case, the peripheral state became a grand instrument of income equalization, not, to be sure, among the various economic strata of peripheral societies, but between the elites of the backward and the advanced industrial societies of the Continent. Not surprisingly perhaps, these methods of income transfer, and the resultant pattern of income inequality, engendered widespread opposition by the taxpaying public. To forestall it, peripheral bureaucracies began to use their administrative leverage to elect candidates sympathetic to their view of the raison d’état. Over the years, the incidents of administrative pressure and fraud multiplied, but around 1875, perhaps in the wake of the downturn in agrarian prices that affected the whole continent, these methods became pivotal to the functioning of political machines. These machines enabled the bureaucratic arm of government, usually under liberal and progressive labels, to turn out predictable majorities, thereby debasing parliamentarism without abandoning it.

Table 2: Relative State Expenditures in U.S. Dollars, 1910 (Adjusted to Purchasing Power Parities) Multiplier Adjusted State Spending per Percent of Coefficienta Capita Core Core Countries U.S. 1.000 9.4 Britain 1.075 18.4 Netherlands 1.266 17.9 Australia 0.794 30.3 19.0 100.0 Average Semiperiphery: Small Countries Switzerland 0.926 7.7 Denmark 0.962 13.2 Norway 0.833 11.0 Sweden 0.926 10.5 10.6 55.8 Average Semiperiphery: Large Countries 1.353b 33.4 AustriaHungary France 1.429 30.2 Germany 1.163 39.4 Italy 1.030 14.6


Average Periphery Turkey Greece Spain Poland (Russia) Average

1.266 1.886 1.613 1.933

286

29.4

154.7

n.a. 18.9 16.3 16.0 17.1

90.0

Source: Colin Clark, The Conditions of Economic Progress (London: Macmillan, 1940), 51.

a. The purchasing power of one dollar in 1925 in given country, b. Calculated from Austrian, Hungarian, and Czech figures. The story of these machines, and the corruption of the electoral systems, shows striking similarities from country to country. One of its first examples can be found in Spain, where the disorderly “praetorianism” of the mid-nineteenth century was liquidated by the Liberal Unionist machine of Premier Leopold O’Donnell (1856–1863).48 It was from his premiership onward, one historian notes, that we can no longer speak of a “parliamentary system with abuses, for abuse became the system itself.”49 In this new system, which was to survive until the 1930s, ministerial changes preceded rather than followed elections the ministry set up a state of “official candidates,” many of them drawn from the public administration, who were then duly returned with the help of the administrative machine, headed by the minister of the interior.50 Much in the same vein, the liberal constitution of Portugal was debased toward the end of the century. Under the regime of Premier João Franco (1906–1908), the slogan of the government became pouca politica, muita administracão (little politics, much administration).51 Elections lost their meaning, especially in the countryside. Indeed, except for Lisbon and Porto, majorities were regularly “made” with little regard for public opinion in the country.52 Half a continent away, a nearly identical system was set up in Hungary by Premier Kalman Tisza (1875–1890). In this system, about 160 of the 413 constituencies of the House of Representatives, inhabited by non-Hungarian national minorities, were turned into rotten boroughs, in which election results were regularly rigged by the bureaucracy. The beneficiary of this rigging was Tisza’s Liberal Party, which due to this advantage became an established government party, winning majorities regularly. In 1918 the old system collapsed, but was soon resuscitated by Premier István Bethlen (1921–1931), whose Unitary Party replaced the old liberals and won successive majorities by maintaining a system of open balloting in 199 rural constituencies, while conceding some of the 46 urban districts to the parties of the opposition.53 The experience of neighboring Romania was similar. In 1876, Premier Ion “I.C.” Bratianu set up a political machine to the benefit of his Liberal Party. Within the machine, the prefects of the counties (judete) were turned by Bratianu into “petty satraps,” whose activities made it “exceedingly difficult to agitate against the government…even for members of the boyar class.”54 Thus, according to another historian, “as the power of the salaried bureaucracy increased, so did in proportion the power of the landed class decrease.”55 In Hungary, the rigging of elections was confined to constituencies inhabited by minorities (and later, to the


287

countryside), but in Romania, the system of “engineering” results by pressure or fraud was universal. In consequence, parliament was usually dominated by a single party, faced only by the token opposition of a handful of deputies.56 In the Balkans, Premier Iovan Ristic of Serbia “gathered around those liberals who were willing to accept his bureaucratic way of operating,” and created a machine to “prevent the formation of true political parties.”57 In Bulgaria, a country extremely proud of its democratic constitution, Stefan Stambulov’s liberals created an administration so tightly organized that “not even a bee could hum” without the knowledge of the premier.58 The regime relied on “prefects, municipal councils and a political machine of hired thugs and rumor mongers to create an atmosphere of fear,” in order to dissuade the majority of voters from casting ballots against his government.59 Under Stambulov only 5 percent of the electorate went to the polls, but the official record showed the participation of two-thirds of those eligible.60 Russia alone had no political machine, for until 1905 no parliament existed, and after 1905 its constitutional powers were so limited that hostile majorities could have little effect on public policy. Overconsumption and political corruption, however, do not tell the entire story, for the elites of the peripheral countries were not impervious to the existence of broader national needs, above all, the need for material progress. As political classes, these elites were dependent on the power and prestige of the state—attributes that required a steadily expanding economic base. Thus, from the very inception of statehood, political classes responded to the exigencies of development by a set of policies often described as “neoliberal.”61 Awkward as this designation may seem, it does capture the essence of the policy. For although the qualifier “neo” tells us that the states were now ready to interpose themselves between domestic and international markets, and to justify the repression of political freedoms by the need for material progress, they remained “liberal” in two significant ways. One, they accepted the legitimacy of markets as distributive instruments of last resort. Two, even at their most oppressive, they maintained some commitment to procedural norms and the objective standards of a legal state, in part because these norms were regarded as the main symbols of “civilized conduct,” and in part because of an instinctive understanding that markets could not function effectively in the absence of a system of law. It was mainly on account of these understandings that radical opponents of the regimes—even assailants and would-be assassins of prefects, prime ministers, and kings—could find protection in the courts of law, or that radical students could seek refuge from the police under the shield of the autonomy of universities. The policies of development, however, became the source of yet another form of strain on these already overburdened economic systems. In order to pursue them effectively, the governments needed funds to create expensive infrastructure—mostly railroads—and then to lay down the foundations of an industrial economy. Foreign capital could do part of the job, and to attract it the governments had to subsidize profit rates directly by grants and indirectly by guaranteeing the flow of cheap comestibles to the industrial labor force. Some governments, namely Serbia and Bulgaria, wavered, but most dealt with the problem by shifting the costs of development to the agricultural sector, and within it, to the agricultural proletariat. This was accomplished by the heavy protection of low-quality domestic manufactures, including textiles and producers’ goods required by agriculture.


288

Further, strikes by agricultural laborers were prohibited, and other draconian measures were imposed, such as the Hungarian and Romanian laws that made the breach of labor contracts punishable by heavy fines and imprisonment. In the short run, these policies may be said to have borne some fruit, for many peripheral countries experienced a spurt of industrial development between 1890 and 1910. In the long run though, they exacted a heavy price, for they produced demoralized, alienated, and rebellious rural populations. Indeed, from the 1890s onward, the frustrations of the peasantry kept the societies of the European periphery in a virtual state of civil war.62 “The Horrors wreaked by the Rumanian peasantry [in 1907], the bloodthirsty land struggles of Ireland, Galicia and Andalusia, the turmoil in Russia,”63 the presistent violence on the Hungarian plain, and the periodic outbursts of peasant fury in Serbia, were all responses to a declining or stagnating rural standard of living, and to contain them governments were forced to redirect some of their resources to create new instruments of coercion. All over the European South and East, the countryside was turned into an armed camp, as governments organized special forces of rural gendarmes that, with their military uniforms, equipment, and tactics, were not so much conventional constabularies as armies of occupation within their own nations. The exact cost of this repression is hard to measure. But force and counterforce were likely links in a chain of causality that led first to the stagnation, then to the decline, of agrarian production throughout the South and the East (including East Prussia and Southern France). In the long run, this decline could not be offset even by the rise of new industrial economies, and would be reflected in the persistence of the pattern of regional income distribution that had been evident since the seventeenth and eighteenth centuries.

Table 3: Yield of Cereals by Region, NW/SE Axis (Metric CTW/HA) 1913 1928 1936 Netherlands 26.7 25.1 25.5 Belgium 24.3 26.5 26.2 Germany, West 22.5 22.9 22.7 Denmark 21.0 22.8 22.4 France, Northeast 19.9 26.3 18.9 19.4 16.8 n.a. Italy, North 18.5 17.0 16.5 Germany, East Sweden 17.9 19.9 20.6 Bohemia (Czech lands) 17.9 16.1 15.6 Austria 15.6 14.3 14.5 Hungary 15.7 15.3 13.6 France, South 11.7 11.5 11.3 Romania 11.7 10.0 11.1 10.5 9.0 n.a. Italy, South Poland 10.5 10.1 9.6 9.9 13.2 12.2 Yugoslavia (Serbia) Spain 8.0 8.3 8.3 Bulgaria 6.8 11.7 13.3


289

Source: Except for Sweden, based on League of Nations, Agrarian Production in Continental Europe (Geneva, 1943), 86–91. For Sweden average of winter wheat, rye, and barley In Historik Statistik for Sverige (Stockholm, 1959), 60. So far, we have focused on economic and political contrasts between the European core and its periphery. Between these two regions, however, we have also recognized the existence of a third, the semiperiphery, the experience of which is congruent with neither the core nor the peripheral model. As we noted above, this region—encompassing much of France and Germany, Northern Italy, Switzerland, Austria, Bohemia, and the Scandinavian countries—was exposed to the agrarian innovations of the core region prior to the critical period from 1780 to 1830. Its agrarian economy thus could best be described as semideveloped. This term implies a higher degree of labor productivity than in the peripheral regions (but lower productivity and capital intensity than in the core) and, at the same time, the prevalence of semicapitalist enterprise, above all the mixing of wage labor with more traditional forms of labor exploitation. Given these forms, the agrarian proprietors of the semiperiphery were not like the traditional seigneurs and peasant farmers of the periphery. Yet they were not an agrarian bourgeoisie either; they retained a separate identity, the survival of which was perhaps best reflected in the rise and continued existence of self-labeled agrarian parties in politics. This experience of the agricultural economy set the stage for a distinct pattern of industrial development. Contrary to Gerschenkron’s famous argument about France and Germany,64 the role of the state in the process was far from negligible. But state intervention left sufficient room for private initiative which, as Gerschenkron correctly points out, had a strong “finance capitalist” component65 in that much of the investment in infrastructure and heavy industry came from private savings, channeled into the economy by a highly developed banking system. This institutional device was a logical response to the needs of the semiperipheral economies, as well as to the unique opportunities they afforded. On the one hand, these areas were less developed than the core: this created pressures to “catch up,” and hence the need for larger concentrations of capital than the individual entrepreneur could provide. On the other hand, the semiperiphery possessed a more developed agricultural sector than the peripheral regions, which made it possible to rely on voluntary savings. The brothers Péreire, who invented the crédit mobiliére, had to invent it in this economic setting. Their invention would have been superfluous in England, and meaningless in, say, Bulgaria or Russia, where the rate of personal savings was low, and where, as Gerschenkron tells us, business ethics were too weak to inspire confidence on the part of institutional lenders.66 This pattern of development created its own peculiar contradictions and social cleavages. The autonomous development of the agrarian sector, and the participation of private capital in the rise of the industrial economy, created a bourgeoisie and civil society much stronger and more differentiated than on the periphery of the Continent. Nowhere does this difference become more obvious than in some of the great social novels of the semiperiphery and the periphery, in comparisons between the characters in Mann’s Buddenbrooks, Balzac’s Les illusions perdues, or Stendhal’s Le rouge et le noir


290

on the one hand, and the heroes of Chekhov’s Cherry Orchard, Tolstoy’s Anna Karenina, or Goncharov’s Oblomov on the other. If the first set of novels describes an ascendant bourgeois society, the second portrays bureaucrats, landowners, military officers, impoverished peasants, and scoundrel entrepreneurs. At the same time, the bourgeoisie and the civil societies of the semiperiphery were more ambivalent and weaker than their counterparts in Britain. In part, this was simply due to the competitive disadvantage these semideveloped economies faced in trying to carve out for themselves a slice of the global market dominated by Britain. But this weakness was further compounded by the fact that, unlike those of Britain, the propertied classes of the Continent were split—between agriculture and industry, between finance capital and family enterprise, as well as between the “ethnic” and the “native” entrepreneur—into groups that viewed themselves with a “mutual incomprehension”67 that left public opinion and parliaments “too divided to parcel out the contested social product.”68 Consequently, while bureaucratic-military machines would never gain the same ascendancy as in Southern and Eastern Europe, they were strong enough to insulate large areas of public affairs from the jurisdiction of parliaments.

Table 4: National Income Growth per Capita of Working Population (International Units) Ca. 1910 Percent of 1925/34 Percent of Percent Core Core Change 1910– 1925/34 Core Countries U.S. 1,211 1,368 Canada 1,182 1,337 Britain 1,072 1,069 Netherlands 975 855 Australia 742 1,016 New Zealand 866 1,202 Average 1,008 100.0 1,141 100.0 13.2 Semiperiphery: Small Countries Switzerland 764 1,018 Denmark 688 680 Norway 522 596 Sweden 567 696 635 63.0 748 65.6 17.8 Average Ca. Percent of 1925/34 Percent of 1910 Core Core Semiperiphery: Large Countries Austria-Hungary 352 386a

Percent Change 1910–1925/34


France Germany Italy Average European Periphery Bulgaria Greece Portugal Russia Romaniab Yugoslaviab Hungaryb Spain Poland Average

786 829 410 594 317 383 310 306 298 271 274 511 320 332

59.0

684 646 346 515

45.1

–13.3

29.7

2.1

291

289 397 342 267 238 327 359

32.9

550c 281 339

Source: Colin Clark, The Conditions of Economic Progress (London: Macmillan, 1940), 40–42, 77, 85, 87, 91, 98, 109, 126, 131–32, 135–36, 139–44, 148, 151.

a. Former territory, b. Substantial territorial changes, c. Estimated to be “in the category 500–600 units,” Clark, 42. This rough equilibrium between state and society, between political and economic classes, was institutionalized in different ways, reflecting the particular traditions and histories of individual countries. In republican France, and under the constitutional monarchy of Italy, bureaucracies exercised their powers de facto, by moving into the political vacuum created by parliamentary immobilisme. While no laws existed to this effect, public finance, defense, and the administration of colonies were largely left to professional officers and civil servants. In imperial Germany and Austria, the division between civil societies (represented in parliaments) and the bureaucratic-military structure of dynastic states was a matter of constitutional law. While in both countries the monarchies recognized parliaments with liberal suffrage and clean elections, they also reserved for themselves vast prerogatives with respect to defense and foreign affairs. Likewise, we find elements of this constitutional dualism in Scandinavia in the first half of the nineteenth century, in that the monarchies still exercised traditional powers over the appointment of ministers and the conduct of war, while the interests of the political classes were manifest in weak attempts to retain control over territories (SchleswigHolstein, Norway) that had been acquired during an earlier period of dynastic absolutism. By its very nature, however, this equilibrium between political and economic classes, bureaucracies and civil societies, was tenuous, and it began to tilt in one direction or another in the second half of the nineteenth century, producing two different political outcomes. In their attempts to explain these differences, historians as well as social scientists have usually turned to variations in the social structure or cultural makeup of


292

individual countries. The presence and survival of traditional economic and social sectors appears to be a particularly appealing hypothesis.69 Still, we may do better overall by approaching the problem within the context of state interaction with the international political environment. Specifically, variations in political outcomes may successfully be linked to the size of a given country and its power potential within the international system. On one end of this spectrum, the elites of the smaller countries were constrained in their choices by a rational assessment of their relative inability to manipulate the international system. The Swedish elites may have come to this conclusion as early as the Napoleonic wars, the Danish after the war with Prussia in 1864. These experiences may well have persuaded the respective political classes that they should improve conditions in their countries not by waging war, but by helping the rising bourgeoisie to adjust to the vagaries of global markets. Thus, over the years, public policy moved through successive stages of economic liberalism, neomercantilism, and finally welfare etatism. This was not the case, however, in the major countries of the semiperiphery, in Germany, France, Italy, and, until 1918, Austria. In these larger and potentially more powerful entities, the political classes could reasonably entertain the strategic alternative of improving their position not by adjusting to the rules of the market, but by adjusting the very rules of the economic game by military and political means.70 The political classes of these societies could not, however, have carried out this strategy on their own. In order to pursue it effectively, they needed allies, and to find them they offered the struggling entrepreneurs of their countries both protection from socialist harassment and a heroic vision of a new continental order in which their bourgeoisie could compete on its own economic terms. In times when the economy was good, the bourgeois classes resisted these blandishments, but in times of hardship, entrepreneurial support for parliamentarism eroded quickly, producing periodic “migrations of bourgeois republicans to the Right over economic and social issues.”71 As the crisis deepened, this “freefloating Right” joined hands with the political classes of the state to form a Bonapartist dictatorship under the leadership of a heroic figure, who promised to raise, or to restore, the nation to its rightful place in the international system. Over time the idiom would change, but the constituent elements of the coalition would remain the same. Bonaparte and Hitler, Boulanger and Mussolini, or Pétain, would appeal to the same social instincts, though their rhetoric and ideological formulae would reflect some fundamental differences between the milieux of the nineteenth and the twentieth centuries. In retrospect, these Bonapartist experiments seem to have been costly exercises in futility. For one, the maintenance of large civilian and military bureaucracies required the diversion of resources from productive investment into partly unproductive state consumption. Thus while the larger countries of the semiperiphery experienced stagnation and relatively low rates of economic growth, the lesser countries prospered and developed democratic institutions cast in the British mold. In the long run, to be sure, even the larger countries attained political democracy and economic prosperity, but only after external forces had reduced them to the status of small powers in the international system. This was the experience, after World War II, of Germany, Italy, and Japan, and of France after the wars in Indochina and Algeria. Ironically, the path of these countries from the semiperiphery to the core was paved not by military victory, but by defeat and the overall reconstitution of the international political system.


293

Revolution and Totalitarianism By the third quarter of the nineteenth century, then, a clear-cut pattern of economic and political differentiation had emerged on the Continent. This enables us to distinguish among three zones, or regions, in terms of different (1) levels of economic productivity and types of labor organization; (2) social mobility and class structure; and (3) structure of public authority. Overall, and within each of these zones, a geographically regressive pattern of productivity is apparent, although material expectations continued to be relatively uniform. This imbalance between expectations and the means of production created increasing pressures on the state to use coercive methods of resource extraction and was, in the last analysis, responsible for ever greater degrees of etatist authoritarianism along the Northwest-Southeast axis of the Continent. At the turn of the century these pressures were further increased by what is frequently referred to as the “rise” of the masses and their “entry” into the political mainstream. This phenomenon is generally associated with the spread of literacy and urbanization, two processes that are seen to be instrumental in “mobilizing” the lower classes and making them “available” for systematic and sustained political participation.72 This may well be an important correlation that is corroborated by impressive empirical evidence, but, like so many other aspects of peripheral politics, the entry of the masses into politics cannot be examined in isolation from major developments in the core of the modern world economy. Most significantly perhaps, during the last decades of the nineteenth century, the lower classes of Britain and of other core societies became enfranchised, and thus a part of the political community. Almost simultaneously, their living standards began to show substantial improvement. After nearly a century of stagnation, urban wages in Britain began to rise, while working hours started to decline.73 These instances of economic and political emancipation were not lost on the working populations of the less developed countries. Indeed, knowledge of these gains spread quickly from country to country, with the laboring classes of the North and West serving as reference groups for those further to the South and the East. Just as for the middle classes a century before, so for the laboring classes, the new standards of living carne to be “accepted as a staple of decency,” as well as a staple of political programs. Although often accused of insensitivity and rigidity, the old political classes responded to the challenge. Operating in an arena of great scarcity, however, they had to make hard choices, and in most instances they attempted to accommodate the demands of the industrial workers rather than the rural population. In part, this was because they perceived them to be more dangerous than peasants, in part because the liberal elites had to contend with market forces that propelled the movement of skilled labor and technical expertise toward the more developed regions of the Continent. Pondering these forces and fears, the governments granted urban workers the right to vote while the peasants remained disenfranchised.74 Still more significantly, urban workers were allowed to organize. By 1905, the right to strike was grudgingly recognized even in autocratic Russia. As a result, urban wages rose. Contemporary labor statistics, of course, are notoriously unreliable and incomplete, but all available evidence points in one direction:


294

the further we travel from the industrialized core region of the Atlantic Northwest, the higher industrial wages became relative to per capita national income, and to the income of the agrarian population. This trend pushed industrial, especially skilled, labor toward the middle of the income structure at the expense of the peasantry, a pattern of income distribution that may well explain the status and ambiguities of the working classes in the periphery, and their reluctance to support left-wing political movements.75 Ironically though, the political benefits gained by rising urban living standards were minimal. While the industrial working classes gained income in comparison to the peasantry of their own countries, they still lagged far behind the standard of their reference groups—the industrial working classes of the advanced societies. Thus while industrial workers did not always support the Left, their politics was nevertheless radical and anti-establishment. The most obvious instances of this occurred in Hungary and Romania, where in the 1905 much of the working-class vote was captured by the Arrow Cross, the Iron Guard, and other parties of the radical Right. The turbulent entry of the lower classes into politics at the end of the nineteenth century was closely related to another important development: the rise of a “radical intelligentsia” in the countries of the periphery. Conventional explanations for this phenomenon usually emphasize economic scarcities: too many people in pursuit of too few jobs, with some of the losers in this race for status and security turning against the established order in bitter disappointment. This interpretation becomes suspect in the light of statistical evidence that points to the truly minuscule size of the educated public in peripheral societies before World War II, and especially before World War I. Thus in the first decade of this century, the number of students enrolled at Bulgarian universities was 2,455. The figure was 5,925 for Romania, 13,737 for the Russian and Austrian provinces of Poland, 11,893 for Hungary, representing only .049, .069, .057, and .053 percent of the respective populations. (The figures for Germany, France, and Italy were .11, .901, and .059 percent.)76 The quantitative argument becomes still weaker if we consider that many of the radicals came from comfortable middle-class backgrounds that could well have served as stepping stones for social mobility through the ordinary avenues of advancement. That they did not take this route was less because of the lack of opportunity than because of the availability of an alternative, more risky, but ultimately more promising, opportunity to enter politics as the leaders of deeply frustrated popular constituencies. Indeed, these men and women were not so much an “intelligentsia” as another class of political entrepreneurs, recruited from the most imaginative and able risk-takers in the talent pool of their societies. These entrepreneurial qualities, much more than formal education, distinguish these new political men and women from the dull, bureaucratized establishments that they sought to replace by mass mobilization and revolutionary violence.77

Table 5: Relative a and Absolute Indus strial Wages (1925–1934) Wages per Industrial Wage as % Industrial Wage as % of Week in U.S. of per Capita Income British Average Industrial $ Wage United 17.3 94 100.0


Kingdom Switzerland Germany Holland Uruguay Czechoslovakia Spain Poland Italy Yugoslavia Mexico Estonia Bulgaria

150 13.8 13.9 12.0 11.0 9.4 8.0 7.1 7.1 6.0 5.2 3.7

79 129 106 127 125 120 148 120 111 130 106 111

295

86.7 79.8 80.3 69.4 63.6 54.3 46.2 41.0 41.0 34.7 30.1 21.4

Source: Colin Clark, The Conditions of Economic Progress (London: Macmillan, 1940), 40–41, 47. Given the goal of mobilization in the context of mass politics, the political style of the radical intelligentsia was agitational. While the populists were “going to the people,” and the Marxists were attempting to “raise the class consciousness of the masses,” the Right went “into the streets,” as did the French Camelots (literally, street hawkers) du Roi in the first decade of the new century. But this new radicalism was not just a matter of style but of substance as well. Unlike the liberal “fathers” who were committed to an incrementalist view of politics and preached adjustment to the existing rules of the global economic and political game the radical “sons” were inspired by grand chiliastic visions of abolishing markets and replacing them with a new global system. If populists and Marxists envisioned a world in which exchange would be freed from the principles of utility and self-interest, their counterparts on the Right set out to replace the “fraud” of the market with the harmony of a universal hierarchy in which the strong would command and the weak would obey. These ideologies, then, provided both logic and justification for the new, totalitarian states of the period following World War I. The promise of a perfectible universe emancipated revolutionary elites from both moral and legal restraints, and once emancipated they proceeded not only to capture the state, but to attack the autonomy of the economy, the culture, and the society as well. Yet the results of this attack varied, reflecting patterns that had evolved through the centuries. If in the East the Soviet state replaced all autonomous social groups and forces, in Central Europe the fascist states of the interwar period, although they declared the totality of their authority, in reality were forced to tolerate an intricate web of de facto social autonomies.78 There is a significant lesson in this. Totalitarian states, like all others, are the products of both continuity and change. These totalitarian states have, on occasion, been described as “developmental” by posterity. No designation could be more misleading. Lenin, Hitler, Mussolini, and their followers on the peripheries did not want to “catch up” with the advanced nations, but


296

wanted to change the existing structure of the world system. True, in time, they and their successors had to formulate economic policies, but their policies of mobilization, whether in Hitler’s Germany or in Stalin’s Russia, were not ends in themselves but means toward the achievement of political ends. It was only much later, in the last decades, that some communist states would seriously contemplate the prospects of correcting economic backwardness by economic means. In doing so, they may well be slowly reverting to some of the institutional features of neoliberalism. In the interim, regional income distribution on the Continent has barely changed. The trip from Amsterdam to Moscow still leads through the same descending path as it did about two centuries ago.

Conclusions This brief examination of European history reveals a strikingly regular geographical pattern of regional income disparities, the origins of which go back to the agrarian revolutions of the early modern age. This pattern of regressive regional income distribution, already evident in the eighteenth century, was thereafter reproduced by the rise of a continent-wide material culture and, in the final analysis, by the ongoing technological revolutions of the Occident. This international demonstration effect, and the broader cultural definition of wants, had two direct consequences. On the one hand, they affected savings: the rate of saving either declined, or, if adequate rates were maintained by force, the costs of coercion tended to exceed the economic benefit. On the other hand, and still more debilitatingly, when some economic gains were registered against all odds, the results tended to be vitiated by invidious comparisons with the accomplishment of the core societies. Although the theory of the demonstration effect is sometimes presented in juxtaposition to other explanations of international polarization and peripheral drift,79 in reality these theories are closely intertwined. Indeed, some aspects of market and state behavior can only be understood in the larger context of changing human needs. Thus, if there are indeed secular trends in the terms of trade and in the relative movement of prices favoring manufactures over primary products, these trends and movements can hardly be divorced from economic demand, and economic demand would hardly be the same in the absence of the Veblen effect. Likewise, peripheral states are not born profligate; they become profligate when the demonstration effect transforms the expectations of classes that control the levers of power and have the potential to convert private need into public policy. Other aspects of market and state behavior are independent of the Veblen effect, but they magnify its adverse consequences by converting global into local inequalities. Thus, when opting for a modern state, the elites of a backward society accept a potential framework for inequality, and when facing global labor markets, peripheral governments are forced to make an agonizing choice among the loss of talent, coercive restrictions on mobility, and the acceptance of domestic income disparities that will further exacerbate tensions in their societies. The pattern of economic differentiation correlated closely with the political differentiation of the Continent into progressively more pluralistic and authoritarian states. The further a society was located from the core, the poorer it was destined to


297

remain, the more power its political classes accumulated, and the greater would be the scope of state authority. This correlation between etatism and material scarcity is consistent with our traditional views of economics and politics; it also explains a great deal about patterns of class formation on the European continent. This is not to say that the political history of Europe can be reduced in toto to the economic categories of scarcity and need: ethnic identities, for instance, were independent of economics, and they largely determined the boundaries of modern states and their destiny as small or large powers. Still, when it comes to explaining political structure, economic explanations are more parsimonious and more powerful than those cast in terms of national cultures and social configurations. For the social scientist then, the most significant question is whether and to what extent this historical experience is likely to replicate itself in the contemporary world. At a first glance around the globe, it is very tempting to suggest that the past will provide us with a good model of the future, that the experience of the European peripheries foreshadowed today’s economic, social, and political outcomes. True, the neatly regressive patterns of Europe may not replicate themselves, but distinctions between core and periphery continue to be relevant, together with the appropriate patterns of class formation and political development. Political classes, pariah entrepreneurs, neotraditional landowners, and progress from bureaucratic-military to revolutionary regimes are, after all, as familiar to the student of the contemporary world as they should be to the historian of Eastern and Southern Europe. And if the politics of the semiperiphery of the nineteenth century was at least in part the product of unique ethnocultural and geographic configurations, the historical record is still useful to highlight the choices available to societies that are located between the core and the periphery of the global economy. Upon closer examination, however, the picture becomes more ambiguous, for along with the numerous similarities we must also account for a few deviant cases that do not follow the patterns set by Europe in the nineteenth and early twentieth centuries. More specifically, while in the European experience regional income inequalities and underdevelopment were reproduced with monotonous regularity, in today’s world some societies have broken out of the iron cage of backwardness and moved up several rungs on the ladder of economic success. The experience of Japan and of the newly industrialized countries of the Pacific Basin are only the most obvious examples of such mobility that require explanation. Following the logic of our initial argument, explanations should be sought both inside the social system and outside it. The internal factors relate to material, geographical, and cultural endowments that had no counterparts in the European experience: a more favorable geographical location, unusual abundance of globally scarce resources, or, to use Veblen’s phrase, an unusual “cultural pedigree”80 that would enable a population and its elites to resist, or to manage successfully, the international demonstration effect. What potential modernizers in the non-Western world need is not just the culture and virtues of Western humanity, or some of their functional equivalents, but virtues far stronger and more resilient than those of the European continent. The external factors meanwhile refer to the terms of economic and political exchange between the core regions and their peripheries. These terms of exchange have in the past been largely circumscribed by technologies of production, transport, and warfare. In the


298

“old” industrial age that began in the eighteenth century, these technologies required costly capital, elaborate skills, and elaborate infrastructures that slowed down the speed with which technology could be diffused from one part of the world to another. Today, however, we may well be on the threshold of a new industrial age. Compared to those of the nineteenth century, many of today’s production technologies require less investment in infrastructure and human capital, and the same can be said about some technologies of mass destruction that, unlike the armies of the nineteenth and early twentieth centuries, can be acquired and utilized without a fully developed industrial base. In the wake of these changes, we may be seeing the rise of a new world system, in which we may well encounter spectacular breakthroughs in potentials for production and destruction that can change the face of whole societies within relatively short spaces of time. For the time being, of course, much of this remains in the realm of speculation, and to turn speculation into theory we will have to engage in further “huge comparisons”81 not only among European societies of the past, but also between their examples and the challenges facing the peripheries today. Such comparisons, we are aptly reminded, are full of intellectual perils. But without them it will be difficult to develop a more comprehensive view of political change than we today possess.

Notes This paper has been prepared under the auspices of the project “Models of Development and Theories of Modernization,” sponsored by the Joint Committee for East European Studies of the Social Science Research Council and the American Council of Learned Societies. 1. See, among others, Immanuel Wallerstein, The Modern World System (New York: Academic Press, 1974); Theda Skocpol, States and Social Revolutions (Cambridge: Cambridge University Press, 1979); Perry Anderson, Lineages of the Absolutist State (London: New Left Books, 1974). 2. See especially, André Gunder Frank, “The Development of Underdevelopment,” and “Economic Dependence, Class Structure, and Underdevelopment Policy,” in James D. Cockroft, André Gunder Frank, and Dale J.Johnson, eds., Dependence and Underdevelopment (New York: Doubleday, 1972), 3–46; and Wallerstein (fn. 1), 302. 3. Gunnar Myrdal, Rich Lands and Poor (New York: Harper, 1958), 23–38; Albert Hirschman, The Strategy of Economic Development (New Haven: Yale University Press, 1962), 183–98. 4. Hirschman (fn. 3), 187. 5. Myrdal (fn. 3), 23. 6. Johann-Gottlieb Fichte, “Der geschlossene Handelsstaat” [The closed trading state] in Fichte, Ausgewählte Werke [Selected works], Vol. III (Darmstadt: Wissenschaftliche Buchgesellschaft, 1964), 417–544; Friedrich List, The National System of Political Economy, trans. Sampson S.Lloyd (London: Longmans, 1916), Alexander Gerschenkron, Economic Backwardness in Historical Perspective (Cambridge, MA: Harvard University Press, 1962), 5–30. 7. This hypothesis is implicit for Perry Anderson, who relates the rise of absolutist states in Europe to “accelerated pressures” for military spending by great powers under conditions of economic backwardness. See Anderson (fn. 1), esp. 494.


299

8. This formula appears in Jean-Philippe Platteau, “Das Paradoxon des Staates in wirtschaftlich rückständigen Ländern” [The paradox of the state in economically backward countries], Oesterreichische Zeitschrift für Soziologie 9, No. 4 (1984), 63–87, esp. 70–74, where the author generously attributes the idea to me. 9. Thorstein Veblen, Imperial Germany and the Industrial Revolution, 3d ed. (New York: Viking, 1954), 147. For Veblen’s thought and the concept of the demonstration effect, see Andrew C.Janos, Politics and Paradigms: The Changing Theories of Change in Social Science (Stanford, CA: Stanford University Press, 1986), 84–92. 10. W.G.Runciman, Relative Deprivation and Social Justice (Berkeley: University of California Press, 1966), 12. 11. For this dimension of the IDE, see Ted Robert Gurr, Why Men Rebel (Princeton: Princeton University Press, 1970), 100–101. One may also refer to Jose Ortega y Gasset’s notion that the dual evils of the modern world are egalitarianism and technicism. See The Revolt of the Masses (New York: Norton, 1957), 56–57. 12. Veblen (fn. 9), 208. 13. Fernand Braudel, Capitalism and Material Life, 1400–1800 (New York: Norton, 1976), 80– 82; H.I.Habakkuk and M.Postan, eds., Cambridge Economic History of Europe, Vol. VI (Cambridge: Cambridge University Press, 1965), 85. 14. For cereal yields in Europe before 1820, see Braudel (fn. 13),81;E.E.Rich and C. H.Wilson, eds., Cambridge Economic History of Europe, Vol. V (Cambridge: Cambridge University Press, 1977), 81; for the nineteenth century, see Alan S.Milward and S.B.Saul, The Development of the Economies of Continental Europe (Cambridge, MA: Harvard University Press, 1977), 234, 349, 379. For the twentieth century, see League of Nations, Agrarian Production in Continental Europe (Geneva, 1943), esp. 86–91. See also Table 3, below. 15. Braudel (fn. 13), 282–83; Habakkuk and Postan (fn. 13), 274–75. 16. Habakkuk and Postan (fn. 13), 313. 17. Neil McKendrick, John Brewer, and J.H.Plumb, Birth of a Consumer Society. The Commercialization of Eighteenth Century England (Bloomington: Indiana University Press, 1982), 29. 18. Ibid. 19. Ibid. 20. Ibid., 51. 21. Faujas de St.Fond, Voyage en Angleterre [Travels in England], Vol. 1,112, quoted in McKendrick et al. (fn. 17), 137. 22. Fé1ix F.Colson, De l’état présent de Moldavie et de la Valachie [The present state of Moldavia and Wallachia], Vol. I (Paris: Paulin, 1841), 210. See also Janos (fn. 9), 88. 23. Johann von Csaplovits, Eme Gemälde von Ungarn [A portrait of Hungary], Vol. I (Pest: Hartleben, 1829), 254. 24. Routh Trouton, Peasant Renaissance in Yugoslavia, 1900–1950 (London: Routledge, 1952), 71–72. 25. Claudio Veliz, The Centralist Tradition of Latin America (Princeton: Princeton University Press, 1980), 168, 172. 26. William P.Glade, The Latin American Economics A Study of Their Institutional Evolution (New York: American Books, 1969), 205. 27. Charles Anderson, Politics and Economic Change in Latin America (Princeton, NJ: Van Nostrand, 1967), 22. 28. McKendrick et al (fn. 17), 24. 29. There is considerable scholarly debate as to whether such price movements did or did not,


30.

31. 32.

33. 34. 35. 36. 37. 38. 39. 40.

41. 42. 43. 44. 45.

300

actually take place, and whether higher demand for manufactures, or disproportionate increases in primary production, were responsible. According to several sources though, the relative price of primary products on the world export market fell from 100 in 1872–76 to 68.2 in 1900, and rose again to 79.4 by 1913. See Milward and Saul (fn. 14), 485–87. For the argument concerning a secular decline of primary products, see League of Nations, Critique of Industrialization and Foreign Trade (Geneva, 1945) and United Nations, The Relative Price of Exports and Imports of Underdeveloped Countries (New York, 1949). For a refutation of the argument, see Paul Bairoch, The Economic Development of the Third World Since 1910, trans. Cynthia Postan (Berkeley: University of California Press, 1977), 111–26. For Latin America, see Veliz (fn. 25), 26, 120–24. For Spain, Richard Herr, The Eighteenth Century Revolution in Spain (Princeton: Princeton University Press, 1958), 120, 146–47. For Eastern Europe, see Jozo Tomasevich, Peasants, Politics, and Economic Change in Yugoslavia (Stanford, CA.Stanford University Press, 1955), 63. For Eastern Europe, see Andrew C.Janos, The Politics of Backwardness in Hungary, 1825–1945 (Princeton: Princeton University Press, 1982), 36–37, Robert Seton-Watson, History of the Roumanians (Cambridge: Cambridge University Press, 1934), 211; Stefan Zeletin, Burghezia româna: origina si rolul ei istoric [The Romanian bourgeoisie Its origins and role in history] (Bucharest: Editura Cultura Nationala, 1925), 47. Celso Furtado, The Economic Growth of Brazil (Berkeley: university of California Press, 1963), 97–98 These patterns apparently prevailed in the German lands between the Weser and Elbe, and in Austria, Tuscany, and central France, among other regions of the continent. See Max Weber, “Die Verhältnisse der Landarbeiter im ostelbischen Deutschland” [The condition of agricultural laborers in Germany east of the Elbe], Schriften des Vereins für Sozialpolitik 55 (1892), 795–99. See also Wallerstein (fn. 1), esp. 103–7. Weber (fn. 32). 797. The term first appears in his Neo-iobagia [Neoserfdom] (Bucharest: Minerva, 1912). Milward and Saul (fn. 14), 343–44. See Reinhard Bendix, Max Weber (Berkeley: University of California Press, 1977), 31–32. The poet Mihaí Eminescu, quoted in Zeletin (fn. 30), 142. Anderson (fn. 27), 224. For this term, see Fred Riggs, Administration in Developing Countries (Boston: Houghton Mifflin, 1966), 188–93. See, among others, Nicholas Spulder, The State and Economic Development in Eastern Europe (New York: Random House, 1966); Hans Schuster, Die Judenfrage in Rumanien [The Jewish question in Romania] (Leipzig: Felix Meiner, 1939); Trouton (fn. 24), 53–54; S.S.Bobchev, “Commercial and Economic Groups,” in Doreen Warriner, ed., Contrasts in Emerging Societies (Bloomington: Indiana University Press), 223, 227–28. Dusan J.Popovic, O Cincarima [About the Tsintsars] (Belgrade: Gregorivic, 1937), 83–84, 318–480. Peter Weber, Portugal: Räumliche Dimension und Abhängigkeit [Portugal: Spatial dimension and dependency] (Darmstadt: Wissenschaftliche Buchgesellschaft, 1982), 5. Gyula Szekfu, A mai Széchenyi [Topical Széchenyi] (Budapest: Magyar Szemle, 1934), 243. Vasa Cubrilovic, Istorija politickc misli u Srbiji XIX veka [History of nineteenth century Serbian political thought] (Belgrade: Prosveta, 1958), 288. Stanley G.Payne, A History of Spain and Portugal, Vol. II (Madison: University of Wisconsin Press, 1973), 518; Joseph Roucek, Balkan Politics (Stanford: Stanford University


301

Press), 58; and W.H.Harold Temperley, History of Serbia (London: Bell, 1917), 276. 46. John R.Lampe and Marvin R.Jackson, Balkan Economic History (Bloomington: Indiana University Press, 1982), 233, and Janos (fn. 30), 95. 47. In 1910, Portugal led the way with foreign indebtedness of 661.1 gold francs per capita. Next came Spain (498.6), Italy (368.5), Greece (328.0), Romania (299.6), Serbia (238.8), and Russia (186.8), in all of which per capita national income varied between 200 and 400 francs. By 1900, some 20 to 35 percent of budget revenue was used to service these debts, and new loans had to be contracted to pay the interest on the old. See Milward and Saul (fn. 14), 444 L.S.Stavrianos, The Balkans since 1453 (New York: Holt, Reinhart & Winston, 1958), 419, Lampe and Jackson (fn. 46), 160, 232–35. 48. Raymond Carr, Spain. 1808–1939 (Oxford: Clarendon Press, 1966), 211, 256. 49. Ibid., 256. 50. Ibid., esp. 210–56, 347–49 51. Payne (fn. 45). 555 52. A.H.Marques de Olivera, History of Portugal (New York: Columbia University Press 972), 53. 53. Janos (fn. 30), 98–101 212–13. 54. Seton-Watson (fn. 30), 354 55. Zeletin (fn. 30), 77 56. G.D.Nicolescu and A.Hermely, Deputatii nostri, 1895–99 [Our deputies, 1895–99] (Bucharest: Muller, 1899); G.D.Nicolescu, Parlamentul român [The Romanian parliament] (Bucharest: Socec, 1903); Corrin Petrescu, Albumul corpurilor leguitoare [The almanac of the legislative bodies] (Bucharest: Unirea, 1910). 57. Gale Stokes, Legitimacy through Liberalism. Vladimir Jovanovic and the Transformation of Serbian Politics (Seattle: University of Washington Press, 1975), 221. 58. Beaman A.Hulme, Stambuloff (London: Bliss, 1895), 63. 59. Slobodan Stoianovich, “Social Foundations of Balkan Politics,” in Charles and Barbara Jelavich, eds., The Balkans in Transition (Berkeley: University of California Press, 1963), 322. 60. Ibid., 322. 61. Stefan Zeletin, Neo-liberaismul. studii asupra istorei si politicei burghezei române [Neoliberalism: Studies in the history and politics of the Romanian bourgeoisie] (Bucharest: Pagini agrare si sociale, 1927), 10–12 62. Milward and Saul (fn. 14), 227–29. 63. Habakkuk and Postan (fn. 13), 461. 64. Gerschenkron (fn. 6), 12–13. 65. Ibid. 66. Ibid. (fn. 6), 19–20. 67. Peter N.Stearns, Paths to Authority: The Middle Class and the Industrial Labor Force in France, 1820–1948 (Urbana: University of Illinois Press, 1978), 54. 68. Charles S.Maier, Recasting Bourgeois Europe (Princeton: Princeton University Press, 69. It should be noted that the semiperiphery is primarily an economic concept, and that only a few of the continental governments presided over “pure” semiperipheral economies. The main culprit here was ethnic nationalism, the rise of which lumped together regions with different economic characters solely because their inhabitants spoke the same language. Thus the heartlands of both the German and the French economies were semiperipheral, but they coexisted uneasily with core-type economies in the Northwest and peripheral-type economies in the South and East of the two countries respectively. In Italy, a semiperipheral


70.

71. 72. 73. 74.

75. 76. 77. 78. 79. 80. 81.

302

North faced an undeveloped, peripheral South. In Czechoslovakia (after 1918), the more advanced, semiperipheral provinces were dominant, but had to coexist with the backward economy of Slovakia. This dichotomy was first suggested, although in a different context, in Peter 1. Katzenstein, Small States in World Markets: Industrial Policy in Europe (Ithaca, NY: Cornell University Press, 1985), 17–38. Maier (fn. 68), 353. See especially Karl Deutsch, “Social Mobilization and Political Development,” American Political Science Review 55 (September 1961), 493–502. G.D.H.Cole and Raymond Postgate, The British People, 1746–1946 (New York: Knopf, 1957), 373. After 1920, most governments of the European periphery extended suffrage to the rural population, although the electoral system in the villages remained as corrupt as before. At the same time, most governments made an attempt to appease the peasant populations by various redistribution schemes. These reforms were, however, largely symbolic. The only exception was Romania, a country that experienced a radical land reform. Even there, the weak peasant economy continued to serve as a resource base for subsidizing the industrialization of the country. See Henry Roberts, Rumania: Political Problems of an Agrarian State (New Haven: Yale University Press, 1951), esp. 22–88; also Zeletin (fn. 61). See Richard V.Burks, The Dynamics of Communism in Eastern Europe (Princeton: Princeton University Press, 1962), esp. 39–42. Désiré Laky, Statistiques des étudiants des universités en 1930 [Statistics of university students in 1930] (Budapest: Patria, 1932), 12–14. See also Janos (fn. 9); 92–93. See “Comments” in Carl J.Friedrich, ed., Totalitarianism (Cambridge, MA: Harvard University Press, 1954), 381. Fernando H.Cardoso and Enzo Faletto, Dependency and Development in Latin America, trans. Marjory M.Urquidi (Berkeley: University of California Press, 1979), 12. Veblen (fn 9), 163. See Charles Tilly, Big Structures. Large Processes, Huge Comparisons (New York: Russell Sage Foundation, 1985).


303

51 Social Democracy as a Historical Phenomenon* Adam Przeworski

* Source: Adam Przeworski, New Left Review, 1980, no. 122, pp. 27–58.

Not to repeat past mistakes: the sudden resurgence of a sympathetic interest in Social Democracy is a response to the urgent need to draw lessons from the history of the socialist movement. After several decades of analyses worthy of an ostrich, some rudimentary facts are finally being admitted. Social Democracy has been the prevalent manner of organization of workers under capitalism. Reformist parties have enjoyed the support of workers. Perhaps even more: for better or worse, Social Democracy is the only political force of the Left that can demonstrate an extensive record of reforms in favour of the workers. Any movement that seeks to transform historical conditions operates under these very conditions. The movement for socialism develops within capitalism and faces definite choices that arise from this very organization of society. These choices have been threefold: (I) whether to seek the advancement of socialism through the political institutions of the capitalist society or to confront the bourgeoisie directly, without any mediation; (2) whether to seek the agent of socialist transformation exclusively in the working class or to rely on multi- and even supra-class support; and (3) whether to seek improvements, reforms, within the confines of capitalism or to dedicate all efforts and energies to its complete transformation. Social democrats choose to participate, to seek supra-class alliances, and to struggle for reforms. Yet these decisions are not independent of each other. What is crucial to understand is the development of social democracy as a process: the manner in which the response to any one of these alternatives opens and closes the subsequent choices. For it may be that any movement that chooses to participate in bourgeois institutions, and specifically in elections, must seek support for socialist transformation beyond the working class and must struggle for all improvements that are possible in the short run without regard for ultimate consequences. Are the decisions to participate and the strategy of supra-class appeal inextricably connected? Is the orientation toward immediate reforms a necessary consequence of broadening the class base? Is an electoral party that would be based exclusively on working class support and dedicated exclusively to ultimate goals even possible? These are the kinds of questions that need to be answered if we are to draw lessons from the social democratic experience. What we need to know is the logic of choices faced by any movement for socialism within capitalist society: the historical possibilities that are opened and closed as each choice is made.

Social Democracy as a Historical Phenomenon

304

The Decision to Participate The reason why involvement in bourgeois politics has never ceased to evoke controversy is that the very act of ‘taking part’ in this system shapes the movement for socialism and its relation to workers as a class. The recurrent question is whether involvement in bourgeois institutions can result in socialism, or must strengthen the capitalist order. Is it possible for the socialist movement to find a passage between the ‘two reefs’ charted by Rosa Luxemburg: ‘abandonment of the mass character or abandonment of the final goals’?1 Participation in electoral politics is necessary if the movement for socialism is to find mass support among workers, yet this participation seems to obstruct the attainment of final goals. Working for today and working toward tomorrow appear as contrasting horns of a dilemma. Participation imprints a particular structure upon the organization of workers as a class. These effects of participation upon internal class relations have been best analysed by Luxemburg: ‘the division between political struggle and economic struggle and their separation is but an artificial product, even if historically understandable, of the parliamentary period. On the one hand, in the peaceful development, “normal” for the bourgeois society, the economic struggle is fractionalized, disaggregated into a multitude of political struggles limited to each firm, to each branch of production. On the other hand, the political struggle is conducted not by the masses through direct action, but, in conformity with the structure of the bourgeois state, in the representative fashion, by the pressure exercised upon the legislative body.’2 The first effect of ‘the structure of bourgeois state’ is thus that wage-earners are formed as a class in a number of independent and often competitive organizations, most frequently as trade-unions and political parties, but also as cooperatives, neighbourhood associations, clubs, etc. One characteristic feature of capitalist democracy is the individualization of class relations at the level of politics and ideology.3 People who are capitalists or wage-earners within the system of production all appear in politics as undifferentiated ‘individuals’ or ‘citizens’. Hence, even if a political party succeeds in forming a class on the terrain of political institutions, economic and political organizations never coincide. A multiplicity of unions and parties represent different interests and compete with each other. Moreover, while the class base of unions is confined to those who are more or less permanently employed, political parties which organize wage-earners must also mobilize people who are not members of unions. Hence there is a permanent tension between the narrower interests of unions and the broader interests represented by parties.4 The second effect is that relations within the class become structured as relations of representation. Parliament is a representative institution: it seats indi-viduals, not masses. A relation of representation is thus imposed upon the class by the very nature of capitalist democratic institutions. Masses do not act directly in defence of their interests; they delegate this defence. This is true of unions as much as of parties: the process of collective bargaining is as distant from the daily experience of the masses as elections. Leaders become representatives. Masses represented by leaders: this is the mode of


305

organization of the working class within capitalist institutions. In this manner participation demobilizes the masses. The organizational dilemma extends even further. The struggle for socialism inevitably results in the embourgeoisement of the socialist movement: this is the gist of Robert Michels’ classical analysis. The struggle requires organization; it demands a permanent apparatus, a salaried bureaucracy; it calls for the movement to engage in economic activities of its own. Hence socialist militants inevitably become bureaucrats, newspaper editors, managers of insurance companies, directors of funeral parlours, and even Parteibudiger—party bar keepers. All of these are petty bourgeois occupations. They impress,’ Michels concluded,‘…a markedly petty bourgeois stamp.’5 As a French dissident wrote recently, The working class is lost in administering its imaginary bastions. Comrades disguised as notables occupy themselves with municipal garbage dumps and school cafeterias. Or are these notables disguised as comrades? I no longer know.’6 A party that participates in elections must forsake some alternative tactics: this is the frequently diagnosed tactical dilemma. As long as workers did not have full political rights, no choice between insurrectionary and parliamentary tactics was necessary. Indeed, political rights could be conquered by those who did not have them only through extraparliamentary activities. César de Paepe, the founder of the Parti Socialiste Brabançon, wrote in 1877 that ‘in using our constitutional right and legal means at our disposal we do not renounce the right to revolution’.7 This statement was echoed frequently, notably by Engels in 1895. Alex Danielsson, a Swedish left-wing socialist, maintained in a more pragmatic vein that Social Democrats should not commit themselves to ‘a dogma regarding tactics that would bind the party to act according to the same routine under all circumstances’.8 That the mass strike should be used to achieve universal (and that meant male) suffrage was not questioned, and both the Belgian and Swedish parties led successful mass strikes that resulted in extensions of suffrage. Yet as soon as universal suffrage was obtained, the choice between the ‘legal’ and the ‘extra-parliamentary’ tactics had to be made. J.McGurk, the Chairman of the Labour Party, put it sharply in 1919: ‘We are either constitutionalists or we are not constitutionalists. If we are constitutionalists, if we believe in the efficacy of the political weapon (and we do, or why do we have a Labour Party?) then it is both unwise and undemocratic because we fail to get a majority at the polls to turn around and demand that we should substitute industrial action.’9 The turning point in the tactics of several parties occurred after the failures of general strikes organized around economic issues. While strikes oriented toward suffrage had been generally successful, the use of mass strikes for economic goals resulted in political disasters in Belgium in 1902,10 Sweden in 1909,11 France in 1920,12 Norway in 1921,13and Great Britain in 1926.14All these strikes were defeated; in the aftermath trade union membership was decimated and repressive legislation was passed. These common experiences of defeat and repression directed socialist parties toward an almost exclusive reliance on electoral tactics. Electoral participation was necessary to protect the movement from repression: this was the lesson drawn by socialist leaders. As Kautsky wrote already in 1891, The economic struggle demands political rights and these will not fall from heaven.’15 To win votes of people other than workers, particularly the petty bourgeosie, to form


306

alliances and coalitions, to administer the government in the interest of workers, a party cannot appear to be ‘irresponsible’, to give any indication of being less than wholehearted about its commitment to the rules and the limits of the parliamentary game. At times the party must even restrain its own followers from actions that would jeopardize electoral progress. Moreover, a party oriented toward partial improvements, a party in which leader-representatives lead a petty bourgeois life style, a party that for years has shied away from the streets cannot ‘pour through the hole in the trenches’, as Gramsci put it, even when this opening is forged by a crisis. The trouble about the revolutionary left in stable industrial societies,’ observed Eric Hobsbawn, ‘is not that its opportunities never came, but that the normal conditions in which it must operate prevent it from developing the movements likely to seize the rare moments when they are called upon to behave as revolutionaries…. Being a revolutionary in countries such as ours just happens to be difficult.’16 This dilemma became even more acute when democracy—representative democracy characteristic of bourgeois society—ceased to be merely a tactic and was embraced as the basic tenet of the future socialist society. Social democratic parties recognized in political democracy a value that transcends different forms of organization of production. Jean Jaurès claimed that: The triumph of socialism will not be a break with the French Revolution but the fulfillment of the French Revolution in new economic conditions.’17 Eduard Bernstein saw in socialism simply ‘democracy brought to its logical conclusion’,18 and ever since then the recurrent theme of social democracy has been precisely the notion of ‘extending’ the democratic principle from the political to the social, in effect principally economic, realm. Representative democracy became for social democrats simultaneously the means and the goal, the vehicle for socialism and the political form of the future socialist society, simultaneously the strategy and the programme, instrumental and prefigurative.19 This commitment made, however, even more crucial the question whether, as Harold Laski put it, capitalist democracy will ‘allow its electorate to stumble into socialism by the accident of the verdict at the polls’.20 The most important reservation toward an exclusive commitment to electoralism stemmed from the tenuous nature of bourgeois legality. Little is to be gained by interpreting and reinterpreting every word Marx wrote about bourgeois democracy for the simple reason that Marx himself, and the people who led the newly founded parties into electoral battles, were not quite certain what to expect of electoral competition. The main question—one which history never resolved because it cannot be resolved once and for all—was whether the bourgeoisie would respect its own legal order in case of an electoral triumph of socialism. If socialists were to use the institution of suffrage—established by the bourgeoisie in its struggle against absolutism—to win elections and to legislate a society toward socialism, would the bourgeoisie revert to illegal means to defend its interests ? This is what happened in France in 1851, and it seemed likely that it would happen again. But on several occasions Marx entertained the possibility that in England or in Holland counter-revolution would not occur if workers won the majority in the parliament. Thus, the essential question facing socialist parties was whether, as Hjalmar Branting posed it in 1886, ‘the upper class [would] respect popular will even when it demanded the abolition of its own privileges’.21 Sterky, the leader of the left wing of the Swedish party, was among those


307

who took a clearly negative view: ‘Suppose that…the working class could send a majority to the legislature; not even by doing this would it obtain power. One can be sure that the capitalist class would then take care not to continue along a parliamentary course but would instead resort to bayonets.’22This was eventually the position defended by Luxemburg in 1900.23 No one could be completely certain: according to Salvadori, Kautsky wobbled each time he approached this question.24 Austrian Socialists promised in their Linz programme of 1926 to ‘govern in strict accordance with the rules of the democratic state’, but they still felt compelled to warn that ‘should the bourgeoisie by boycotting revolutionary forces attempt to obstruct the social change which the labour movement in assuming power is pledged to carry out, then social democracy will be forced to employ dictatorial means to break such resistance.’25 The main doubt about electoral participation was whether revolution would not be necessary in any case, as August Bebel put it in 1905, ‘as a purely defensive measure, designed to safeguard the exercise of power legitimately acquired through the ballot.’26 Dictatorship of the proletariat, and revolutionary violence, might be necessary even if the party adhered strictly to its electoral commitment. Tactical dualism could not be easily foresaken.27 Hence social democrats faced a dilemma, dramatized by Gay in his biography of Bernstein. ‘Is democratic socialism, then, impossible? Or can it be achieved only if the party is willing to abandon the democratic method temporarily to attain power by violence in the hope that it may return to parliamentarism as soon as control is secure. Surely this second alternative contains tragic possibilities: a democratic movement that resorts to authoritarian methods to gain its objective may not remain a democratic movement for long. Still, the first alternative—to cling to democratic procedures under all circumstances—may doom the party to continual political impotence.’28 Social Democracy’s Forward March In spite of all the ambivalence, in spite of the pressure of short-term preoccupations, socialists entered into bourgeois politics to win elections, to obtain an overwhelming mandate for revolutionary transformations, and to legislate the society into socialism. This was their aim and this was their expectation. Electoral participation was based on the belief that democracy is not only necessary but that it is sufficient for reaching socialism. ‘If one thing is certain,’ Engels wrote in 1891 (a letter that was to meet with Lenin’s acute displeasure), ‘it is that our Party and the working class can only come to power under the form of democratic republic. This is even the specific form of the dictatorship of the proletariat.’29 Jaurès saw in democracy ‘the largest and most solid terrain on which the working class can stand…the bedrock that the reactionary bourgeoisie cannot dissolve without opening fissures in the earth and throwing itself into them’.30 Millerand was, as always, most incisive: To realize the immediate reforms capable of relieving the lot of the working class, and thus fitting it to win its own freedom, and to begin, as conditioned by the nature of things, the socialization of the means of production, it is necessary and sufficient for the socialist party to endeavour to capture the government through universal suffrage.’31 Socialists entered elections because they were concerned about immediate


308

improvements of workers’ conditions. Yet they also entered in order to bring about socialism. Was this divergence between cause and purpose a symptom of rationalization? Was the pathos of final goals just a form of self-deception? Such questions are best left for psychologists to resolve. But one thing is certain. Those who led socialist parties into electoral battles believed that dominant classes can be ‘beaten at their own game’. Socialists were deeply persuaded that they would win elections, that they would obtain for socialism the support of an overwhelming numerical majority. They put all of their hopes and their efforts into electoral competition because they were certain that electoral victory was within reach. Their strength was in numbers, and elections are an expression of numerical strength. Hence, universal suffrage seemed to guarantee socialist victory, if not immediately then certainly within the near future. Revolution would be made at the ballot box. Among the many expressions of this conviction is the striking apologia delivered by Engels in 1895: The German workers… showed the comrades in all countries how to make use of universal suffrage…. With the successful utilization of universal suffrage…an entirely new method of proletarian struggle came into operation, and this method quickly developed even further. It was found that state institutions, in which the rule of the bourgeoisie is organized, offer the working class still further opportunities to fight these very state institutions.’ And Engels offered a forecast: ‘If it [electoral progress] continues in this fashion, by the end of the century we shall…grow into the decisive power in the land, before which all other powers will have to bow, whether they like it or not.’32 The grounds for this conviction were both theoretical and practical. Already in The Communist Manifesto, Marx and Engels described socialism as the movement of ‘the immense majority’.33 In an 1850 article on The Chartists‘ in the New York Daily Tribune and then again in 1867 in the Polish emigre newspaper Glos Wolny, Marx repeated that ‘universal suffrage is the equivalent of political power for the working class of England, where the proletariat forms the large majority of the population….’34 Kautsky’s The Class Struggle, probably the most influential theoretical statement of the early socialist movement, maintained that the proletariat already constituted the largest class ‘in all civilized countries’.35 And even if the first electoral battles would not end in triumph, even if the proletariat was not yet the majority, electoral victory seemed only a matter of time because capitalism was swelling the ranks of the proletarians. The development of factory production and its corollary concentration of capital and land were leading rapidly to proletarianization of craftsmen, artisans, merchants, and small agricultural proprietors. Even ‘the physician, the lawyer, the priest, the poet, the man of science’ were being converted into proletarians, according to The Communist Manifesto. This growth of the number of people who sold their labour power for a wage was not accidental, temporary, or reversible: it was viewed as a necessary feature of capitalist development. Hence, it was just a question of time before almost everyone, ‘all but a handful of exploiters’, would become proletarians. Socialism would be in the interest of almost everyone, and the overwhelming majority of the people would electorally express their will for socialism. A young Swedish theoretician formulated this syllogism as follows in 1919: ‘The struggle for the state is political. Its outcome is therefore to a very great extent contingent upon the possibility open to society’s members—whose proletarianism has been brought about by the capitalist process—to exercise their proper influence on


309

political decision-making. If democracy is achieved, the growth of capitalism means a corresponding mobilization of voices against the capitalist system itself. Democracy therefore contains an automatically operative device that heightens the opposition to capitalism in proportion to the development of capitalism.36 Indeed, while those who eventually became communists saw in the Russian Revolution the proof that successful insurrection is always possible, for social democrats the necessity to rely on an insurrection of a minority meant only that conditions for socialism were not yet mature.37 While Branting, for example, shared Gramsci’s first reaction to the October Revolution38 when he maintained that ‘the whole developmental idea of socialism is discarded in Bolshevism’, he drew precisely the conclusion that socialists should wait until conditions ripen to the point that an overwhelming majority of the people would electorally express their will for socialist transformations.39 Since they were thoroughly persuaded that such conditions would be brought about by the development of capitalism, social democrats were not chagrined by electoral reversals, which were interpreted only to mean that the point had not yet arrived. Even when they had to relinquish control over the government, social democrats were not tempted to hasten the course of history. History spoke through the people, who spoke in elections, and no one doubted that history would make people express their will for socialism. These expectations, based on the conviction about the future course of history, were almost immediately vindicated by the electoral progress of socialist parties. The German party posed by Engels as the model to be followed—despite years of depression grew from 125,000 votes in 1871 to 312,000 in 1881, 1,427,000 in 1890, to 4,250,400 on the eve of World War I. Indeed, as soon as the Anti-Socialist laws were allowed to lapse, the SPD became in 1890 the largest party in Germany with 19.7 per cent of the vote. By 1912 their share of 34.8 per cent was more than twice that of the next largest party. No wonder that Bebel in 1905 could make ‘explicit the widely held assumption of his fellow socialists that the working class would continue to grow and that the party would one day embrace a majority of the population….’40 Several parties entered even more spectacularly into the competition for votes. In 1907, Finnish Social Democrats won the plurality, 37 per cent, in the first election under universal suffrage. The Austrian Social Democrats won 21.0 per cent after male franchise was made universal in 1907, 25.4 per cent in 1911, and the plurality of 40.8 per cent in 1919. The Belgian Parti Ouvrier won 13.2 per cent when the regime censitaire was abolished in 1894 and kept growing in jumps to win in 1925 the plurality of 39.4 per cent, a success which ‘stimulated them to hope that continuing industrialization would produce an increasing socialist workingclass electorate’.41 Even in those countries where the first steps were not equally dramatic, electoral progress seemed inexorable. In the religiously politicized Netherlands, socialism marched in big steps, from 3 per cent of the total vote in 1896 to 9.5 per cent, 11.2 per cent, 13.9 per cent and 18.5 per cent in 1913. The Danish party obtained 4.9 per cent in 1884, the first election it contested, only 3.5 per cent in 1889; from this moment on the party never failed to increase the share of the vote until 1935 when it won 46.1 per cent. There again, ‘there was a general expectation that as the sole party representing the labour movement, it would achieve power through an absolute majority of the electorate’.42 The Swedish party began meekly, offering candidates on joint lists with Liberals, it won 3.5 per cent in 1902, 9.5 per cent in 1905, 14.6 per cent in 1908, jumped


310

to 28.5 per cent in 1911 after suffrage was extended, increased its share to 30.1 per cent and 36.4 per cent in the two successive elections of 1914, and together with its left-wing off-shoot won the plurality of the vote, 39.1 per cent in 1917. The Norwegian Labour Party grew about 5 per cent in each election from 1897 when it obtained 0.6 per cent onward to 1915 when its share reached 32.1 per cent. Practice was confirming the theory. From election to election the forces of socialism were growing in strength. Each round was a new success. From a few thousand, at best, during the first difficult moments, socialists saw their electorate extend into millions. The progress seemed inexorable; the majority, and the mandate for socialism embodied therein, were only a matter of a few years, a couple of elections away. One more effort and humanity would be ushered into a new era by the overwhelming expression of popular will. ‘I am convinced,’ Bebel spoke at the Erfurt Congress, ‘that the fulfillment of our aims is so close that there are few in this hall who will not live to see the day.’43 Social Democracy and the Working Class The Socialist party was to be the working class organized. As Bergounioux and Manin observed, ‘workers’ autonomy outside politics or a political emancipation that would not be specifically workers’, such were the two tendencies at the moment when Marx and Engels contributed to the founding of the International Workingmen’s Association’.44 Marx’s decisive influence was a synthesis of these two positions; socialism as a movement of the working class in politics. The orientation Marx advocated was new: to organize a ‘party’ but one that would be distinctly of workers, independent from and opposed to all other classes. The organization of workers ‘into a class, and consequently into a political party’45 was necessary for workers to conquer political power and, in Marx’s view, it should not and would not affect the autonomy of the working class as a political force. The emancipation of the working class should be,’ in the celebrated phrase, ‘the task of the working class itself.’ We know why Marx expected workers to become the moving force for socialism by virtue of their position within the capitalist society; workers were simultaneously the class that was exploited in the specifically capitalist manner and the only class that had the capacity to organize production on its own once capitalist relations were abolished.46 Yet this emphasis on the ‘organic relation between socialism and the working class—the relation conceived of as one between the historical mission and the historical agent—does not explain by itself why socialists sought during the initial period to organize only workers and all the workers. The reasons for this privileged relation between socialist parties and the working class were more immediate and more practical than those that could be found in Marx’s theory of history. First, capitalism is a system in which workers compete with each other unless they are organized as a class. Similarity of class position does not necessarily result in solidarity since the interests which workers share are precisely those which put them in competition with one another, primarily as they bid down wages in quest of employment. Class interest is something attached to workers as a collectivity rather than as a collection of individuals, their ‘group’ rather than ‘serial’ interest.47 A general increase of wages is in


311

the interest of all workers, but it does not affect relations among them. Alternatively, a law establishing a minimal level of wages, extending compulsory education, advancing the age of retirement, or limiting working hours affects the relations among workers without being necessarily in the interest of each of them. Indeed, some workers would prefer to work beyond their normal retirement age even if they were excluding other workers from work; some people who do not find employment would be willing to be hired for less than the minimal wage even if it lowered the general level of wages; some would be willing to fulfill their historical mission of emancipating the entire society. In his Address to the Communist League in 1850 Marx emphasized that workers ‘must themselves do the utmost for their final victory by clarifying their minds as to what their class interests are, by taking their position as an independent party as soon as possible and by not allowing themselves to be seduced for a single moment by the hypocritical phrases of the democratic petty bourgeoisie into refraining from the independent organization of the party of the proletariat.’48 Rosenberg reports the tendency of German socialism in the 1860s to ‘isolate itself and to emphasize these qualities that differentiated it from all the groups and tendencies of the wealthy classes. At this stage the radical proletarian movement tended particularly to see the nobility and the peasants, the manufacturers and the intellectuals as “a uniform reactionary mass”.’49 The same was true of the first labour candidates who competed in the Paris election of 1863.50 The notion of ‘one single reactionary mass’ underlied the Gotha Programme of 1875 and reappeared in the Swedish programme of 1889.51 Still in 1891, when Engels was asked to comment on Kautsky’s draft of the Erfurt Programme, he objected to a reference to ‘the people in general’ by asking ‘who is that?’52 And with his typical eloquence, Jules Guesde argued in Lille in 1890: ‘The Revolution which is incumbent upon you is possible only to the extent that you will remain yourselves, class against class, not knowing and not wanting to know the divisions that may exist in the capitalist world.’53 Indeed, the initial difficulty which socialists faced was that workers were distrustful of any influences originating outside their class. Socialism seemed an abstract and an alien ideology in relation to daily experience. It was not apparent to workers that an improvement of their conditions required that the very system of wage labour must be abolished. Bergounioux and Manin report that according to a study of French workers at the beginning of the Third Republic there was a resistance among workers to the socialist message, an emphasis on the direct conflict between workers and employers, and a neglect of politics.54 In Belgium, a party bearing a socialist label, Parti socialiste beige, was founded in 1879 but had difficulty persuading workers’ associations to affiliate. According to Landauer workers were mistrustful of socialist propaganda, and de Paepe argued that ‘the word “socialist” frightens many workers’.55Thus was born in 1885 the Parti ouvrier beige: a workers’ party in place of a socialist one. In Great Britain, tradeunionists objected to, and until 1918 were successful in preventing, the Labour Party from admitting members of other classes on an individual basis. If socialists were to be successful, theirs had to be a workers’ party. In Sweden, the first local cells of the Social Democratic Party were in fact called Arbetarkommuner, Workers’ Communes.56 Socialists were anxious to emphasize the class character of the movement and were willing to make doctrinal compromises to implant socialism among workers.


312

The Dilemma of Proletarian Electoralism The majority which socialists expected to win in elections was to be formed by workers. The proletariat—acting upon its interests and conscious of its mission—was to be the social force precipitating society into socialism. But this proletariat was not, and never became, a numerical majority of voting members of any society. The prediction that the displaced members of the old middle classes would either become proletarians or join the army of the unemployed did not materialize. The old middle classes, particularly the independent agricultural proprietors, almost vanished as a group in most Western European countries, but their sons and daughters were more likely to find employment in an office or a store than in a factory. Moreover, while the proportion of the adult population engaged in any activity outside the household drastically fell in the course of capitalist development, those excluded from gainful activities did not become a reserve proletariat. Extended compulsory education, forced retirement, large standing armies, effective barriers to economic participation of women all had the effect of reducing entry into the proletariat.57 AS a result, from 1890 to 1980 the proletariat continued to be a minority of the population. In Belgium, the first European country to have built substantial industry, the proportion of workers did break the magic number of the majority when it reached 50.1 per cent in 1912. Since then it has declined systematically, down to 19.1 per cent in 1971. In Denmark, the proportion of workers in the electorate never exceeded 29 per cent. In Finland, it never surpassed 24 per cent. In France, this proportion declined from 39.4 per cent in 1893 to 24.8 per cent in 1968. In Germany, workers increased as a proportion of the electorate from 25.5 per cent in 1871, to 36.9 per cent in 1903, and since then has constituted about one third of the electorate. In Norway, workers constituted 33 per cent of the electorate in 1894 and their proportion peaked in 1900 at 34.1 per cent. In Sweden, the proportion of workers in the electorate grew from 28.9 per cent in 1908 to 40.4 per cent in 1952; then it declined to 38.5 per cent in 1964. The rules of the democratic game, while universal and at times fair, show no compassion. If a party is to govern alone, unburdened by the moderating influence of alliances and the debts of compromise, it must obtain some specific proportion of the vote, not much different from 50 per cent. Electoral institutions preceded the birth of parties which sought to use them as the vehicle toward socialism, and those institutions carried within themselves the fundamental rule which makes the victory of an isolated minority impossible. A party representing a class which has fewer members than the other classes combined cannot win electoral battles. The combination of minority status with majority rule constitutes the historical condition under which socialists have to act. This objective condition imposes upon socialist parties a choice: socialists must choose between a party homogeneous in its class appeal, but sentenced to perpetual electoral defeats, and a party that struggles for electoral success at the cost of diluting its class character. This choice is not between revolution and reform. There is no a priori reason, and no historical evidence, to suppose that an electoral class-pure party of workers would be any more revolutionary than a


313

party heterogeneous in its class base. Indeed, class-pure electoral parties of workers, of which the SPD during the Weimar period is probably the prime example,58 can be totally committed to the defence of particularistic interests of workers within the confines of capitalist society. Such class parties can easily become mere electoral interest groups, pressuring for a larger share of the national product without any concern for the manner in which it is produced. A pure party of workers who constituted a majority of the electorate would perhaps have maintained its ultimate commitment without a compromise, as socialists said they would when they saw the working class as majoritarian. But to continue as a minority party dedicated exclusively to ultimate goals, in a game in which one needs a majority—more, an overwhelming mandate—to realize these goals, would have been absurd. To gain electoral influence for whatever aims, from the ultimate to the most immediate working class parties must seek support from members of other classes. Given the minority status of workers within the class structure of capitalist societies, the decision to participate in elections thus alters the very logic of the problem of revolutionary transformation. The democratic system played a perverse trick on socialist intentions: the emancipation of the working class could not be the task of workers themselves if this emancipation was to be realized through elections. The only question left was whether a majority for socialism could be recruited by seeking electoral support beyond the working class. There is a peculiar tendency among contemporary observers, to see the strategy of appealing to a heterogeneous class base as a relatively recent effect of the ‘deradicalization’ of socialist movements. The German Mittelklasse Strategic is seen as the prototype of this new orientation and Kurt Schumacher as its architect.59 In this interpretation socialist parties began to enlist support from groups other than workers only after they have given up their socialist goals. This view is simply inaccurate. Socialists sought support beyond the working class as soon as the prospect of electoral victory became real and ever since they have continued to go back and forth between a search for allies and the emphasis on the working class. That triumphant forcecast made by Engels in 1895 which predicted that socialists would become a force before which ‘all powers will have to bow’ was conditional in his view upon the success of the party in ‘conquering the greater part of the middle strata of society, petty bourgeoisie and small peasants’. His advice to the French party—advice the French did not need since they were already heeding it60—was the same: recruit the small peasants. The Erfurt Programme of 1891 set the tone in which appeals to ‘the middle classes’ were couched: their interests ‘paralleled’ those of the proletariat; they were the ‘natural allies’ of the proletariat.61 Guesdists in France began to advocate alliances as soon as Guesde was elected to the Parliament in 1893.62 In Belgium, the first programme adopted in 1894 by the Parti ouvrier appealed to the lower middle class and the intelligentsia.63 in Sweden, a multi-class strategy was debated as early as 1889, and the party kept moving toward a heterogeneous class orientation until its full acceptance in 1920.64 The British Labour Party did defeat, in 1912, a proposal to open the membership, on an individual basis, to ‘managers, foremen, [and] persons engaged in commercial pursuits on their own account.’65 But in 1918, as it took a programmatic turn to the left, Labour opened its ranks to ‘workers by brain’. Indeed, in his polemic with Beer,


314

McKibbin interprets the very emphasis on socialism in the 1918 programme as an attempt to capture the ‘professional middle classes’.66 Revisionists everywhere asserted that workers were not a majority and that the party must seek support beyond the working class. Bernstein, Jaurès, and MacDonald came to this conclusion independently: once a party committed itself to electoral competition they had to embrace this conclusion. By 1915, Michels could already characterize social democratic strategy as follows: ‘For motives predominantly electoral, the party of the workers seeks support from the petty bourgeois elements of society, and this gives rise to more or less extensive reactions upon the party itself. The Labour Party becomes the party of the ‘people’. Its appeals are no longer addressed to the manual workers, but to “all producers”, to the “entire working population”, these phrases being applied to all the classes and all the strata of society except the idlers who live upon the income from investments.’67 The post-war orientation of several social democratic parties toward broadly understood middle strata is not a result of a new strategic posture but rather a reflection of the changing class structure of Western Europe. The proportion of the population engaged in agriculture declined during the twentieth century, more rapidly during the 1950s than during any of the preceding decades. The ‘new middle classes’ almost replaced the ‘old’ one numerically. Party strategies reflected, albeit with some lag, the numerical evolution of class structure. What is relatively new, therefore, is only the explicit indication of salaried employees as a pool of potential socialist support. It was Bernstein after all who introduced the notion of the Volkspartei, not Schumacher or Brandt. The search for allies is inherent to electoralism.

Dissolving the Class Appeal Once they decided to compete for the votes of ‘natural allies’, whether these were the old or the new middle classes, socialists were appealing to the overwhelming majority of the population. Branting’s estimate in 1889 that the ‘people’ constituted ninety-five per cent of the Swedish society was probably only slightly exaggerated, given his definition of ‘the people’.68 Seeking an equitable distribution of the burden of World War I debt, Labour and the New Social Order, a programmatic document of the party, asserted that ‘In this manner the Labour Party claims the support of four-fifths of the whole nation.’69 There is no reason to doubt that today the working class together with its allies comprise around eighty per cent of the population of France or of the United States.70 If to industrial workers we add white-collar employees, petty bourgeois, housewives, retirees, and students, almost no one is left to represent interests antagonistic to socialism. Exploiters remain but a handful: ‘the businessman with a tax-free expense account, the speculator with tax-free capital gains and the retiring company director with a tax-free redundancy payment,’ in the words of the 1959 Labour Party electoral manifesto.71 Yet social democratic parties have never obtained the votes of four-fifths of the electorate in any country. Only in a few instances have they won the support of one-half of the people who actually went to the polls. They are far from obtaining the votes of all whom they claim to represent. Moreover, they cannot even win the votes of all workers— the proletariat in the classical sense of the word. In several countries as many as one-third


315

of manual workers vote for bourgeois parties. In Belgium as many as one half of the workers do not vote socialist.72 in the United Kingdom, the Labour Party lost 49 per cent of the working class vote in the 1979 election. Social democrats appear condemned to minority status when they are a class party, and they seem equally relegated when they seek to be the party of the masses, of the entire nation. As a pure party of workers they cannot win the mandate for socialism, but as a party of the entire nation they have not won it either. Some of the reasons why no political party ever won a majority with a programme of socialist transformation are undoubtedly external to the electoral system. Yet social democratic parties face a purely electoral dilemma. Class shapes the political behaviour of individuals only as long as people who are workers are organized politically as workers. If political parties do not mobilize people qua workers but as ‘the masses’, ‘the people’, ‘consumers’, ‘taxpayers’, or simply ‘citizens’, then workers are less likely to identify themselves as class members and, eventually, less likely to vote as workers. By broadening their appeal to the ‘masses’, social democrats weaken the general salience of class as a determinant of the political behaviour of individuals. The strategies oriented toward broad electoral support have an effect not only upon the relation between workers and other classes but primarily within the class, upon the relations among workers. In order to be successful in electoral competition, social democratic parties must present themselves to different groups as an instrument for the realization of their immediate economic interests, immediate in the sense that these interests can be realized when the party is victorious in the forthcoming election. Supraclass alliances must be based on a convergence of immediate economic interests of the working class and of other groups. Social democrats must offer credits to the petty bourgeoisie, pensions to salaried employees, minimal wages to workers, protection to consumers, education to the young, family allowances to families. This convergence cannot be found in measures that strengthen the cohesion and combativeness of workers against other classes. When social democrats extend their appeal, they must promise to struggle not for objectives specific to workers as a collectivity—those that constitute the public goods for workers as a class—but only those which workers share as individuals with members of other classes. The common grounds can be found in a shift of the tax burden from indirect to direct taxation, in consumer protection laws, in spending on public transport, and the like. These are concerns which workers as individuals share with others who receive low incomes, who purchase consumer products, who travel to work. They are not interests of workers as a class but of the poor, of consumers, commuters, etc. None of this implies that the party no longer represents workers when it appeals to the masses. Although the convergence is never perfect and some interests of workers are often compromised, the party continues to represent those interests which workers as individuals share with other people. Hence social democratic parties oriented toward ‘the people’ continue to be parties of workers as individuals. But they cease to be the organization of workers as a class which disciplines individuals in their competition with each other by posing them against other classes. It is the very principle of class conflict— the conflict between internally cohesive collectivities that becomes compromised as parties of workers become parties of the masses.


316

Differentiation of class appeal, however, affects not only the organization of workers as a class. It has a fundamental effect on the form of political conflicts in capitalist societies since it reinstates a classless vision of politics. When social democratic parties become parties ‘of the entire nation’, they reinforce the vision of politics as a process of defining the collective welfare of ‘all members of the society’. Politics, once again, is defined on the dimension individual-nation, not in terms of class. This de-emphasis of class conflict in turn affects workers. As class identification becomes less salient, socialist parties lose their unique appeal to workers. Social democratic parties are no longer qualitatively different from other parties; class loyalty is no longer the strongest base of self-identification. One can no longer recall, as Vivian Gornick did of her childhood, that: ‘Before I knew I was Jewish or a girl I knew that I was a member of the working class.’73 Workers see society as composed of individuals; they view themselves as members of collectivities other than class; they behave politically on the basis of religious, ethnic, regional, or some other affinity. They become Catholics, Southerners, Francophones, or simply ‘citizens’. It is now clear that the dilemma comes back with a vengeance within the very system of electoral competition. The choice between class purity and broad support must be lived continually by social democratic parties because when they attempt to increase their electoral support beyond the working class these parties reduce their capacity to mobilize workers. This choice was not made once and for all by any party; nor does it represent a unidirectional evolution. Indeed, if there exists an electoral trade-off between appealing to the masses and recruiting workers, then strategic shifts are imperative from the purely electoral point of view. Histories of particular parties are replete with strategic reversals, with major changes of direction, controversies, schisms, and scissions. SPD returned to an emphasis on class in 1905; Swedish Social Democrats temporarily abandoned their attempt to become a multi-class party once in 1926, and then again in 1953; the Norwegian Labour Party emphasized its class orientation in 1918; German young socialists launched a serious attack on the Mittleklass Strategic a decade ago; conflicts between an ouvrierist and a multiclass tendency today wrench several parties. In terms of purely electoral considerations social democrats face a dilemma. They are forced to go back and forth between an emphasis on class and an appeal to the nation. They seem unable to win either way, and they behave the way rational people do when confronted with dilemmas: they bemoan and regret, change their strategies, and once again bemoan and regret. Social Democrats have not succeeded in turning elections into an instrument of socialist transformation. To be effective in elections they have to seek allies who would join workers under the socialist banner, yet at the same time they erode exactly that ideology which is the source of their strength among workers. They cannot remain a party of workers alone and yet they can never cease to be a workers’ party.

Reform and Revolution Socialists entered into elections with ultimate goals. The Hague Congress of the First International proclaimed that the ‘organization the proletariat into a political party is


317

necessary to ensure the victory of social revolution and its ultimate goal—the abolishment of classes’.74 The first Swedish programme specified that ‘Social Democracy differs from other parties in that it aspires to completely transform the economic organization of bourgeois society and bring about the social liberation of the working class….’75 Even the most reformist among revisionists, Millerand, admonished that ‘whoever does not admit the necessary and progressive replacement of capitalist property by social property is not a socialist’.76 These were the goals that were to be reached through legislation, upon a mandate of an electorally expressed majority, as the will of universal suffrage. Socialists were going to abolish exploitation, to destroy the division of society into classes, to remove all economic and political inequalities, to end the wastefulness and anarchy of capitalist production, to eradicate all sources of injustice and prejudice. They were going to emancipate not only workers but humanity, to build a society based on cooperation, to rationally orient energies and resources toward satisfaction of human needs, to create social conditions for an unlimited development of the personality. Rationality, justice and freedom were the guiding goals of the social democratic movement. These were ultimate goals: they could not be realized immediately, for economic as well as political reasons. And social democrats were unwilling to wait for the day when these aims could finally be accomplished. They claimed to represent the interests of workers and of other groups not only in the future but as well within ‘present-day’, that is, capitalist society. The Parti Socialiste Française, led by Jaurès, proclaimed at its Tours Congress of 1902: The Socialist Party, rejecting the policy of all or nothing, has a programme of reforms whose realization it pursues forthwith’, and listed 54 specific demands concerning democratization, secularization, organization of justice, family, education, taxation, protection of labour, social insurance, nationalization of industries, and foreign policy ,77 The first programme of the Swedish Social Democrats in 1897 demanded direct taxation, development of state and municipal productive activities, public credit including direct state control of credit for farmers, legislation concerning work conditions, old age, sickness, and accident insurance, legal equality, and freedoms of organization, assembly, speech, and press.78 This orientation toward immediate improvements was never seen by its architects as a departure from ultimate goals. Since socialism was thought to be inevitable, there would be no reason why immediate measures should not be advocated by socialist parties: there was no danger, not even a possibility, that such measures could prevent the advent of the inescapable. As Kautsky put it, ‘it would be a profound error to imagine that such reforms could delay the social revolution’.79 Ultimate goals were going to be realized because History was on the side of socialism. Revisionists within the movement were, if anything, even more deterministic than those who advocated insurrectionary tactics. Millerand argued, for example, in his Saint-Mandé speech, that: ‘Men do not and will not set up collectivism; it is setting itself up daily; it is, if I may be allowed the phrase, being secreted by the capitalist regime.’80 Even when social democratic movements left the protection of history to rediscover a justification of socialism in ethical values, no dilemma appeared in the consciousness of socialist leaders. Bernstein’s famous renunciation of final goals did not imply that they would remain unfulfilled, but only that the way to realize them was to concentrate on


318

proximate aims. Jaurès, speaking about the conquest of political power by workers, provided the classical image: ‘I do not believe, either, that there will necessarily be an abrupt leap, the crossing of the abyss; perhaps we shall be aware of having entered the zone of the Socialistic State as navigators are aware of having crossed the line of a hemisphere—not that they have been able to see as they crossed a cord stretched over the ocean warning them of their passage, but that little by little they have been led into a new hemisphere by the progress of their ship.’81 Indeed, for social democrats immediate reforms constitute ‘steps’ in the sense that gradually they accumulate toward a complete restructuring of society. Anticipating Bernstein’s argumentation, George von Vollmar, the leader of the Bavarian wing of the SPD, declared at the Erfurt Congress: ‘Beside the general or ultimate goal we see a nearer aim: the advancement of the most immediate needs of the people. For me, the achievement of the most immediate demands is the main thing, not only because they are of great propagandist value and serve to enlist the masses, but also because, in my opinion, this gradual progress, this gradual socialization, is the method strongly indicated for a progressive transition.’82 Reform and revolution do not require a choice within the social democratic view of the world. To bring about ‘social revolution’—the phrase which before 1917 connoted transformation of social relations but not necessarily an insurrection—it is sufficient to follow the path of reforms. Reforms are thought to be cumulative and irreversible: there was nothing strange in Jaurès’ argument that: ‘Precisely because it is a party of revolution…the Socialist Party is the most actively reformist….’83 The more reforms, the faster they are introduced, the nearer the social revolution, the sooner the socialist ship would sail into the new world. And even when times are not auspicious for new steps to be made, even when political or economic circumstances require that reforms be postponed, eventually each new reform would build upon the past accomplishments. Mitigating the effects of capitalism and transforming it piece by piece would eventually lead to a complete restructing of society. Reviewing Miliband’s (1969) book, Benjamin Barber best expressed this perspective: ‘surely at some point mitigation becomes transformation, attenuation becomes abolition; at some point capitalism’s “concessions” annihilate capitalism…. This is not to say that such a point has been reached, only that there must be such a point.84 Welfare Displaces Socialization The ‘social revolution’ envisioned by social democrats was necessary because capitalism was irrational and unjust. And the fundamental cause of this inefficiency and inequity was private property of the means of production. While private property was occasionally seen as the source of most disparate evils—from prostitution and alcoholism to wars—it was always held directly responsible for the irrationality of the capitalist system and for the injustice and poverty that it generated. Already in Socialism: Utopian and Scientific, one of the most important theoretical sources of the socialist movement, Engels emphasized that the increasing rationality of capitalist production within each firm is accompanied, and must be accompanied, by the chaos and anarchy of production at the societal scale. ‘The contradiction between


319

socialized production and capitalist appropriation,’ Engels wrote, ‘now presents itself as an antagonism between the organization of production in the individual workshop and the anarchy of production in society generally.’85 Speaking in 1920, Branting repeated that: ‘In the basic premises of the present social order there are no satisfactory guarantees either that production as an entity is given the most rational orientation possible, or that profit in the various branches is used in the way that is best from the national economic and social point of view.’86 The second effect of private property is the unjust distribution of material rewards which it generates. The economic case for socialism,’ wrote a Labour Party theoretician, ‘is largely based on the inability of capitalism to bring about any equitable or even practicable distribution of commodities in an age of mechanization and massproduction.’87 Even the most decisive break with the Marxist tradition, the SPD’S Bad Godesberg programme of 1959, maintained that the ‘Market economy does not assure of itself a just distribution of income and property’. Given this analysis, socialization or nationalization of the means of production was the principal method of realizing socialist goals and hence the first task to be accomplished by social democrats after the conquest of power. ‘Social revolution,’ writes Tingsten ‘was always understood to mean systematic, deliberate socialization under the leadership of the Social Democratic working class.’88 Socialization or nationalization—a terminological ambiguity which was significant—was the manner by which socialist revolution would be realized. Until World War I, as socialist parties concentrated their efforts on winning suffrage and organizing workers as a class, little if any concrete thought was devoted to the means by which socialization was to be accomplished. The very possibility of actually being in a position to pursue a programme of socialization caught all socialist parties by surprise when the war destroyed the established order, unleashed spontaneous movements of factory occupations, and opened the doors to government participation. Indeed, the wave of factory occupations which occurred in Austria, Germany, Finland, Italy, and Sweden appeared to the established socialist parties and trade-unions as almost as much a threat to their own authority and organization as to the capitalist order.89 As these spontaneous movements were repressed or exhausted, the logic of parliamentarism re-established its grip on the social democratic movement. Nationalization efforts turned out to be so similar in several countries that their story can be summarized briefly. The issue of socialization was immediately placed on the agenda of social democratic parties in Austria, Finland, Germany, Great Britain, Holland, Italy, and Sweden and of the CGT in France. In several countries, notably Germany, Great Britain, and Sweden, ‘socialization commissions’ were established by respective parliaments, while in France Léon Blum introduced in the Chamber a bill to nationalize the railway industry. The commissions were supposed to prepare detailed programmes of socialization—in some cases for all basic industries and in others for specific ones, typically coal. The British commission finished its career quickly as Lloyd George simply ignored its recom-mendations; in Germany the issue of coal nationalization lingered after the resignation of the first commission; and in Sweden the socialization committee worked 16 years, spending most of its time studying similar efforts elsewhere, and expired without making any recommendations. Although social democrats formed or


320

entered governments in several countries, the global result of these first attempts at socialization was null: with the exception of the French armament industry in 1936, not a single company was nationalized in Western Europe by a social democratic government during the entire inter-war period. How did it happen that the movement that aimed to revolutionize society by changing the very base of its productive organization ended the period of integration into the political institutions of capitalism without even touching its fundaments? When Marx described in 1850 the anatomy of capitalist democracy, he was certain that, unless withdrawn, universal suffrage would lead from ‘political to social emancipation’; that once endowed with political rights, workers would proceed immediately to destroy the ‘social power’ of capitalists by socializing the means of production.90 Still in 1928, Wigforss saw this outcome as inevitable: The universal suffrage is incompatible with a society divided into a small class of owners and a large class of unpropertied. Either the rich and the propertied will take away universal suffrage, or the poor, with the help of their right to vote, will procure for themselves a part of the accumulated riches.91 And yet while social democrats held power in Austria, Belgium, Denmark Finland, France, Germany, Great Britain, Norway, and Sweden, riches remained nearly intact, and certainly private property in the means of production was not disturbed. One can cite a number of reasons. Not negligible was the theoretical ambiguity of the very project of the ‘expropriation of expropriators’. One difficulty lay in that ambiguous relation between ‘socialization’—the turning over of industries to their employees—and ‘nationalization’—their general direction by the state, on the one hand, as Korsch,92 Wigforss,93 and others pointed out, direct control of particular firms by the immediate producers would not remove the antagonism between producers and consumers, that is, workers in other firms. On the other hand, transfer to centralized control of the state would have the effect of replacing the private authority of capital by the bureaucratic authority of the government, and the Soviet example loomed largely as a negative one. The ‘gestionnaire’ tendency dominated in Germany, where the principle was even incorporated into the Constitution, and Sweden; the ‘planiste’ tendency found its most important articulation in Belgium and France under the influence of Henri de Man. A veritable wave of constitution writing ensued in the immediate aftermath of World War I: Otto Bauer in Austria (1919), Karl Kautsky in Germany (1925), G.D.H. Cole in Great Britain (1919), Henri de Man in Belgium—all rushed to devise some way of combining rationality at the level of the society as a whole with the control of the immediate producers over their own activities. Yet this burst of theoretical activity came rather belatedly in relation to the demands of practical politics. The fact, frequently admitted by social democratic politicians, was that they did not know how to proceed to the realization of their programme. The choice of industries which were to be nationalized, methods of financing, techniques of management, and the mutual relations among sectors turned out to be technical problems for which social democrats were unprepared. Hence they formed study commissions and waited. Nevertheless, the cause of the social democratic inertia was much more profound than the ambiguity of their plans. Socialists never won a sufficient number of votes to obtain a parliamentary majority and hence to be able to legislate anything without support, or at


321

least consent, of other parties. Remarkably, and quite to their surprise, socialist parties in several countries were invited to take office as minority governments or to enter governments as members of multi-party coalitions. And the question of what to do as a minority government presented itself as the following choice: either the party would pursue its socialist objectives and be promptly defeated or it would behave like any other party, administering the system and introducing only those few reforms for which it could obtain a parliamentary majority. Each strategy was viewed in terms of its long-term effects. Proponents of the maximalist strategy argued that the party would educate the electorate about its socialist program and would expose the reactionary character of the bourgeois parties. They claimed that the people would then return the party to office with a majority and the mandate to pursue its socialist programme. Only in Norway was this strategy adopted; the government lasted three days in 1928; and the party was returned to office four years later only after it had moderated its socialist objectives. Proponents of a minimal programme argued that the most important task a party could accomplish was to demonstrate that it was ‘fit to govern’, that it was a governmental party. ‘We are not going to under take of office to prepare for a General Election,’ said MacDonald in 1924 ‘we are going to take office in order to work.’94 Their expectation, in turn, rested on the belief that reforms were irreversible and cumulative. As Lyman put it, ‘Gradualists imagined that socialism could be achieved by instalments, each instalment being accepted with no more serious obstruction on the part of the Conservatives than Labour opposition generally gave to Tory governments. Each instalment would then remain, unharmed by interludes of Tory rule, and ready to serve as the foundation on which the next Labour government would resume construction of the socialist commonwealth.’95 Hence the party would come into office, introduce those reforms and only those reforms for which it could muster the support of a parliamentary majority, and then leave to return when a new mandate issued from the electorate. ‘We hope to continue only as long in office, but certainly as long in office, as will enable us to do some good work that will remove many obstacles which would have hampered future governments if they found the problems that we know how to face’: this was the intention of the Labour Party in 1924, according to MacDonald.96 Hence Blum introduced a distinction between the ‘exercise of power’ and the ‘conquest of power’: as a minority socialists could only exercise it, but they should exercise it in such a way that would eventually lead to its conquest.97 If socialists could not pursue an immediate programme of nationalization, what could they do in the meantime ? They could and did pursue ad hoc measures designed to improve the conditions of workers: develop housing programmes, institute some protection from unemployment, introduce minimum wage laws, income and inheritance taxes, old age pensions. Such measures, although they favoured workers, were neither politically unfeasible nor economically shocking—they continued the tradition of reforms associated with Bismarck, Disraeli, and Giolitti. These measures neither modified the structure of the economy nor the political balance of forces. The fact is that until the 1930s social democrats did not have any kind of economic policy of their own. The economic theory of the Left was the theory that criticized capitalism, claimed the superiority of socialism, and led to a programme of


322

nationalization of the means of production. Once this programme was suspended—it was not yet abandoned—no socialist economic policy was left.98 Socialists behaved like all other parties: with some distributional bias toward their constituency but full of respect for the golden principles of the balanced budget, deflationary anti-crisis policies, gold standard, and so on. Of Blum it is said that he ‘could envisage no intermediate stage between pure doctrinaire socialism and the free play of capitalism…,’99 and it seems that neither could anyone else. The only known theory of reforms was that which called for nationalization; no other coherent alternative existed. Such an alternative did emerge in response to the Great Depression. In Sweden, Norway, and to a lesser extent France, socialist governments responded to unemployment with a series of anti-cyclical policies that broke the existing economic orthodoxy. It remains a matter of controversy whether the Swedish policies were developed autonomously, from Marx via Wicksell, or were an application of the already circulating ideas of Keynes.100 The fact is that social democrats everywhere soon discovered in Keynes’ ideas, particularly after the appearance of his General Theory, something they urgently needed: a distinct policy for administering capitalist economies. The Keynesian revolution—and this is what it was—provided social democrats with a goal and hence the justification of their governmental role and simultaneously transformed the ideological significance of distributive policies that favoured the working class. From the passive victim of economic cycles, the state became transformed almost overnight into an institution by which society could regulate crises to maintain full employment. Describing the policies of the Swedish government of 1952, Gustav Möller, the architect of the unemployment programme, emphasized that previously unemployment relief was a ‘system meant only to supply bare necessities to the unemployed, and did not have the purpose of counteracting the depression…. Economic cycles, it was said, follow natural economic laws, and governmental interference with them is, by and large, purposeless and, from a financial point of view, even dangerous in the long run’.101 Both Möller and Wigforss102 described how the Swedish Social Democrats discovered that unemployment could be reduced and the economy invigorated if the state followed anti-cyclical policies, allowing deficits to finance productive public works during depressions and paying back the debts during periods of expansion. Society is not helpless against the whims of the capitalist market, the economy can be controlled, and the welfare of citizens can be continually enhanced by the active role of the state: this was the new discovery of social democrats. And this was not all: Keynesianism was not only a theory that justified socialist participation in government but, even more fortuitously from the social democratic point of view, it was a theory that suddenly granted a universalistic status to the interests of workers. Earlier, all demands for increased consumption were viewed as inimical to the national interest: higher wages meant lower profits and hence a reduced opportunity for investment and future development. The only conceivable response to crises was to cut costs of production, that is, wages. This was still the view of the Labour Party in 1929. But in the logic of Keynes’ theory higher wages, particularly if the wage fund was increased by raising employment rather than the wage rate (which did not rise in Sweden until 1936), meant an increase of aggregate demand, which implied increased expectations of profit, increased investment, and hence economic stimulation. The


323

significance of increasing wages changed from being viewed as an impediment to national economic development to being its stimulus. Corporatist defence of the interests of workers, a policy social democrats pursued during the ‘twenties, and the electoral strategy toward the ‘people’ now found ideological justification in a technical economic theory. The Keynesian turn soon led social democrats to develop a full-fledged ideology of the ‘welfare state’.103 Social democrats defined their role as that of modifying the play of the market forces, in effect abandoning the project of nationalization altogether. The successful application of Keynesian instruments was seen as the demonstration that nationalization—full of problems and uncertainties as it proved to be—was not only impossible to achieve in a parliamentary way but was simply unnecessary. Keynes himself wrote: ‘It is not the ownership of the instruments of production which it is important for the state to assume. If the state is able to determine the aggregate amount of resources devoted to augmenting the instruments and the basic rate of reward to those who own them, it will have accomplished all that is necessary.’104 AS Wigforss argued further, state ownership of particular industries would only result in the socialist government being forced to behave as a capitalist firm, subject to ‘the chaos of the market’, while by indirect control the state could rationalize the economy as a whole and orient it toward the general welfare.105 The theoretical underpinning of this new perspective was the distinction between the concept of property as the authority to manage and property as legal possession. Already Bernstein claimed that ‘the basic issue of socialization is that we place production, economic life, under the control of the public weal.’106 Instead of direct ownership, the state could achieve all the socialist goals by influencing private industry to behave in the general interest. The essence of nationalization,’ wrote de Man in 1934 ‘is less the transfer of property than the transfer of authority….’107 If the state could regulate private industry when necessary and if it could mitigate the effects of the free play of market forces, then direct ownership would be unnecessary and inadvisable: this became the motto of social democracy in the aftermath of the Keynesian revolution. In sum, unable as minority governments to pursue the social programme, in the midthirties, social democracy found a distinct economic policy which justified its governmental role, which specified a number of intermediate reforms that could be successively accomplished within the confines of capitalism, and which provided in several countries a successful electoral platform. Caught in the twenties in an all-ornothing position, social democrats discovered a new path to reform by abandoning the project of nationalization for that of general welfare.

The Abandonment of Reformism The abandonment of programmatic nationalization of the means of production did not imply that the state would never become engaged in economic activities. In contemporary Western European countries between 5 and 20 per cent of gross product is now being produced by enterprises of which the state is in some form a complete owner.108 The paths by which this ‘public sector’ developed are too varied to recount here. In Italy and


324

Spain the public sector constitutes mainly a fascist legacy; in Austria it consists predominantly of confiscated German properties; in Great Britain and France a wave of nationalizations followed World War II. Characteristically, state enterprises are limited to credit institutions, coal, iron and steel, energy production and distribution, transport, and communication. Outside these sectors only those companies which are threatened with bankruptcy and hence a reduction of employment pass into public hands. Instances in which the state would be engaged in producing and selling final-demand goods are extremely rare; they seem to be limited to the automobile industry. The state engages in those economic activities which are necessary for the economy as a whole and sells its products and services mainly to private firms. Hence, the state does not compete with private capital but rather provides the inputs necessary for the profitable functioning of the economy as a whole. This division between the state and the market has been recently enshrined in the ‘public goods theory of the state’.109 This theory assumes that the capitalist market is a natural form of economic activity; the existence of the market and its laws are taken as given. The role of the state is supposed to be limited to the provision of so-called ‘public goods’: those that are indivisible and which must be supplied to everyone if they are supplied to anyone. It is proper for the state to construct public roads or to train the labour force: rational private entrepreneurs will not provide such goods since they cannot prevent people from using roads or from selling their newly acquired skills to competitors. The role of the state is thus supposed to be limited to those activities that are unprofitable for private entrepreneurs yet needed for the economy as a whole. Hence, the structure of the capitalist systems built by social democrats turned out to be the following: (1) the state operates those activities which are unprofitable for private firms but necessary for the economy as a whole; (2) the state regulates, particularly by pursuing anti-cyclical policies, the operation of the private sector; and (3) the state mitigates, through welfare measures, the distributional effects of the operation of the market. The regulatory activities of the state are based on the belief that private capitalists can be induced to allocate resources in a manner desired by citizens and expressed at the polls. The basic notion is that in a capitalistic democracy resources are allocated by two mechanisms: the ‘market’, in which the weight of preferences of decision-makers is proportional to the resources they control, and the state, in which the weight of preferences is distributed equally to persons qua citizens. The essence of contemporary social democracy is the conviction that the market can be directed to those allocations of any good, public or private, that are preferred by citizens and that by gradually rationalizing the economy the state can turn capitalists into private functionaries of the public without altering the juridical status of private property. Having made the commitment to maintain private property of the means of production, to assure efficiency, and to mitigate distributional effects, social democracy ceased to be a reformist movement.110 Reformism always meant a gradual progression toward structural transformations; reformism was traditionally justified by the belie that reforms are cumulative, that they constitute steps, that they lead in some direction. The current policy of social democrats by its very logic no longer permits the cumulation of reforms. The abandonment of reformism is a direct consequence of those reforms that have been


325

accomplished. Since the state is engaged almost exclusively in those activities which are unprofitable from the private point of view, it is deprived of financial resources needed to continue the process of nationalization. Having nationalized deficitary sectors social democrats undermined their very capacity to gradually extend the public realm. At the same time, having strengthened the market, social democrats perpetuate the need to mitigate the distributional effect of it operation. Welfare reforms do not even have to be ‘undone’ by bourgeois governments. It is sufficient that the operation of the market is left to itself for any length of time and inequalities increase, unemployment fluctuates, shifts of demand for labour leave new groups exposed to impoverishment, etc. As Martin put it with regard to Great Britain, ‘The “basic structure of the full employment welfare state” did not prove as durable as Crosland’s analysis would lead us to expect. However, this was not because Conservative governments between 1951 and 1964 proceeded to dismantle it…. All that was necessary to undermine the full employment welfare state was for the Conservative governments simply to do nothing to counteract these processes.’111 Mitigation does not become transformation: indeed, without transformation the need to mitigate becomes eternal. Social democrats find themselves in the situation which Marx attributed to Louis Bonaparte: their policies seem contradictory since they are forced at the same time to strengthen the productive power of capital and to counteract its effects. The final result of this orientation is that social democrats again find themselves without a distinct alternative of their own as they face a crisis of the international system. When in office they are forced to behave like any other party, relying on deflationary, cost-cutting measures to ensure private profitability and the capacity to invest. Measures oriented to increase democracy at the work-place—the recent rediscovery of social democrats112—not surprisingly echo the posture of the movement in the 1920s, another period when the Left lacked any macro-economic approach of its own.

Economic Bases of Class Compromise As soon as social democrats formed governments after World War I, they discovered that their concern with justice was not immediately compatible with the goal of increased productivity. In Wigforss’ words, ‘Because Social Democracy works for a more equal and more just distribution of property and incomes, it must never forget that one must produce before one has something to distribute.113 The concern for restoring and extending industrial productive capacity quickly came to dominate the first discussions of socialization of industry in Germany and Sweden.114Certainly a just distribution of poverty was not the socialist promise, and to enhance affluence social democrats had to focus their efforts on increasing productivity. But without nationalization of the means of production, increases of productivity require profitability of private enterprise. As long as the process of accumulation is private, the entire society is dependent upon maintaining private profits and upon the actions of capitalists allocating these profits. Hence the efficacy of social democrats—as of any other party—in regulating the economy and mitigating the social effects depends upon the profitability of the private sector and the willingness of capitalists to cooperate.


326

The very capacity of social democrats to regulate the economy depends upon the profits of capital. This is the structural barrier which cannot be broken: the limit of any policy is that investment and thus profits must be protected in the long run. The basic compromise of social democrats with private capital is thus an expression of the very structure of capitalist society. Once private property of the means of production was left intact, it became in the interest of wage-earners that capitalists appropriate profits. As Chancellor Schmidt put it, The profits of enterprises today are the investments of tomorrow, and the investments of tomorrow are the employment of the day after’ (Le Monde, July 6, 1976). This expectation—that current profits would be transformed into future improvements of material conditions of wage-earners—became the foundation of the social democratic consent to capitalism.115 Social democrats consent to the right of capitalists to withhold a part of societal product because the profits appropriated by capital are expected to be saved, invested, transformed into productive capacity, and partly distributed as gains to other groups. Social democrats protect profits from revindicative demands of the masses because radical redistributive policies are not in the interest of wage-earners. This is why social democrats trade-off the abolition of private property of the means of production for cooperation of capitalists in increasing productivity and distributing its gains. This is why social democrats not only attempt to produce capitalism but struggle to improve it even against the resistance of capitalists. Nationalization of the means of production has turned out to be electorally unfeasible; radical redistributive polices result in economic crises which are not in the interest of wage-earners; and general affluence can be increased only if capitalists are made to cooperate and wage-earners are continually disciplined to wait.

Crisis and the Workers Government Social democrats will not lead European societies into socialism. Even if workers would prefer to live under socialism, the process of transition must lead to a crisis before socialism could be organized. To reach higher peaks one must traverse a valley, and this descent will not be completed under democratic conditions. Suppose that social democrats win elections and attempt to use their position for a democratic transition to socialism. Given the social structure of capitalist societies, such an electoral victory is possible only if support can be obtained from several groups: industrial workers, non-manual employees, petite bourgeoisie, farmers, housewives, retired people, and/or students. Hence pressures for a significant improvement of material conditions erupt from several groups. Wages, particularly the minimal or ‘vital’ wages (sueldo vital in Chile, SMIC in France), must be increased. Unemployment must be reduced. Transfers, particularly family allowances, must be raised. Credit for small enterprises and farms must become cheaper and available at a higher risk. These demands can be financed by (1) a redistribution of personal incomes (both through direct taxation and a reduction of wage differentials), (2) increased utilization of latent capacity, (3) spending of foreign reserves or borrowing, and/or (4) reduction of the rate of profit.116 The sum of the first three sources will not be sufficient to satisfy the demands.


327

Redistribution of top incomes does not have much of a quantitative effect, and it cannot reach too far down without threatening the electoral support of salaried employees. Forced to pay higher wages, and to keep employment beyond the efficient level, capitalists can respond only by increasing the prices of wage goods. Inflation is also fueled by balance of payment difficulties resulting from the necessity to import wage goods and from speculatory pressures. Hence, either an inflationary dynamic sets into motion or, if prices are controlled, scarcities appear, a black market is organized, and so on. Eventually nominal wage increases become eroded, as they were in France in 1936,117 in Chile and in Portugal. Under normal circumstances it can be expected that the increase of aggregate demand should stimulate investment and employment. Redistributional measures, even if they include inorganic emission, are usually justified not only by appeals to justice but also to efficiency. As lower incomes increase, so does the demand for wage goods. The utilization of latent capacity and foreign reserves are seen as a cushion that would protect prices from increased demand during the short period before investment picks up and eventually when supply rises. It is expected that profits from a larger volume of sales will be rein vested and thus the economy will be stimulated to develop at a faster pace. This was, for example, the Vuskovic programme in Chile—not at all unreasonable under normal circumstances. Such a program cannot be successful, however, when economic demands grow spontaneously and when they are accompanied by structural transformations. Wage demands are likely to become confiscatory under such circumstances, and capitalists expect that these demands will be enforced, or at least condoned, by the government. Measures of nationalization, distribution of land, and monopolization of credit and foreign exchange by the state threaten the very institution of private profit. Under such circumstances, rational private capitalists will not invest. A transition to socialism must therefore generate an economic crisis. Investment falls sharply, prices increase, nominal wage gains become eroded, and eventually output falls, demand slackens, unemployment reappears as a major problem. What is not possible is thus the programme articulated by Allende when he said that ‘the political model toward socialism that my government is applying requires that the socio-economic revolution take place simultaneously with an uninterrupted economic expansion.’118What is not possible is the realization of Blum’s belief ‘that a better distribution…would revive production at the same time that it would satisfy justice’.119 What is not possible is a transition to socialism that begins with ‘une augmentation substantielle des salaires et traitment.’120 Faced with an economic crisis, threatened with loss of electoral support, concerned about the possibility of a fascist counter-revolution, social democrats abandon the project of transition or at least pause to wait for more auspicious times. They find the courage to explain to the working class that it is better to be exploited than to create a situation which contains the risk of turning against them. They refuse to stake their fortunes on a worsening of the crisis. They offer the compromise; they maintain and defend it. The question which remains is whether there exists a way to escape the alternative defined for the Left by Olof Palme: ‘Either to return to Stalin and Lenin, or take the road that joins the tradition of social democracy’.121


328

Notes 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33.

Dick Howard, ‘Re-reading Rosa Luxemburg’, Telos, 18, Winter 1973–4, pp. 89–107. Rosa Luxemburg, ‘The Mass Strike, the Political Party, and the Trade Unions’, in M.A.Waters (ed.), Rosa Luxemburg Speaks, New York 1970, p. 202. Georg Lukacs, History and Class Consciousness, London 1971, pp. 65–6; Nicos Poulantzas, Political Power and Social Classes, NLB, London 1973. Cf.Ralph Miliband, Marxism and Politics, London 1977, p. 129. Robert Michels, Political Parties, New York 1962, p. 270. Guy Konopniki, Vive le centenaire du PCF, Paris 1979, p. 53. Carl Landauer, European Socialism, 1, Berkeley 1959, p. 457. Herbet Tingsten, The Swedish Social Democrats, Totowa 1973, p. 362. Ralph Miliband, Parliamentary Socialism, London 1975, p. 69. Landauer, European Socialism, 1, pp. 472–3. Berndt Schiller, ‘Years of Crisis, 1906–14’, in Steven Koblick (ed.), Sweden’s Development from Poverty to Affluence, 1750–1970, Minneapolis 1975, pp. 208–17. Charles S.Maier, Recasting Bourgeois Europe, Princeton 1975, p. 158. William A.Lafferty, Economic Development and the Response of Labour in Scandinavia, Oslo 1971, p. 191. Miliband, Parliamentary Socialism, p. 148. Karl Kautsky, The Class Struggle, New York 1971, p. 186. Eric Hobsbawm, Revolutionaries, New York and London 1973, pp. 14–15. Jean Jaurès, L’esprit du socialisme, Paris 1971, p. 71. Eduard Bernstein, Evolutionary Socialism, New York 1961. For the views of Kautsky and Luxemburg, who were somewhat more cautious, see Massimo Salvadori, Karl Kautsky and the Socialist Revolution 1880–1938, NLB, London 1979 and Norman Geras, The Legacy of Rosa Luxemburg, NLB, London 1976. Harold Laski, Democracy in Crisis, Chapell Hill 1935, p. 77. Tingsten, The Swedish Social Democrats, p. 361. Ibid. Rosa Luxemburg, Reform of Revolution, New York 1970, p. 28. Salvadori, Karl Kaustky and the Socialist Revolution, pp. 66 and 80. Norbett Lesor,’ Austro-Marxism: a Reappraisal’, Journal of Contemporary History, II, 1976, pp. 133–48, p. 145. Carl E.Shorske, German Social Democracy 1905–1917, New York 1955, p. 43. Cf.Geras, The Legacy of Rosa Luxemburg. Peter Gay, The Dilemma of Democratic Socialism, New York 1970, p. 7. Frederick Engels and Karl Marx, Correspondence 1846–95, Moscow 1935, p. 486. Leslie Derfler, Socialism since Marx: A Century of the European Left, New York 1973, p. 59. R.C.K.Ensor, Modern Socialism as Set Forth by the Socialists in their Speeches, Writings and Programs, New York 1908, p. 54. Frederick Engels, ‘Introduction (1985) to Karl Marx’, The Class Struggles in France, 1848–50, Moscow 1960, p. 22. Karl Marx and Frederick Engels, ‘The Communist Manifesto’, The Revolutions of 1848, Penguin/NLR Marx Library, London 1976.

The state: critical concepts 34. 35. 36. 37. 38. 39. 40. 41.

42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57.

58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69.

329

Karl Marx, Surveys from Exile, Penguin/NLR Marx Library, London 1976. Kautsky, The Class Struggle, p. 43. Tingsten, The Swedish Social Democrats, p. 402. Karl Kautsky, Terrorisme et communisme, Paris 1919. Giuseppe Fiori, Antonio Gramsci: Life of a Revolutionary, NLB, London 1973, p. 112. Tingsten, The Swedish Social Democrats, p. 405. Shorske, German Social Democracy, p. 43. Xavier Mabille and Val R.Lorwin, ‘The Belgian Socialist Party’, in William E. Paterson and Alastair H.Thomas (eds.), Social Democratic Parties in Western Europe, London 1977, p. 392. Paterson and Thomas, Social Democratic Parties, p. 240. Derfler, Socialism since Marx, p. 58. Alain Bergounioux and Bernard Manin, La social-democratie ou le compromis, Paris 1979, p. 27. Marx and Engels, ‘Communist Manifesto’, The Revolutions of 1848. Ernest Mandel, The Formation of the Economic Thought of Karl Marx, New York and London 1972, p. 23. Jean-Paul Sartre, Critique of Dialectical Reason, NLB, London 1976. Karl Marx, ‘Address to the Communist League’, The Revolutions of 1848. Arthur Rosenberg, Democracy and Socialism, Boston 1965, p. 161. Rosenberg, Democracy and Socialism, p. 165. Tingsten, The Swedish Social Democrats, p. 567. Frederick Engels, A Contribution to the Critique of the Social Democratic Draft Programme of 1891, Moscow n.d., p. 56. Jean-Jacques Fiechtier, Le socialisme français: de l’affaire Drefyus á la grande guerre, Geneva 1965, p. 258. Bergounioux and Manin, La social-democratic ou le compromis, p. 25. Landauer, European Socialism, I, pp. 457–8. Raymond Fusilier, La parti socialiste suedois. Son organisation, Paris 1954, p. 29. Adam Przeworski and Ernest Underhill, The Process of Class Formation from Karl Kautsky’s The Class Struggle to Recent Controversies’, Politics and Society, 7 1977 pp. 343–402; and The Evolution of European Class Structure during the Twentieth Century’, unpublished MSS, University of Chicago 1979. Richard N.Hunt, German Social Democracy, 1918–33, Chicago 7970. William E.Patterson, The German Social Democratic Party’, in Paterson and Thomas, Social Democratic Parties in Western Europe. Carl Landauer, The Guesdists and the Small Farmer: Early Erosion of French Marxism’, International Review of Social History, 6 (1961), pp. 212–25. Kautsky, The Class Struggle. Derfler, Socialism since Marx, p. 48. Landauer, European Socialism, I, p. 468. Tingsten, The Swedish Social Democrats. Ross McKibbin, The evolution of the Labour Party, London 1974, p. 95. Samuel Beer, British Politics in the Collectivist Age, New York 1969 (2nd ed.); McKibbin, The Evolution of the Labour Party, p. 97. Michels, Political Parties, p. 254. Tingsten, The Swedish Social Democrats, p. 135. Arthur Henderson, The Aims of Labor, New York 1918 (2nd ed.), p. 125.

Social Democracy as a Historical Phenomenon 70.

71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89.

90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100.

330

Parti Communiste Français, Traite d’economie politique le capitalisme monopoliste d’etat, 2 vols., Paris 1971; Erik Olin Wright, ‘Class Boundaries in Advanced capitalism’, NLR 98 (1976), pp. 3–42. F.W.S.Craig (ed.), British Parliamentary Election Results, 1918–1949, Glasgow 1969, p. 130. Keith Hill, ‘Belgium: Political Change in a Segmented Society’, in Richard Rose (ed.), Electoral Behavior: a Comparative Handbook, New York 1974, p. 83. Vivian Gornick, The Romance of American Communism, New York 1977, p. 3. Szymon Chodak (ed.), Systemy Partyjne Wspovczesnego Kaptializmu, Warsaw 1962, p. 39. Tingsten, The Swedish Social Democrats, pp. 118–9. Ensor, Modern Socialism, p. 51. Ensor, Modern Socialism, p. 345. Tingsten, The Swedish Social Democrats, pp. 119–20. Kautsky, The Class Struggle, p. 93. Ensor, Modern Socilaism, p. 50. Ensor, Modern Socialism, p. 121. Gay, The Dilemma of Democratic Socialism, p. 258. Fiechtier, Le socialisme français, p. 163. Benjamin Barber. Frederick Engels, ‘Socialism: Utopian and Scientific’, In L.Feuer (ed.), Marx and Engels, Basic Writings on Politics and Philosophy, Garden City 1959, pp. 97–8. Tingsten, The Swedish Social Democrats, p. 239. Sir Stafford Cripps, ‘Democracy or Dictatorship?’—the Issue for the Labour Party’, Political Quarterly, 1933, pp. 467–81. Tingsten, p. 131. Charles S.Maier, Recasting Bourgeois Europe, Princeton 1975, p. 63; Gwyn Williams, Proletarian Order: Antonio Gramsci, Factory Councils and the Origins of Communism in Italy, 1911–21, London 1975, pp. 121–45; Paulo Spriano, Storia del Partito Communista Italiano, I, Da Bordiga a Gramsci, Turin 1967, pp. 50–63; Ernst Wigforss, ‘Industrial Democracy in Sweden’, International Labour Review, 1924, 9, pp. 667–79, p. 672. Karl Marx, The Class Struggles in France, 1848–50, Moscow 1952, p. 62. Tingsten, The Swedish Social Democrats, pp. 274–5. Karl Korsh, ‘What is Socialization?’, New German Critique, 6 (1965), pp. 60–82. Tingsten, The Swedish Democrats, p. 208. Miliband, Parliamentary Socialism, p. 101. Richard Lyman, ‘The British Labour Party: the Conflict between Socialist Ideals and Practical Politics between the Wards’, Journal of British Studies, 5 (1965), pp. 140–52. Richard Lyman, The First Labour Government 1924, London 1957, p. 106; for a similar statement by Branting see Tingsten, The Swedish Social Democrats, p. 238. Joel Colton, ‘Leon Blum and the French Socialists as a Government Party’, Journal of Politics, 15 (1953), pp. 517–543. Bergounioux and Manin, La social-democratie ou le compromis, p. 100. Irwin M.Wall, ‘The Resignation of the First Popular Front Government of Leon Blum, June 1937’, French Historical Studies 6, (1970), pp. 538–54. Karl-Gustaf Landgren, Den ‘Nya Ekonomin’ I Sverige’, Stockholm 1960; Otto Steigler, Studien Zur Enststehung Der Neuen Wirtshaftslehre in Schweden: Eine Anti-Kritik, Berlin 1971. Bo Gustafsson, ‘A Perennial of Doctrinal History: Keynes and the “Stockholm School’”, Economy and History, 17 (1973), pp. 114–28.


331

101. Gustav Möller, ‘The Unemployment Policy’, Annals of the American Academy of Political and Social Science, 197 (1938), pp. 47–72, p. 49. 102. Ernst Wigforss, The Financial Policy During Depression and Boom’, Annals of the American Academy of Political and Social Science, 197 (1938), pp. 25–40. 103. Asa Briggs, The Welfare State in Historical Perspective’, European Journal of Sociology, 2 (1961), pp. 221–258. 104. John Maynard Keynes, The General Theory of Employment, Interest and Money, New York 1964, p. 378. 105. Leif Lewin, ‘The Debate on Economic Planning in Sweden, in Steven Koblick (ed.), Sweden’s Development from Poverty to Affluence, 1750–1970, Minneapolis 1975, p. 286. 106. Quoted by Korsh, ‘What is Socialization?’, New German Critique, 6 (1975), p. 65. 107. Bergounioux and Manin, La social-democratie ou le compromis, p. 114. 108. Le Dossier des Nationalisations, Le Monde, Paris 1977. 109. Paul A.Samuelson, ‘The Pure Theory of Capital Expenditure’, in Joseph E.Steiglitz (ed.), The Collected Scientific Papers of Paul A.Samuelson, Cambridge (USA) 1966; Richard A.Musgrave, ‘Provision for Social Goods in the Market System’, Public Finance, 26 (1971), pp. 304–320. 110. See particularly Brandt’s view in Willy Brandt, Bruno Kreisky and Olaf Palme, La socialdemocratic et l’avenir, Paris 7976. 111. Andrew Martin, ‘Is Democratic Control of Capitalist Economies Possible?’, in Leon N.Lindberg et al., Stress and Contradiction in Modern Capitalism, Lexington 1975, pp. 13–56, p. 28. 112. Brandt, Kreisky and Palme, La social-democratie et l’avenir. 113. Timothy A.Tilton, ‘A Swedish Road to Socialism: Ernst Wigforss and the Ideological Foundations of Swedish Social Democracy’, American Political Science Review, 73 (1979), pp. 505–520, p. 516. 114. Maier, Recasting Bourgeois Europe, p. 194; Tingsten, The Swedish Social Democrats, p. 230. 115. Adam Przeworski, ‘Material Bases of Consent: Economic and Politics in a Hegemonic System’, Political Power and Social Theory, I (1979), pp. 21–63. 116. Serge-Christof Kolm, La transition socialiste, Paris 1977. 117. Michael Kalecki, The Lesson of the Blum Experiment’, Economic Journal, 48 (1938), pp. 26–41. 118. Stefan De Vylder, Allende’s Chile: the Political Economy of the Rise and Fall of the Unidad Popular, Cambridge 1976, p. 53. 119. Etienne Weil-Raynall, ‘Les obstacles economiques a l’experience Blum’, La Revenue Socialiste, 98 (1956). 120. Parti Communiste Français, Parti Socialiste Français, Mouvement des Radicaux de Gauche, Programme Commun du Gouvernement, Paris 1972, p. 1. 121. Brandt, Kreisky and Palme, La social-democratic et l’avenir, p. 120.


332

52 Transitions to Democracy: Toward a Dynamic Model* Dankwart A.Rustow

* Source: Comparative Politics, 1970, vol. 2, pp. 337–363.

I What conditions make democracy possible and what conditions make it thrive? Thinkers from Locke to Tocqueville and A.D.Lindsay have given many answers. Democracy, we are told, is rooted in man’s innate capacity for self-government or in the Christian ethical or the Teutonic legal tradition. Its birthplace was the field at Putney where Cromwell’s angry young privates debated their officers, or the more sedate House at Westminster, or the rock at Plymouth, or the forest cantons above Lake Lucerne, or the fevered brain of Jean Jacques Rousseau. Its natural champions are sturdy yeomen, or industrious merchants, or a prosperous middle class. It must be combined with strong local government, with a two-party system, with a vigorous tradition of civil rights, or with a multitude of private associations. Recent writings of American sociologists and political scientists favor three types of explanation, One of these, proposed by Seymour Martin Lipset, Philips Cutright, and others, connects stable democracy with certain economic and social background conditions, such as high per capita income, widespread literacy, and prevalent urban residence. A second type of explanation dwells on the need for certain beliefs or psychological attitudes among the citizens. A long line of authors from Walter Bagehot to Ernest Barker has stressed the need for consensus as the basis of democracy—either in the form of a common belief in certain fundamentals or of procedural consensus on the rules of the game, which Barker calls “the Agreement to Differ.” Among civic attitudes required for the successful working of a democratic system, Daniel Lerner has proposed a capacity for empathy and a willingness to participate. To Gabriel Almond and Sidney Verba, on the other hand, the ideal “civic culture” of a democracy suggests not only such participant but also other traditional or parochial attitudes.1 A third type of explanation looks at certain features of social and political structure. In contrast to the prevailing consensus theory, authors such as Carl J. Friedrich, E.E.Schattschneider, Bernard Crick, Ralf Dahrendorf, and Arend Lijphart have insisted that conflict and reconciliation are essential to democracy.2 Starting with a similar assumption, David B.Truman has attributed the vitality of American institutions to the citizens’ “multiple membership in potential groups”—a relationship which Lipset has

Transitions to Democracy: Toward a Dynamic Model

333

called one of “crosscutting politically relevant associations.”3 Robert A.Dahl and Herbert McClosky, among others, have argued that democratic stability requires a commitment to democratic values or rules, not among the electorate at large but among the professional politicians—each of these presumably linked to the other through effective ties of political organization.4 Harry Eckstein, finally, has proposed a rather subtle theory of “congruence”: to make democracy stable, the structures of authority throughout society, such as family, church, business, and trade unions, must prove the more democratic the more directly they impinge on processes of government.5 Some of these hypotheses are compatible with each other, though they may also be held independently—for example, those about prosperity, literacy, and consensus. Others—such as those about consensus and conflict—are contradictory unless carefully restricted or reconciled. Precisely such a synthesis has been the import of a large body of writing. Dahl, for instance, has proposed that in polyarchy (or “minorities rule,” the closest real-life approximation to democracy) the policies of successive governments tend to fall within a broad range of majority consensus.6 Indeed, after an intense preoccupation with consensus in the World War II years, it is now widely accepted that democracy is indeed a process of “accommodation” involving a combination of “division and cohesion” and of “conflict and consent”—to quote the key terms from a number of recent book titles.7 The scholarly debate thus continues, and answers diverge. Yet there are two notable points of agreement. Nearly all the authors ask the same sort of question and support their answers with the same sort of evidence. The question is not how a democratic system comes into existence. Rather it is how a democracy, assumed to be already in existence, can best preserve or enhance its health and stability. The evidence adduced generally consists of contemporary information, whether in the form of comparative statistics, interviews, surveys, or other types of data. This remains true even of authors who spend considerable time discussing the historical background of the phenomena that concern them—Almond and Verba of the civic culture, Eckstein of congruence among Norwegian social structures, and Dahl of the ruling minorities of New Haven and of oppositions in Western countries.8 Their key propositions are couched in the present tense. There may be a third feature of similarity underlying the current American literature of democracy. All scientific inquiry starts with the conscious or unconscious perception of a puzzle.9 What has puzzled the more influential authors evidently has been the contrast between the relatively smooth functioning of democracy in the English-speaking and Scandinavian countries and the recurrent crises and final collapse of democracy in the French Third and Fourth Republics and in the Weimar Republic of Germany. This curiosity is of course wholly legitimate. The growing literature and the increasingly subtle theorizing on the bases of democracy indicate how fruitful it has been. The initial curiosity leads logically enough to the functional, opposed to the genetic, question. And that question, in turn, is most readily answered by an examination of contemporary data about functioning democracies—perhaps with badly functioning democracies and nondemocracies thrown in for contrast. The functional curiosity also comes naturally to scholars of a country that took its crucial steps toward democracy as far back as the days of Thomas Jefferson and Andrew Jackson. It accords, moreover, with some of the characteristic trends in American social science in the last generation or


334

two—with the interest in systematic equilibria, in quantitative correlations, and in survey data engendered by the researcher’s own questions. Above all, it accords with a deepseated prejudice against causality. As Herbert A.Simon has strikingly put it, “…we are wary, in the social sciences, of asymmetrical relations. They remind us of pre-Humeian and pre-Newtonian notions of causality. By whip and sword we have been converted to the doctrine that there is no causation, only functional interrelation, and that functional relations are perfectly symmetrical. We may even have taken over, as a very persuasive analogy, the proposition ‘for every action, there is an equal and opposite reaction.’ “10 Students of developing regions, such as the Middle East, Southern Asia, tropical Africa, or Latin America, naturally enough have a somewhat different curiosity about democracy. The contrast that is likely to puzzle them is that between mature democracies, such as the United States, Britain, or Sweden today, and countries that are struggling on the verge of democracy, such as Ceylon, Lebanon, Turkey, Peru, or Venezuela. This will lead them to the genetic question of how a democracy comes into being in the first place.11 The question is (or at least was, until the Russian invasion of Czechoslovakia in 1968) of almost equal interest in Eastern Europe. The genesis of democracy, thus, has not only considerable intrinsic interest for most of the world; it has greater pragmatic relevance than further panegyrics about the virtues of Anglo-American democracy or laments over the fatal illnesses of democracy in Weimar or in several of the French Republics. In the following sections of this article I should like to examine some of the methodological problems involved in the shift from functional to genetic inquiry and then proceed to outline one possible model of the transition to democracy.

II What changes in concept or method does the shift from functional to genetic inquiry imply? The simplest answer would be, “None at all.” The temptation is to make the functional theories do double duty as genetic theories, to extend the perspective of Westminster and Washington versus Weimar and Paris to Ankara, Caracas, and Bucharest as well. If conditions such as consensus or prosperity will help to preserve a functioning democracy, it may be argued, surely they will be all the more needful to bring it into existence. Alas, the simple equation of function and genesis is a little too simple, and the argument a fortiori is, in fact, rather weak.12 The equation certainly does not seem to hold for most other types of political regimes. Military dictatorships, for instance, typically originate in secret plotting and armed revolt but perpetuate themselves by massive publicity and by alliances with civilian supporters. Charismatic leaders, according to Max Weber, establish their claim to legitimacy by performing seeming miracles but preserve it through routinization. A hereditary monarchy rests most securely on the subjects’ unquestioning acceptance of immemorial tradition; it evidently cannot be erected on such a principle. Communist regimes have been installed by revolutionary elites or through foreign conquest but consolidated through the growth of domestic mass parties and their bureaucracies. From physics and chemistry, too, the distinction between


335

the energy required to initiate and to sustain a given reaction is familiar. These arguments from analogy of course are just as inconclusive as the supposedly a fortiori one. Still, they shift the burden of proof to those who assert that the circumstances which sustain a mature democracy also favor its birth. The best known attempts to apply a single world-wide perspective to democracy, whether nascent or mature, are the statistical correlations compiled by Lipset and by Cutright.13 But Lipset’s article well illustrates the difficulty of applying the functional perspective to the genetic question. Strictly interpreted, his data bear only on function. His statistical findings all take the form of correlations at a given single point in time. In the 1950s his “stable democracies” generally had substantially higher per capita incomes and literacy rates than did his “unstable democracies,” or his unstable and stable authoritarianisms. Now, correlation evidently is not the same as causation—it provides at best a clue to some sort of causal connection without indicating its direction. Lipset’s data leave it entirely open, for example, whether affluent and literate citizens make the better democrats; whether democracies provide superior schools and a more bracing climate for economic growth; whether there is some sort of reciprocal connection so that a given increase in affluence or literacy and in democracy will produce a corresponding increment in the other, or whether there is some further set of factors, such as the industrial economy perhaps, that causes both democracy and affluence and literacy. A corresponding objection can be urged against the findings of Almond, Verba, and others that are based mainly on contemporary opinion or attitude surveys. Only further investigation could show whether such attitudes as “civic culture,” an eagerness to participate, a consensus on fundamentals, or an agreement on procedures are cause or effect of democracy, or both, or neither. Lipset’s title is true to his functional concern. He is careful to speak of “Some Social Requisites,” not prerequisites, “of Democracy,” and thus to acknowledge the difference between correlation and cause. But the subtlety has escaped many readers who unthinkingly translate “requisites” into “preconditions.”14 The text of the article, moreover, encourages the same substitution, for it repeatedly slips from the language of correlation into the language of causality. Significantly, on all those occasions economic and social conditions become the independent, and democracy the dependent, variable. A genetic theory will have to be explicit about distinguishing correlate from cause. This does not commit us to any old-fashioned or simple-minded view of causality, whereby every effect has but one cause and every cause but one effect. It does not preclude the “probabilistic” view recently argued by Almond and, indeed, espoused by every social statistician since Emile Durkheim and before.15 It does not rule out somewhat more sophisticated causal concepts such as Gunnar Myrdal’s spiral, Karl W.Deutsch’s quorum of prerequisites, Hayward R.Alker’s nonlinear correlations, or the notion of a threshold which Deane Neubauer recently applied to Lipset’s and Cutright’s propositions.16 Above all, a concern for causality is compatible with—indeed is indispensable to—a sceptical view that attributes human events to a mixture of law and chance. Such semideterminism is tantamount to an admission that the social scientist will never know enough to furnish a complete explanation, that he is at least as unlikely as the natural scientist to rival Laplace’s Demon. Nor do scholars who would theorize about the genesis of democracy need to concur in all their epistemology and metaphysics. But to be


336

geneticists at all they do have to inquire into causes. Only by such inquiry, I would add, can the social scientist accomplish his proper task of exploring the margins of human choice and of clarifying the consequences of the choices in that margin.17 It probably is no simple confusion between correlate and cause that leads Lipset’s readers astray, and, on occasion, the author as well. Rather it seems to be a tacit assumption that social and economic conditions are somehow more basic, and that we must look for the significant relations in this deeper layer rather than in the “superstructure” of political epiphenomena. Our current emphasis in political science on economic and social factors is a most necessary corrective to the sterile legalism of an earlier generation. But, as Lipset (together with Bendix) has himself warned in another context, it can easily “explain away the very facts of political life.”18 We have been in danger of throwing away the political baby with the institutional bathwater. Note that this widespread American economicism goes considerably beyond Marx and Engels, who saw the state as created by military conquest, economic regimes defined by their legal relations of property, and changes from one to the next brought about through political revolution. If they proclaimed themselves materialists or talked like economic determinists, it was mainly in protest against the wilder flights of Hegelian “idealism.” Any genetic theory of democracy would do well to assume a two-way flow of causality, or some form of circular interaction, between politics on the one hand and economic and social conditions on the other. Wherever social or economic background conditions enter the theory, it must seek to specify the mechanisms, presumably in part political, by which these penetrate to the democratic foreground. The political scientist, moreover, is entitled to his rights within the general division of labor and may wish to concentrate on some of the political factors without denying the significance of the social or economic ones. With Truman, Dahl, and others, I would tend to see the patterns of conflict and of recurrent or changing alignments as one of the central features of any political system. With Apter, I would consider choice as one of the central concerns of the political process.19 What goes for economics and sociology goes for psychology as well. Here, too, the relationship with politics is one of interaction and interdependence, so that political phenomena may have psychological consequences as well as vice versa. In explaining the origins of democracy we need not assume—as does much of the current survey research literature—that beliefs unilaterally influence actions. Rather, we may recognize with Leon Festinger and other social psychologists of the “cognitive dissonance” school that there are reciprocal influences between beliefs and actions.20 Many of the current theories about democracy seem to imply that to promote democracy you must first foster democrats—perhaps by preachment, propaganda, education, or perhaps as an automatic byproduct of growing prosperity. Instead, we should allow for the possibility that circumstances may force, trick, lure, or cajole non-democrats into democratic behavior and that their beliefs may adjust in due course by some process of rationalization or adaptation. To seek causal explanations, as I insisted earlier, does not imply simple-mindedness. Specifically, we need not assume that the transition to democracy is a world-wide uniform process, that it always involves the same social classes, the same types of political issues, or even the same methods of solution. On the contrary, it may be well to


337

assume with Harry Eckstein that a wide variety of social conflicts and of political contents can be combined with democracy.21 This is, of course, in line with the general recognition that democracy is a matter primarily of procedure rather than of substance. It also implies that, as among various countries that have made the transition, there may be many roads to democracy. Nor does a model of transition need to maintain that democratic evolution is a steady process that is homogeneous over time. Such a notion of temporal continuity and presumably of linear correlation seems to lurk behind much of the literature of the LipsetCutright genre. Temporal discontinuity, on the contrary, is implicit in the basic distinction drawn earlier in this article between the functional and genetic questions. The same discontinuity may be carried into the genetic scheme itself. For instance, it may be useful to single out certain circumstances as background factors and to proceed step-bystep to other factors that may become crucial in the preparation, decision, and consolidation phases of the process. Even in the same country and during the same phase of the process, political attitudes are not likely to be spread evenly through the population. Dahl, McClosky, and others have found that in mature democracies there are marked differences in the attitudes of professional politicians and of common citizens.22 Nor can we take it for granted that the politicians will all share the same attitudes. In so far as democracy is based on conflict, it may take two attitudes to make a quarrel. All these differences are likely, moreover, to be compounded during the formative period when part of the quarrel must ex hypothesi be between democrats and nondemocrats. Finally, a dynamic model of the transition must allow for the possibility that different groups—e.g., now the citizens and now the rulers, now the forces in favor of change and now those eager to preserve the past—may furnish the crucial impulse toward democracy.

III The methodological argument I have been advancing may be condensed into a number of succinct propositions. 1. The factors that keep a democracy stable may not be the ones that brought it into existence: explanations of democracy must distinguish between function and genesis. 2. Correlation is not the same as causation: a genetic theory must concentrate on the latter. 3. Not all causal links run from social and economic to political factors. 4. Not all causal links run from beliefs and attitudes to actions. 5. The genesis of democracy need not be geographically uniform: there may be many roads to democracy. 6. The genesis of democracy need not be temporally uniform: different factors may become crucial during successive phases. 7. The genesis of democracy need not be socially uniform: even in the same place and time the attitudes that promote it may not be the same for politicians and for common citizens. My refrain, like Sportin’ Life’s, has been, “It ain’t necessarily so.” Each proposition


338

pleads for the lifting of some conventional restriction, for the dropping of some simplifying assumption made in the previous literature, for the introduction of complicating diversifying factors. If the argument were to conclude on this sceptical note, it would set the researcher completely adrift and make the task of constructing a theory of democratic genesis well-nigh unmanageable. Fortunately, the genetic perspective requires or makes possible a number of new restrictions that more than compensate for the loss of the seven others. We may continue the listing of summary propositions before elaborating this second part of the methodological argument. 8. Empirical data in support of a genetic theory must cover, for any given country, a time period from just before until just after the advent of democracy. 9. To examine the logic of transformation within political systems, we may leave aside countries where a major impetus came from abroad. 10. A model or ideal type of the transition may be derived from a close examination of two or three empirical cases and tested by application to the rest. That diachronic data, covering more than a single point in time, are essential to any genetic theory should be obvious. Such a theory, moreover, must be based on cases where the process is substantially complete. Although control data on nondemocracies and on abortive and incipient cases may become important at a later stage of theorizing, it is more convenient to start out by studying a phenomenon where it actually has come into existence. The “advent” of democracy must not, of course, be understood as occurring in a single year. Since the emergence of new social groups and the formation of new habits are involved, one generation is probably the minimum period of transition. In countries that had no earlier models to emulate, the transition is likely to have come even more slowly. In Britain, for example, it may be argued that it began before 1640 and was not accomplished until 1918. For an initial set of hypotheses, however, it may be best to turn to countries where the process occurred relatively rapidly. The study of democratic transitions will take the political scientist deeper into history than he has commonly been willing to go. This implies many changes in method— beginning with suitable substitutions for survey data and for interviews. Even reliable statistics are harder to come by early in any democratic experiment. The United States Constitution (Article 1, Section 2) reminds us that our decennial census was introduced at that very time so that we might begin to govern ourselves by an accurate count of noses. Whatever the difficulties in the vastly increased use of historical data by social scientists, at least three arguments can be made in extenuation and encouragement. Man did not become a political animal in 1960 or in 1945, as much of our recent literature pretends to suppose. History, to paraphrase Georges Clemenceau, is far too important a topic to be left just to historians. And recently scholars in comparative politics have turned with increasing zest to historical themes. The list includes Almond, Leonard Binder, Dahl, Samuel P.Huntington, Lipset, Robert E. Ward, and Myron Weiner—not to speak of those like Friedrich and Deutsch to whom a political-historical perspective was natural to start with.23 The next restriction—the omission early in the inquiry of cases where the major impulse to democratization came from the outside—is in accord with the conventional division of labor between the subfields of comparative politics and international relations.


339

There are topics such as the theory of modernization where that division should be transcended from the start.24 In tracing the origins of democracy, too, both perspectives may be applied at once, as witness the suggestive work of Louis Hartz, the masterly synthesis by Robert Palmer, and the current research by Robert Ward on JapaneseAmerican interaction in the shaping of the 1947 constitution.25 But for a first attempt at a general theory it may be preferable to stick to countries where the transition occurred mainly within a single system. To speak of “major impulses from outside” or transitions “mainly within the system” acknowledges that foreign influences are almost always present. Throughout history, warfare has been a major democratizing force, because it has made necessary the marshalling of additional human resources.26 Democratic ideas moreover, have proved infectious whether in the days of Rousseau or of John F.Kennedy. And the violent overthrow of one oligarchy (e.g., France in 1830, Germany in 1918) has often frightened another into peaceful surrender (e.g., Britain in 1832, Sweden in 1918). From such ever present international influences we may distinguish situations where people arriving from abroad took an active part in the internal political process of democratization. A theory of democratic origins, that is to say, should leave aside at the beginning those countries where military occupation played a major role (postwar Germany and Japan), where democratic institutions or attitudes were brought along by immigrants (Australia and New Zealand), or where in these and other ways immigration played a major role (Canada, the United States, and Israel). The preference expressed earlier for relatively rapid instances of transition and the omission of immigrant countries amount to a very serious restriction, for they leave out of account, at this first stage of theorizing, all the English-speaking democracies. The reasons, however, seem cogent. Indeed, it may well be that American social scientists have added to their difficulties in understanding transitions to democracy by paying undue attention to Britain and the United States, which for the reasons just suggested prove to be among the hardest instances to analyze in genetic terms. The total of eight provisional exclusions still leaves (among extant democracies) about twenty-three cases which to base a comparative analysis, thirteen of which are in Europe: Austria, Belgium, Ceylon, Chile, Colombia, Costa Rica, Denmark, Finland, France, Iceland, Ireland, India, Italy, Lebanon, Luxembourg, Netherlands, Norway, Philippines, Sweden, Switzerland, Turkey, Uruguay, Venezuela.27 Among these twenty-odd democracies, the last methodological proposition urges an even narrower selection at this preliminary stage of theorizing. What is here involved is a choice between three research strategies: inclusion of all relevant cases, concentration on a single country, or some intermediate course. Completeness is of course desirable, and all the more so where the “universe” consists of no more than twenty or thirty cases. But the more nearly complete the coverage, the shallower it will have to be. The number of possible variables is so enormous (economic conditions, social cleavages, political alignments, psychological attitudes) that they could be handled only by means of the kind of simplifying assumptions that we rejected earlier on logical grounds. A test, no matter how complete, of a fallacious set of propositions would hardly yield convincing results. The country monograph would avoid this danger. Nor does it deliberately have to be


340

antitheoretical or “merely descriptive.” Any country study nevertheless sacrifices the advantages of comparison, the social scientist’s nearest substitute for a laboratory. No such study can tell us which strands in a tangle of empirical factors represent the development of democracy and which the national idiosyncrasies of Monographistan. The middle course avoids the twin dangers of inconclusive scholasticism and of factgrubbing. Instead, it can offer a more balanced and hence more fruitful blend of theory and empiricism. The many possible variables that can affect the origins of democracy and the even more complex relations among them can best be sorted out by looking at their total configuration in a limited number of cases—perhaps no more than two or three at the start. What will emerge from this exercise is a model, or as Weber used to call it, an “ideal type,” of the transition from oligarchy to democracy. Being an ideal type, it deliberately highlights certain features of empirical reality and deliberately distorts, simplifies, or omits others. Like any such construct, it must be judged initially by its internal coherence and plausibility but ultimately by its fruitfulness in suggesting hypotheses applicable to a wide variety of other empirical cases.28 It is at this further stage of testing that the demand for completeness comes once again into its own. The model I should like to sketch in the next few pages is based in large part on my studies of Sweden, a Western country that made the transition to democracy in the period from 1890 to 1920 and of Turkey, a Westernizing country where that process began about 1945 and is still underway. The choice of these two is accidental—except in terms of an autobiographical account for which this is not the occasion. I am now in the early stages of a study that will seek to refine the same set of hypotheses in the light of materials from a slightly larger and less arbitrary selection of countries.

IV A.Background Condition The model starts with a single background condition—national unity. This implies nothing mysterious about Blut und Boden or daily pledges of allegiance, about personal identity in the psychoanalyst’s sense, or about a grand political purpose pursued by the citizenry as a whole. It simply means that the vast majority of citizens in a democracy-tobe must have no doubt or mental reservations as to which political community they belong to. This excludes situations of latent secession, as in the late Habsburg and Ottoman Empires or in many African states today, and, conversely, situations of serious aspirations for merger as in many Arab states. Democracy is a system of rule by temporary majorities. In order that rulers and policies may freely change, the boundaries must endure, the composition of the citizenry be continuous. As Ivor Jennings phrased it tersely, “the people cannot decide until somebody decides who are the people.”29 National unity is listed as a background condition in the sense that it must precede all the other phases of democratization but that otherwise its timing is irrelevant. It may have been achieved in prehistoric times, as in Japan or Sweden; or it may have preceded the other phases by centuries, as in France, or by decades, as in Turkey. Nor does it matter by what means national unity has been established. The geographic


341

situation may be such that no serious alternative has ever arisen—Japan once again being the best example. Or a sense of nationality may be the product of a sudden intensification of social communication in a new idiom developed for the purpose. On the other hand, it may be the legacy of some dynastic or administrative process of unification. The various hypotheses proposed by Deutsch clearly become relevant here.30 I have argued elsewhere that in an age of modernization men are unlikely to feel a preponderant sense of loyalty except to a political community large enough to achieve some considerable degree of modernity in its social and economic life.31 This sort of hypothesis must be examined as part of a theory of nationhood, not of one of democratic development. What matters in the present context is only the result. I hesitate to call this result a consensus, for at least two reasons. First, national unity, as Deutsch argues, is the product less of shared attitudes and opinions than of responsiveness and complementarity. Second, “consensus” connotes consciously held opinion and deliberate agreement. The background condition, however, is best fulfilled when national unity is accepted unthinkingly, is silently taken for granted. Any vocal consensus about national unity, in fact, should make us wary. Most of the rhetoric of nationalism has poured from the lips of people who felt least secure in their sense of national identity—Germans and Italians in the past century and Arabs and Africans in the present, never English men, Swedes, or Japanese. To single out national unity as the sole background condition implies that no minimal level of economic development or social differentiation is necessary as a prerequisite to democracy. These social and economic factors enter the model only indirectly as one of several alternative bases for national unity or for entrenched conflict (see B below). Those social and economic indicators that authors are fond of citing as “background conditions” seem somewhat implausible at any rate. There are always nondemocracies that rank suspiciously high, such as Kuwait, Nazi Germany, Cuba or Congo-Kinshasa. Conversely, the United States in 1820, France in 1870, and Sweden in 1890 would have been sure to fail one or another of the proposed tests of urbanization or per capita income—not to speak of newspaper copies in circulation, or doctors movies, and telephones available to each one thousand inhabitants. The model thus deliberately leaves open the possibility of democracies (properly so called) in premodern, prenationalist times and at low levels of economic development. To find a meaningful definition of democracy that would cover modern parliamentary systems along with medieval forest cantons, ancient city states (the ones where slavery and metics were absent), and some of the pre-Colombian Indians may prove difficult. It is not a task that forms part of the present project; still, I should not like to foreclose the attempt. B.Preparatory Phase I hypothesize that, against this single background condition, the dynamic process of democratization itself is set off by a prolonged and inconclusive political struggle. To give it those qualities, the protagonists must represent well-entrenched forces (typically social classes), and the issues must have profound meaning to them. Such a struggle is likely to begin as the result of the emergence of a new elite that arouses a depressed and


342

previously leaderless social group into concerted action. Yet the particular social composition of the contending forces, both leaders and followers, and the specific nature of the issues will vary widely from one country to the next and in the same country from period to period. In Sweden at the turn of the century, it was a struggle first of farmers and then of an urban lower-middle and working class against a conservative alliance of bureaucrats, large landowners, and industrialists; and the issues were tariffs, taxation, military service, and suffrage. In Turkey in the last twenty years it has mainly been a contest of countryside versus city, more precisely of large and middling-size farmers (supported by most of the peasant electorate) against the heirs of the Kemalist bureaucratic-military establishment; the central issue has been industrialization versus agricultural development. In both these examples, economic factors have been of prime importance, yet the direction of causality has varied. In Sweden, it was a period of intense economic development that created new political tensions; at one crucial point rising wages enabled the Stockholm workers to overcome the existing tax barrier for the franchise. In Turkey, conversely, the demand for rural development was the consequence, not the cause, of beginning democratization.32 There may be situations where economic factors have played a much lesser role. In India and in the Philippines the prolonged contest between nationalist forces and an imperial bureaucracy over the issue of self-government may have served the same preparatory function as did class conflict elsewhere. In Lebanon the continuing struggle is mainly between denominational groups and the stakes are mainly government offices. Although political struggles of this sort naturally have their economic dimensions, only a doctrinaire economic determinist would derive colonialism or religious divisions from solely economic causes. James Bryce found in his classic comparative study that, “One road only has in the past led into democracy, viz., the wish to be rid of tangible evils.”33 Democracy was not the original or primary aim; it was sought as a means to some other end or it came as a fortuitous byproduct of the struggle. But, since the tangible evils that befall human societies are legion, Bryce’s single road dissolves into many separate paths. No two existing democracies have gone through a struggle between the very same forces over the same issues and with the same institutional outcome. Hence, it seems unlikely that any future democracy will follow in the precise footsteps of any of its predecessors. As Albert Hirschman has warned in his discussion of economic development, the search for ever more numerous preconditions or prerequisites may end up by proving conclusively that development always will be impossible and always has been.34 More positively, Hirschman and other economists have argued that a country can best launch into a phase of growth not by slavishly imitating the example of nations already industrialized, but rather by making the most of its particular natural and human resources and by fitting these accurately into the international division of labor.35 Similarly, a country is likely to attain democracy not by copying the constitutional laws or parliamentary practices of some previous democracy, but rather by honestly facing up to its particular conflicts and by devising or adapting effective procedures for their accommodation. The serious and prolonged nature of the struggle is likely to force the protagonists to


343

rally around two banners. Hence polarization rather than pluralism, is the hallmark of this preparatory phase. Yet there are limitations implicit in the requirement of national unity—which, of course, must not only preexist but also continue. If the division is on sharply regional lines, secession rather than democracy is likely to result. Even among contestants geographically interspersed there must be some sense of community or some even balance of forces that makes wholesale expulsion or genocide impossible. The Turks are beginning to develop a set of democratic practices among themselves, but fifty years ago they did not deal democratically with Armenians or Greeks. Crosscutting cleavages have their place in this preparatory phase as a possible means of strengthening or preserving that sense of community. Dahl notes wistfully that “one perennial problem of opposition is that there is either too much or too little.”36 The first two elements of the model between them will ensure that there is the right amount. But struggle and national unity cannot simply be averaged out, since they cannot be measured along the same scale. Strong doses of both must be combined, just as it may be possible to combine sharp polarization with crosscutting cleavages. Furthermore, as Mary Parker Follett, Lewis A.Coser, and others have insisted, certain types of conflict in themselves constitute creative processes of integration.37 What infant democracy requires is not a lukewarm struggle but a hot family feud. This delicate combination implies, of course, that many things can go wrong during the preparatory phase. The fight may go on and on till the protagonists weary and the issues fade away without the emergence of any democratic solution along the way. Or one group may find a way of crushing the opponents after all. In these and other ways an apparent evolution toward democracy may be deflected, and at no time more easily than during the preparatory phase. C.Decision Phase Robert Dahl has written that, “Legal party opposition…is a recent and unplanned invention,”38 This accords with Bryce’s emphasis on the redress of specific grievances as democracy’s vehicle and with the assumption here that the transition to democracy is a complex process stretching over many decades. But it does not rule out suffrage or freedom of opposition as conscious goals in the preparatory struggle. Nor does it suggest that a country ever becomes a democracy in a fit of absentmindedness. On the contrary, what concludes the preparatory phase is a deliberate decision on the part of political leaders to accept the existence of diversity in unity and, to that end, to institutionalize some crucial aspect of democratic procedure. Such was the decision in 1907, which I have called the “Great Compromise” of Swedish politics, to adopt universal suffrage combined with proportional representation.39 Instead of a single decision there may be several. In Britain, as is well-known, the principle of limited government was laid down in the compromise of 1688, cabinet government evolved in the eighteenth century, and suffrage reform was launched as late as 1832. Even in Sweden, the dramatic change of 1907 was followed by the further suffrage reform of 1918 which also confirmed the principle of cabinet government. Whether democracy is purchased wholesale as in Sweden in 1907 or on the installment plan as in Britain, it is acquired by a process of conscious decision at least on the part of


344

the top political leadership. Politicians are specialists in power, and a fundamental power shift such as that from oligarchy to democracy will not escape their notice. Decision means choice, and while the choice of democracy does not arise until the background and preparatory conditions are in hand it is a genuine choice and does not flow automatically from those two conditions. The history of Lebanon illustrates the possibilities of benevolent autocracy or of foreign rule as alternative solutions to entrenched struggles within a political community.40 And of course a decision in favor of democracy, or some crucial ingredient of it, may be proposed and rejected—thus leading to a continuation of the preparatory phase or to some sort of abortive outcome. The decision in favor of democracy results from the interplay of a number of forces. Since precise terms must be negotiated and heavy risks with regard to the future taken, a small circle of leaders is likely to play a disproportionate role. Among the negotiating groups and their leaders may be the protagonists of the preparatory struggle. Other participants may include groups that split off from one or the other side or new arrivals on the political stage. In Sweden these new and intermediate groups played a crucial role. Conservatives and Radicals (led by industrialists on one side and intellectuals on the other) had sharpened and crystallized the issues throughout the 1890s. Then came a period of stalemate when discipline in all the recently formed parliamentary parties broke down—a sort of randomization process in which many compromises, combinations, and permutations were devised and explored The formula that carried the day in 1907 included crucial contributions from a moderately conservative bishop and a moderately liberal farmer, neither of whom played a very prominent role in politics before or after this decision phase. Just as there can be different types of sponsors and different contents of the decision, so the motives from which it is proposed and accepted will vary from case to case. The forces of conservatism may yield from fear that continued resistance may lose them even more ground in the end. (Such thoughts were on the minds of British Whigs in 1832 and of Swedish conservatives in 1907.) Or they may belatedly wish to live up to principles long proclaimed; such was the Turkish transition to a multiparty system announced by President Inönü in 1945. The radicals may accept the compromise as a first installment, confident that time is on their side and that future installments are bound to follow. Both conservatives and radicals may feel exhausted from a long struggle or fearful of a civil war. This consideration is likely to loom large if they have been through such a war in recent memory. As Barrington Moore has aptly proposed, the English civil war was a crucial “contribution of early violence to later gradualism.”41 In short, democracy, like any collective human action, is likely to stem from a large variety of mixed motives. The decision phase may well be considered an act of deliberate, explicit consensus. But, once again, this somewhat nebulous term should be carefully considered and perhaps replaced with less ambiguous synonyms. First of all, as Bryce suggests, the democratic content of the decision may be incidental to other substantive issues. Second, in so far as it is a genuine compromise it will seem second-best to all major parties involved—it certainly will not represent any agreement on fundamentals. Third, even on procedures there are likely to be continuing differences of preference. Universal suffrage with proportional representation, the content of the Swedish compromise of 1907, was about equally distasteful to the conservatives (who would rather have continued the old


345

plutocratic voting system) and to the liberals and socialists (who wanted majority rule undiluted by proportional representation). What matters at the decision stage is not what values the leaders hold dear in the abstract, but what concrete steps they are willing to take. Fourth, the agreement worked out by the leaders is far from universal. It must be transmitted to the professional politicians and to the citizenry at large. These are two aspects of the final, or habituation, phase of the model. D. Habituation Phase A distasteful decision, once made, is likely to seem more palatable as one is forced to live with it. Every day experience can supply concrete illustrations of this probability for each of us. Festinger’s theory of “cognitive dissonance” supplies a technical explanation and experimental support.42 Democracy, moreover, is by definition a competitive process, and this competition gives an edge to those who can rationalize their commitment to it, and an even greater edge to those who sincerely believe in it. The transformation of the Swedish Conservative Party from 1918 to 1936 vividly illustrates the point. After two decades those leaders who had grudgingly put up with democracy or pragmatically accepted it retired or died and were replaced by others who sincerely believed in it. Similarly, in Turkey there is a remarkable change from the leadership of Ismet Inönü, who promoted democracy out of a sense of duty, and Adnan Menderes who saw in it an unprecedented vehicle for his ambition, to younger leaders in each of their parties who understand democracy more fully and embrace it more wholeheartedly. In short, the very process of democracy institutes a double process of Darwinian selectivity in favor of convinced democrats: one among parties in general elections and the other among politicians vying for leadership within these parties. But politics consists not only of competition for office. It is above all, a process for resolving conflicts within human groups—whether these arise from the clash of interests or from uncertainty about the future. A new political regime is a novel prescription for taking joint chances on the unknown. With its basic practice of multilateral debate, democracy in particular involves a process of trial and error, a joint learning experience. The first grand compromise that establishes democracy, if it proves at all viable, is in itself a proof of the efficacy of the principle of conciliation and accommodation. The first success, therefore, may encourage contending political forces and their leaders to submit other major questions to resolution by democratic procedures. In Sweden, for instance, there had been a general political stalemate in the last third of the nineteenth century over the prime issues of the day—the taxation and conscription systems inherited from the sixteenth century. But in the two decades after 1918, when democracy was fully adopted by the Swedes, a whole host of thorny questions was wittingly or unwittingly resolved. The Social Democrats surrendered their earlier pacifism, anticlericalism, and republicanism, as well as the demand for nationalization of industry (although they found it hard to admit this last point). The conservatives, once staunchly nationalist, endorsed Swedish participation in international organizations. Above all, conservatives and liberals fully accepted government intervention in the economy and the social welfare state. Of course, the spiral that in Sweden went upward to greater and greater successes for


346

the democratic process may also go downward. A conspicuous failure to resolve some urgent political question will damage the prospects of democracy; if such a failure comes early in the habituation phase, it may prove fatal. Surveying the evolution of political debate and conflict in the Western democracies over the last century, it is striking to observe the difference between social and economic issues, which democracies handled with comparative ease, and issues of community, which have proved far more troublesome.43 With the advantage of a century’s hindsight, it is easy to see that Marx’s estimate was wrong at crucial points. In nationality he saw a cloak for bourgeois class interests. He denounced religion as the opiate of the masses. In economics, by contrast, he foresaw very real and increasingly bitter struggles that would end by bringing bourgeois democracy crashing down. But in fact democracy has proved most effective in resolving political questions where the major divisions have been social and economic, as in Britain, Australia, New Zealand, and the Scandinavian countries. It has been the fight among religious, national, and racial groups, instead, that has proved most tenacious and has caused recurrent bitterness, as in Belgium, Holland, Canada, and the United States. The reasons are not hard to find. On the socioeconomic front Marxism itself became a sufficient force in Europe to serve to some extent as a self-disconfirming prophecy. But beyond this there is a fundamental difference in the nature of the issues. On matters of economic policy and social expenditures you can always split the difference. In an expanding economy, you can even have it both ways: the contest for higher wages, profits, consumer savings, and social welfare payments can be turned into a positive-sum game. But there is no middle position between Flemish and French as official languages, or between Calvinism, Catholicism, and secularism as principles of education. The best you can get here is an “inclusive compromise”44—a log-rolling deal whereby some government offices speak French and some Flemish, or some children are taught according to Aquinas, some, Calvin, and some, Voltaire. Such a solution may partly depoliticize the question. Yet it also entrenches the differences instead of removing them, and accordingly it may convert political conflict into a form of trench warfare. The difficulty that democracy finds in resolving issues of community emphasizes the importance of national unity as the background condition of the democratization process. The hardest struggles in a democracy are those against the birth defects of the political community. The transition to democracy, it was suggested earlier, may require some common attitudes and some distinct attitudes on the part of the politician and of the common citizen. The distinction is already apparent during the decision phase when the leaders search for compromise while their followers wearily uphold the banners of the old struggle. It becomes even more readily apparent during the habituation phase, when three sorts of process are at work. First, both politicians and citizens learn from the successful resolution of some issues to place their faith in the new rules and to apply them to new issues. Their trust will grow more quickly if in the early decades of the new regime, a wide variety of political tendencies can participate in the conduct of affairs, either by joining various coalitions or by taking turns as government and opposition. Second, as we just saw, experience with democratic techniques and competitive recruitment will confirm the politicians in their democratic practices and beliefs. Third, the population at


347

large will become firmly fitted into the new structure by the forging of effective links of party organization that connect the politicians in the capital with the mass electorate through the country. These party organizations may be a direct continuation of those that were active during the preparatory, or conflict, phase of democratization, and a suffrage extension at the time of the democratic “decision” may now have given them a free field. It is possible, on the other hand, that no parties with a broad popular base emerged during the conflict phase and that the suffrage extension was very limited. Even under such conditions of partial democratization of the political structure, a competitive dynamic that completes the process may have been set off. The parliamentary parties will seek support from constituency organizations to insure a steady supply of members for their group in future parliaments. Now this and now that political group may see a chance to steal a march on its opponents by enlarging the electorate or by removing other obstacles to majority control. This, roughly, would seem to have been the nature of British developments between 1832 and 1918. Complete democratization, of course, is the only logical stopping point for such a dynamic.

V The model here presented makes three broad assertions. First, it says that certain ingredients are indispensable to the genesis of democracy. For one thing, there must be a sense of national unity. For another, there must be entrenched and serious conflict. For a third, there must be a conscious adoption of democratic rules. And, finally, both politicians and electorate must be habituated to these rules. Secondly, the model asserts that these ingredients must be assembled one at a time. Each task has its own logic and each has its natural protagonists—a network of administrators or a group of nationalist literati for the task of unification, a mass movement of the lower class, perhaps led by upper class dissidents, for the task of preparatory struggle, a small circle of political leaders skilled at negotiation and compromise for the formulation of democratic rules, and a variety of organization men and their organizations for the task of habituation. The model thus abandons the quest for “functional requisites” of democracy; for such a quest heaps all these tasks together and thus makes the total job of democratization quite unmanageable. The argument here is analogous to that which has been made by Hirschman and others against the theory of balanced economic growth. These economists do not deny that the transition from a primitive subsistence economy to a mature industrial society involves changes on all fronts—in working skills, in capital formation, in the distribution system, in consumption habits, in the monetary system, and so forth. But they insist that any country that attempted all these tasks at once would in practice find itself totally paralysed—that the stablest balance is that of stagnation. Hence the economic developer’s problem, in their view, becomes one of finding backward and forward “linkages,” that is, of devising a manageable sequence of tasks. Thirdly, the model does suggest one such sequence fom national unity as background, through struggle, compromise, and habitation, to democracy. The cogency of this


348

sequence is brought home by a deviant development in Turkey in the years after 1945. The Turkish commitment to democracy was made in the absence of prior overt conflict between major social groups or their leading elites. In 1950 there was the first change of government as the result of a new electoral majority, but in the next decade there was a drift back into authoritarian practices on the part of this newly elected party, and in 1960– 1961 the democratic experiment was interrupted by a military coup. These developments are not unconnected: Turkey paid the price in 1960 for having received its first democratic regime as a free gift from the hands of a dictator. But after 1961 there was a further evolution in the more appropriate sequence. The crisis of 1960–1961 had made social and political conflict far more acceptable, and a full range of social and economic issues was debated for the first time. The conflict that shaped up was between the military on one side and the spokesmen of the agrarian majority on the other—and the compromise between these two allowed the resumption of the democratic experiment on a more secure basis by 1965. In the interests of parsimony, the basic ingredients of the model have been kept to four, and the social circumstances or psychological motivations that may furnish each of them have been left wide open. Specifically, the model rejects what are sometimes proposed as preconditions of democracy, e.g., high levels of economic and social development or a prior consensus either on fundamentals or on the rules. Economic growth may be one of the circumstances that produces the tensions essential to the preparatory or conflict phase—but there are other circumstances that might also serve. Mass education and social welfare services are more likely to be the result of democratization. Consensus on fundamentals is an implausible precondition. A people who were not in conflict about some rather fundamental matters would have little need to devise democracy’s elaborate rules for conflict resolution. And the acceptance of those rules is logically a part of the transition process rather than its prerequisite. The present model transfers various aspects of consensus from the quiescent state of preconditions to that of active elements in the process. I here follow the lead of Bernard Crick, who has strikingly written: It is often thought that for this “master science” [i.e., democratic politics] to function, there must already be in existence some shared idea of a “common good,” some “consensus” or consensus juris. But this common good is itself the process of practical reconciliation of the interests of the various… aggregates, or groups which compose a state; it is not some external and intangible spiritual adhesive…. Diverse groups hold together, firstly, because they have a common interest in sheer survival, and, secondly, because they practise politics—not because they agree about ‘fundamentals,’ or some such concept too vague, too personal, or too divine ever to do the job of politics for it. The moral consensus of a free state is not something mysteriously prior to or above politics: it is the activity (the civilizing activity) of politics itself.45 The basis of democracy is not maximum consensus. It is the tenuous middle ground between imposed uniformity (such as would lead to some sort of tyranny) and implacable hostility (of a kind that would disrupt the community in civil war or secession). In the


349

process of genesis of democracy, an element of what might be termed consensus enters at three points at least. There must be a prior sense of community, preferably a sense of community quietly taken for granted that is above mere opinion and mere agreement. There must be a conscious adoption of democratic rules, but they must not be so much believed in as applied, first perhaps from necessity and gradually from habit. The very operation of these rules will enlarge the area of consensus step-by-step as democracy moves down its crowded agenda. But new issues will always emerge and new conflicts threaten the newly won agreements. The characteristic procedures of democracy include campaign oratory, the election of candidates, parliamentary divisions, votes of confidence and of censure—a host of devices, in short, for expressing conflict and thereby resolving it. The essence of democracy is the habit of dissension and conciliation over ever-changing issues and amidst ever-changing alignments. Totalitarian rulers must enforce unanimity on fundamentals and on procedures before they can get down to other business. By contrast, democracy is that form of government that derives its just powers from the dissent of up to one half of the governed.

Notes This article was presented at the annual meeting of the American Political Science Association, New York City, September 1969. The author is grateful for financial support at various stages of his researches into democracy from the John Simon Guggenheim Foundation, the Ford Foundation, and the National Science Foundation. He jealously claims the full blame for his errors, foibles, and follies as revealed in this essay. 1. Ernest Barker, Reflections on Government (Oxford, 1942), p. 68; Daniel Lerner et al., The Passing of Traditional Society (Glencoe, 1958), pp. 49ff., 60ff.; Gabriel Almond and Sidney Verba, The Civic Culture (Princeton, 1968) 2. Carl J.Friedrich, The New Belief in the Common Man (Boston, 1942); E.E. Schattschneider, The Semi-Sovereign People (New York, 1960); Bernard Crick, In Defence of Politics, rev. ed. (Penguin Books, 1964); Ralf Dahrendorf, Class and Class Conflict in Industrial Society (Stanford, 1959); Arend Lijphart, The Politics of Accommodation (Berkeley and Los Angeles, 1968). 3. David B.Truman, The Governmental Process (New York, 1951), p. 514; S.M. Lipset, Political Man (New York, 1960), pp. 88ff. Already A.Lawrence Lowell had spoken of the need for a party alignment where “the line of division is vertical,” cutting across the horizontal division of classes. Government and Parties in Continental Europe (Boston, 1896), vol. 2, pp. 65ff. 4. Robert A.Dahl, Who Governs? (New Haven, 1961); Herbert McClosky “Consensus and Ideology in American Politics,” American Political Science Review, LVIII [June 1964); James W.Prothro and Charles M.Grigg, “Fundamental Principles of Democracy: Bases of Agreement and Disagreement,” Journal of Politics, XXII (May 1960). 5. Harry Eckstein, The Theory of Stable Democracy (Princeton, 1961) and Division and Cohesion in a Democracy (Princeton, 1965). 6. Robert A.Dahl, A Preface to Democratic Theory (Chicago, 1956). 7. Lijphart; Eckstein; Dahl, Pluralist Democracy in the United States: Conflict and Consent


350

(Chicago, 1967). 8. Almond and Verba; Eckstein; Dahl, Who Governs? and ed. Political Oppositions in Western Democracies (New Haven, 1966). 9. See Thomas Kuhn, The Structure of Scientific Revolutions (Chicago, 1962). 10. Herbert A.Simon, Models of Man: Social and Rational (New York, 1957). 11. For a general discussion of the question of democracy in the context of recent modernizing countries, see Rustow, A World of Nations: Problems of of Political Modernization (Washington, 1967), Ch. 7, which states some of the present argument in summary form. 12. “…a political form may persist under conditions normally adverse to the emergence of that form” (Lipset, p. 46). 13. Seymour Martin Lipset, “Some Social Requisites of Democracy: Economic Development and Political Legitimacy,” American Political Science Review, LIII (March 1959); idem, Political Man; Philips Cutright, “National Political Development: Measurement and Analysis,” American Sociological Review, XXVIII (April 1963). 14. Rupert Emerson, From Empire to Nation (Cambridge, 1960), p 278, paraphrases Lipset to this effect. M.Rejai, in his useful anthology and commentary Democracy: The Contemporary Theories (New York, 1967), includes an excerpt from the article under the heading, “Socioeconomic Preconditions” (pp. 242–247). 15. Gabriel A.Almond and James S.Coleman, eds. The Politics of the Developing Areas (Princeton, 1960), Introduction. 16. Gunnar Myrdal, An American Dilemma (New York, 1944), Appendix; Hay ward R. Alker, Jr., “The Long Road to International Relations Theory: Problems of Statistical NonAdditivity,” World Politics, XVIII (July 1966): Deane Neubauer “Some Conditions of Democracy,” American Political Science Review, LXI (December 1967). 17. This statement of the function of the social scientist is taken from Rustow, A World of Nations, p. 17; the next two paragraphs paraphrase ibid., pp. 142ff. 18. Reinhard Bendix and Seymour Martin Lipset, “Political Sociology,” Current Sociology, VI, No. 2 (1957), 85. 19. David E.Apter, The Politics of Modernization (Chicago, 1985). 20. Leon Festinger, A Theory of Cognitive Dissonance (Stanford, 1957). 21. Eckstein, Division and Cohesion, pp. 183–85. 22. Dahl, Who Governs?; McClosky; Prothro and Grigg. 23. Almond, current study on nineteenth-century Britain; Leonard Binder, ed. Politics in Lebanon (New York, 1966); Dahl, see nn. 4,7, and 8; Karl W.Deutsch, Nationalism and Social Communication (New York, 1953) and Deutsch et al., Political Community and the North Atlantic Area (Princeton, 1957); Carl J.Friedrich, Constitutional Government and Democracy (Boston 1950); Samuel P.Huntington, “Political Modernization: America vs. Europe.’ World Politics, XVIII (April 1966); S.M.Lipset, The First New Nation (New York, 1963), and Lipset and Stein Rokkan, eds. Party Systems and Voter Alignments (New York, 1967); Robert E.Ward and D.A.Rustow, eds. Political Modernization in Japan and Turkey (Princeton, 1964), Myron Weiner, current study on nineteenth-century social history of the Balkans. 24. In this combination lies the strength of Cyril E.Black’s Dynamics of Modernization (New York, 1966) compared to most of the other literature on the subject. 25. Louis Hartz et al., The Founding of New Societies (New York, 1964); R.R.Palmer, The Age of the Democratic Revolution, 2 vols. (Princeton, 1959–64). 26. See, e.g., Bertrand de Jouvenel, On Power (New York, 1948). 27. This list, together with the eight omissions noted (Australia, Canada, Germany Israel, Japan.


28. 29. 30. 31. 32.

33. 34. 35.

36. 37.

38. 39. 40. 41. 42. 43. 44. 45.

351

New Zealand, United Kingdom, United States), corresponds to the one I gave in A World of Nations, pp. 290ff., with the following exceptions: Greece has been omitted because democracy was superseded by a military coup in 1967; Mexico was omitted because, on second thought, I do not believe that it meets the criterion of “a government based on three or more consecutive, popular, and competitive elections”—the problems of course being the severe de facto restrictions on competition; Turkey and Venezuela have been added because they now have begun to meet the criterion. For a recent, lucid restatement of the rationale for such models or ideal types, see T.B.Bottomore, Elites and Societies (New York, 1965), p. 32. W.Ivor Jennings, The Approach to Self-Government (Cambridge, 1956), p. 56. Deutsch, Nationalism and Social Communication; Deutsch et al., Political Community and the North Atlantic Area. Rustow, A World of Nations, pp. 30ff. and International Encyclopedia of the Social Sciences, s.v. “Nation.” For developments in Sweden see Rustow, The Politics of Compromise: A Study of Parties and Cabinet Government in Sweden (Princeton, 1955), Chs. 1–3, and Douglas A. Verney, Parliamentary Reform in Sweden, 1866–1921 (Oxford, 1957). On Turkey see Ward and Rustow and the following essays by Rustow: “Politics and Islam in Turkey,” in R.N.Frye, ed. Islam and the West (The Hague, 1957), pp. 69–107; “Turkey: The Tradition of Modernity,” in Lucian W.Pye and Verba, eds. Political Culture and Political Development (Princeton, 1965), pp. 171–198: “The Development of Parties in Turkey,” in Joseph LaPalombara and Myron Weiner, eds. Political Parties and Political Development (Princeton, 1966), pp. 107–133; and “Politics and Development Policy,” in F.C.Shorter, ed. Four Studies in the Economic Development of Turkey (London, 1967), pp. 5–31. James Bryce, Modern Democracies (London, 1921), vol. 2, p. 602. Albert O.Hirschman, Journeys Toward Progress (New York, 1963), pp. 6ff. Ibid., and Hirschman, The Strategy of Economic Development (New Haven, 1958), and Hirschman, “Obstacles to Development: A Classification and a Quasi-Vanishing Act,” Economic Development and Cultural Change, XIII (July 1965), 385–393. Dahl et al., Political Oppositions in Western Democracies, p. 397. Mary Parker Follett, The New State (New York, 1989) and Creative Experience (New York, 1924); Lewis A.Coser, The Function of Social Conflict (Glencoe, 1965), p. 121 and passim. A widespread contrary position has recently be restated by Edward Shils, who write in reference to Lebanon: “Civility will not be strengthened by crisis. It can only grow slowly and in a calm atmosphere. The growth of civility is a necessary condition for Lebanon’s development…into a genuinely democratic system” (in Binder et al., Politics in Lebanon, p. 10). I find it hard to think of situations where there have been any notable advances in either civility or democracy except as the result of crisis. Dahl et al., Political Oppositions in Western Democracies, p. xi. Rustow, The Politics of Compromise, p. 69. Binder, ed., Politics in Lebanon. Barrington Moore, Jr., Social Origins of Dictatorship and Democracy (Boston, 1966), p. 3. Festinger, A Theory of Cognitive Dissonance. The contrast emerges implicitly from the country studies in Dahl, ed., Political Oppositions in Western Democracies. Rustow, Politics of Compromise, p. 231. Crick, In Defence of Politics, p. 24.


352

53 Dilemmas of Democratization in Latin America* Terry Lynn Karl

* Source: Comparative Politics, 1990, vol. 23, pp. 1–21.

The demise of authoritarian rule in Argentina, Bolivia, Brazil, Chile, Ecuador, Peru, and Uruguay, when combined with efforts at political liberalization in Mexico and the recent election of civilian presidents in Guatemala, El Salvador, Honduras, and Nicaragua, represents a political watershed in Latin America. This wave of regime changes in the 1980s places a number of questions on the intellectual and political agenda for the continent. Will these newly emergent and fragile democracies in South America be able to survive, especially in the context of the worst economic recession since the 1930s? Can the liberalization of authoritarian rule in Central America and the possible prospect of honest competitive elections in Mexico be transformed into genuine democratic transitions? Will previously consolidated political democracies such a.c Venezuela and Costa Rica be able to extend the basic principles of citizenship into economic and social realms, or will they be “deconsolidated” by this challenge and revert to a sole preoccupation with survivability?1 Behind such questions lies a central concern expressed by Dankwart A.Rustow almost twenty years ago: “What conditions make democracy possible and what conditions make it thrive?”2 This article addresses Rustow’s query by arguing the following: First, the manner in which theorists of comparative politics have sought to understand democracy in developing countries has changed as the once-dominant search for prerequisites of democracy has given way to a more process-oriented emphasis on contingent choice. Having undergone this evolution, theorists should now develop an interactive approach that seeks explicitly to relate structural constraints to the shaping of contingent choice. Second, it is no longer adequate to examine regime transitions writ large, that is, from the general category of authoritarian rule to that of democracy. Such broad-gauged efforts must be complemented by the identification of different types of democracy that emerge from distinctive modes of regime transition as well as an analysis of their potential political, economic, and social consequences. Before these issues and their implications for the study of Latin America can be addressed, however, a definition of democracy must be established.

Defining Democracy Defining democracy is no simple task because the resolution of a number of disputes over

Dilemmas of Democratization in Latin America

353

both its prospects and evaluation rests on how the term itself is operationalized. If, for example, democracy is defined in a Schumpeterian manner as a polity that permits the choice between elites by citizens voting in regular and competitive elections, the militarized countries of Central America could be classified as political democracies by many scholars, just as they are (with the exception of Sandínista Nicaragua) by U.S. policymakers.3 But if the definition is expanded to include a wider range of political conditions—from lack of restrictions on citizen expression, to the absence of discrimination against particular political parties, to freedom of association for all interests, to civilian control over the military—these same countries (with the exception of Costa Rica) could scarcely be classified under this rubric. The problem is compounded when a number of substantive properties—such as the predominance of institutions that faithfully translate individual preferences into public policy through majoritarian rule, the incorporation of an ever-increasing proportion of the population into the process of decision making, and the continuous improvement of economic equity through the actions of governing institutions—are included either as components or empirical correlates of democratic rule.4 Approaches that stipulate socioeconomic advances for the majority of the population and active involvement by subordinate classes united in autonomous popular organizations as defining conditions intrinsic to democracy are hard-pressed to find “actual” democratic regimes to study. Often they are incapable of identifying significant, if incomplete, changes towards democratization in the political realm. Moreover, they are cut off from investigating empirically the hypothetical relationship between competitive political forms and progressive economic outcomes because this important issue is assumed away by the very definition of regime type. While these substantive properties are ethically desirable to most democrats, such conceptual breadth renders the definition of democracy virtually meaningless for practical application.5 For these reasons, I will settle for a middle-range specification of democracy. It is defined as “a set of institutions that permits the entire adult population to act as citizens by choosing their leading decision makers in competitive, fair, and regularly scheduled elections which are held in the context of the rule of law, guarantees for political freedom, and limited military prerogatives.” Specified in this manner, democracy is a political concept involving several dimensions: (1) contestation over policy and political competition for office; (2) participation of the citizenry through partisan, associational, and other forms of collective action; (3) accountability of rulers to the ruled through mechanisms of representation and the rule of law; and (4) civilian control over the military. It is this latter dimension, so important in the Latin American context, which sets my definition apart from Robert Dahl’s classic notion of a “procedural minimum.”6 A middle-range definition of this sort avoids the Scylla of an overly narrow reliance on the mere presence of elections without concomitant changes in civil-military relations and the Charybdis of an overly broad assumption of social and economic equality. While perhaps less than fully satisfactory from a normative perspective, it has the advantage of permitting a systematic and objective investigation of the relationship between democratic political forms and the long-range pursuit of equity.


354

The Futile Search for Democratic Preconditions If the questions raised by democratization remain relatively unchanged from the past, the answers that are offered today come from a different direction. This becomes evident through a brief comparison of the divergent theories about the origins of democratic regimes that have dominated the study of Latin America. The scholarship that preceded the new wave of democratization in the 1980s argued that a number of preconditions were necessary for the emergence of a stable democratic polity. First, a certain degree of wealth or, better said, level of capitalist development was considered a prerequisite of democracy. Market economies in themselves were not enough; a country had to cross (and remain beyond) a minimum threshold of economic performance before political competition could be institutionalized. “The more well-to-do a nation,” Seymour Martin Lipset claimed, “the greater the chances that it will sustain democracy.”7 A wealthy economy made possible higher levels of literacy, education, urbanization, and mass media exposure, or so the logic went, while also providing resources to mitigate the tensions produced by political conflict.8 A second set of preconditions that underlay traditional approaches to democracy was derived from the concept of political culture, that is, the system of beliefs and values in which political action is embedded and given meaning. The prevalence of certain values and beliefs over others was said to be more conducive to the emergence of democracy. Thus, for example, Protestantism allegedly enhanced the prospects for democracy in Europe while Catholicism, with its tradition of hierarchy and intolerance, was posited to have the opposite effect in Latin America.9 Although arguments based only on the link between different religious systems and experiences with democracy have been dismissed by most scholars, more sophisticated claims sought to identify political cultures characterized by a high degree of mutual trust among members of society, a willingness to tolerate diversity, and a tradition of accommodation or compromise because such cultures were considered necessary for the subsequent development of democratic institutions. That a “civic culture” of this sort necessarily rested on a widely differentiated and articulated social structure with relatively autonomous social classes, occupational sectors, and ethnic, religious, or regional groups was an unspoken assumption. In other words, a prodemocratic consensus and set of values was considered the main prerequisite of political democracy.10 Third, specific domestic historical conditions and configurations were said to be prerequisites of democracy. Theorists of “crises and sequences” argued that the order in which various crises of modernization appeared and were settled determined whether economic and social transformations were conducive to the development of democracy. Democratic regimes were more likely to emerge if problems of national identity were resolved prior to the establishment of a central government and if both of these events preceded the formation of mass parties.11 In a different, yet still historically grounded vein, Barrington Moore, Jr. contended that democracies were more likely to appear where the social and economic power of the landed aristocracy was in decline relative to that of the bourgeoisie and where labor-


355

repressive agriculture was not the dominant mode of production. When this occurred as a result of the commercialization of agriculture that transformed a traditional peasantry into either a class of small farmers or a rural proletariat, the prognosis for democracy was strong indeed.12 A version of Moore’s approach has been used to explain the different political trajectories in Central America. Specifically, democracy is said to have emerged in Costa Rica due to the creation of a yeoman farmer class, while the persistence of authoritarian rule in Guatemala and El Salvador is attributed to the continued dominance of the landed aristocracy.13 Finally, some scholars treated external influences as another set of preconditions on the grounds that these could be decisive in determining whether a polity became democratic or authoritarian. Dependency theorists in Latin America and the United States contended that the continent’s particular insertion into the international market made democratization especially problematic at more advanced stages of import-substituting capitalist development and even enhanced the necessity for authoritarian rule under specific circumstances. In a logic that ran counter to Lipset’s “optimistic equation,” both Guillermo O’Donnell and Fernando Henrique Cardoso argued that, as dependent economies became more complex, more penetrated by foreign capital and technology, and more reliant upon low wages to maintain their competitive advantage in the international economy, professional militaries, technocrats, and state managers moved to the forefront of the decision-making process, forcibly replacing unruly, “populist” parties and trade unions in order to establish a supposedly more efficient form of rule.14 Inversely, using an argument based on external influences of a qualitatively different sort, proponents of an aggressive U.S. foreign policy towards the region declared that the rise and decline of democracy was directly related to the rise and decline of the global power of the United States rather than to market mechanisms or accumulation processes. In Samuel Huntington’s view, the dramatic increase in authoritarian rule during the 1960s and 1970s was a direct reflection of the waning of U.S. influence. Specifically, it was due to the decreased effectiveness of efforts by U.S. officials to promote democracy as a successful model of development. Concomitantly, he argued, the spate of democratic transitions in the 1980s could be credited to the Reagan administration’s renewed effort to “restore American power” through the rollback of revolutions and the promotion of electoral reforms. This position, so ideologically convenient for policymakers, located the roots of democracy outside Latin America.15 The experience of Latin American countries in the 1980s challenged all of these presumptions about preconditions. The hypothetical association between wealth and democracy might be called upon to “explain” the transition to democracy in Brazil after a protracted economic boom, but it could hardly account for the case of Peru, whose transition was characterized by stagnant growth rates, extreme foreign debt, persistent balance of payments problems, and a regressive distribution of income. Nor could it explain the anomaly of Argentina, where relatively high levels of per capita GDP were persistently accompanied by authoritarian rule. If the political cultures of Argentina, Uruguay, and Brazil all tolerated, admittedly to varying degrees, the practice of official state terror and widespread violations of human rights, how could they suddenly become sufficiently “civic” and “tolerant” to support a democratic outcome? As the Catholic church took an increasingly active role in opposing authoritarian rule, especially in


356

Brazil, Chile, Peru, Central America, and Panama, the argument about the so-called “anti-democratic bias” of Catholicism became increasingly implausible.16 The predictability of approaches emphasizing the influence of the international system fared little better. While the manner of a country’s insertion into the world capitalist economy is now considered essential in explaining its subsequent political and economic development, as dependency theorists claimed, criticisms of other scholars plus the democratic transitions in Brazil and Chile demonstrated that there was no direct or inevitable correlation between capital deepening and authoritarian rule.17 The general trends towards recession in export earnings, debt crises, diminishing U.S. support for human rights, and the frequent resort to military instruments under the foreign policy of the Reagan administration boded ill for the emergence of democracies in the 1980s, yet emerge they did. The pattern of their appearance presented an undeniable challenge to Huntington’s thesis linking democratization with the rise of U.S. power. In the southern zone, where influence from the north is not especially high, military rulers generally made way for civilian authority. In Central America, Panama, and Haiti, where the overriding historical role of the U.S. is indisputable, militaries either permitted elections to occur without limiting their own prerogatives, or they refused to leave power altogether. Indeed, where the decline in U.S. hegemony was greatest, democracy seemed to appear even though dictatorship “should” have been the more appropriate response! These anomalies suggest the pressing need for important revisions, even reversals, in the way democratization in contemporary Latin America is understood. First, there may be no single precondition that is sufficient to produce such an outcome. The search for causes rooted in economic, social, cultural/psychological, or international factors has not yielded a general law of democratization, nor is it likely to do so in the near future despite the proliferation of new cases.18 Thus, the search for a set of identical conditions that can account for the presence or absence of democratic regimes should probably be abandoned and replaced by more modest efforts to derive a contextually bounded approach to the study of democratization. Second, what the literature has considered in the past to be the preconditions of democracy may be better conceived in the future as the outcomes of democracy. Patterns of greater economic growth and more equitable income distribution, higher levels of literacy and education, and increases in social communication and media exposure may be better treated as the products of stable democratic processes rather than as the prerequisites of its existence. A “civic” political culture characterized by high levels of mutual trust, a willingness to tolerate diversity of opinion, and a propensity for accommodation and compromise could be the result of the protracted functioning of democratic institutions that generate appropriate values and beliefs rather than a set of cultural obstacles that must be initially overcome. There is evidence for this contention in the fact that most democracies in Europe and Latin America’s oldest democracy in Costa Rica have emerged from quite “uncivic” warfare. In other words, what have been emphasized as independent variables in the past might be more fruitfully conceived as dependent variables in the future.

From Contingent Choice to Structured Contingency


357

The failure to identify clear prerequisites, plus the hunch that much of what had been thought to produce democracy should be considered as its product, has caused theorists of comparative politics to shift their attention to the strategic calculations, unfolding processes, and sequential patterns that are involved in moving from one type of political regime to another, especially under conditions of nonviolence, gradualism, and social continuity. For Guillermo O’Donnell and Philippe Schmitter, democratization is understood as a historical process with analytically distinct, if empirically overlapping, stages of transition consolidation, persistence, and eventual deconsolidation.19 A variety of actors with different followings, preferences, calculations, resources, and time horizons come to the fore during these successive stages. For example, elite factions and social movements seem to play the key roles in bringing about the demise of authoritarian rule; political parties move to center stage during the transition itself; and business associations, trade unions, and state agencies become major determinants of the type of democracy that is eventually consolidated.20 What differentiates these stages above all, as Adam Przeworski points out, is the degree of uncertainty which prevails at each moment. During regime transitions, all political calculations and interactions are highly uncertain. Actors find it difficult to know what their interests are, who their supporters will be, and which groups will be their allies or opponents. The armed forces and the civilian supporters of the incumbent authoritarian regime are characteristically divided between “hard-line” and “soft-line” factions. Political parties emerge as privileged in this context because, despite their divisions over strategies and their uncertainties about partisan identities, the logic of electoral competition focuses public attention on them and compels them to appeal to the widest possible clientele. The only certainty is that “founding elections” will eliminate those who make important miscalculations. The absence of predictable “rules of the game” during a regime transition expands the boundaries of contingent choice. Indeed, the dynamics of the transition revolve around strategic interactions and tentative arrangements between actors with uncertain power resources aimed at defining who will legitimately be entitled to play in the political game, what criteria will determine the winners and losers, and what limits will be placed on the issues at stake. From this perspective, regime consolidation occurs when contending social classes and political groups come to accept some set of formal rules or informal understandings that determine “who gets what, where, when, and how” from politics. In so doing, they settle into predictable positions and legitimate behaviors by competing according to mutually acceptable rules. Electoral outcomes may still be uncertain with regard to person or party, but in consolidated democracies they are firmly surrounded by normative limits and established patterns of power distribution. The notion of contingency (meaning that outcomes depend less on objective conditions than subjective rules surrounding strategic choice) has the advantage of stressing collective decisions and political interactions that have largely been underemphasized in the search for preconditions. But this understanding of democracy has the danger of descending into excessive voluntarism if it is not explicitly placed within a framework of structural-historical constraints. Even in the midst of the tremendous uncertainty provoked by a regime transition, where constraints appear to be most relaxed and where a wide range of outcomes appears to be possible, the decisions made by various actors


358

respond to and are conditioned by the types of socioeconomic structures and political institutions already present. These can be decisive in that they may either restrict or enhance the options available to different political actors attempting to construct democracy. For example, certain social structures seem to make the emergence of political democracy highly improbable; inversely, it is reasonable to presume that their absence may make accommodative strategies more viable and reinforce the position of democratic actors. Political democracies have lasted only in countries where the landed class, generally the most recalcitrant of interests, has played a secondary role in the export economy, for example Venezuela and Chile, or where non-labor-repressive agriculture has predominated, for example Costa Rica, Argentina, and Uruguay. Thus the survivability of political democracy does seem to depend on a structural space defined in part by the absence of a strong landowner elite engaged in labor-repressive agriculture or its subordination to interests tied to other economic activities.21 The cases of Venezuela and Chile better make the point. In Venezuela, dependence upon petroleum as the leading source of foreign exchange had the (unintended) effect of hastening the decline of that country’s already stagnant agriculture and, with it, the landowning elite. Faced with overvalued exchange rates that hurt agro-exports and abundant foreign reserves for importing cheap foodstuffs, landowners sold their property to oil companies and converted themselves into a commercial and financial urban bourgeoisie. This largely voluntary self-liquidation removed the incentive for them to commercialize rural areas, to subordinate the peasantry through repressive means, and eventually to maintain authoritarian rule. It also removed the social base for an antisystem party of the right. Thus, actors designing pact-making strategies in Venezuela during the regime transition in 1958 did not face powerfully organized antidemocratic rural elites.22 Social dynamics in Chile, though different, had the same effect. Conservative elements based in a system of labor-repressive agriculture eventually supported the expansion of the suffrage in the nineteenth century as a means of combating the rising power of industrialists and capas medias, who were tied to the state and supported by revenues from copper.23 In effect, the social impact of the dominant presence of mineral exports meant that, when compared to the cases of Central America, both Venezuela and Chile were able to institutionalize democratic agreements with relative ease. These cases illustrate the limits, as well as the opportunities, that social structures place upon contingent choice. If the focus in explaining the emergence of democracy had been solely on the forging of institutional compromises, that is, conceptualizing the establishment of democracy as only the product of strategic interactions, the pact-making that characterized the Venezuelan transition and the gradual expansion of the suffrage in Chile would appear to be simply the result of skilful bargaining by astute political leaders.24 Instead, by focusing on the internal social dynamics produced by a mineralbased insertion into the international economy, it becomes evident how oil- or copperinduced structural change makes such “statecraft” possible. This is not to argue that individual decisions made at particular points in time or all observable political outcomes can he specifically and neatly linked to preexisting structures, but it is claimed that historically created structures, while not determining which one of a limited set of


359

alternatives political actors may choose, are “confining conditions” that restrict or in some cases enhance the choices available to them. In other words, structural and institutional constraints determine the range of options available to decision makers and may even predispose them to choose a specific option. What is called for, then, is a path-dependent approach which clarifies how broad structural changes shape particular regime transitions in ways that may be especially conducive to (or especially obstructive of) democratization. This needs to be combined with an analysis of how such structural changes become embodied in political institutions and rules which subsequently mold the preferences and capacities of individuals during and after regime changes. In this way, it should be possible to demonstrate how the range of options available to decision makers at a given point in time is a function of structures put in place in an earlier period and, concomitantly, how such decisions are conditioned by institutions established in the past. The advantages of this method are evident when compared to a structural approach alone, which leads to excessively deterministic conclusions about the origins and prospects of democracy, or to a sole focus on contingency, which produces overly voluntaristic interpretations.25 Modes of Transition to Democracy Once the links between structures, institutions, and contingent choice are articulated, it becomes apparent that the arrangements made by key political actors during a regime transition establish new rules, roles, and behavioral patterns which may or may not represent an important rupture with the past, These, in turn, eventually become the institutions shaping the prospects for regime consolidation in the future. Electoral laws, once adopted, encourage some interests to enter the political arena and discourage others. Certain models of economic development, once initiated through some form of compromise between capital and labor, systematically favor some groups over others in patterns that become difficult to change. Accords between political parties and the armed forces set out the initial parameters of civilian and military spheres. Thus, what at the time may appear to be temporary agreements often become persistent barriers to change, barriers that can even scar a new regime with a permanent “birth defect.” These observations have important implications for studying democracy in Latin America. Rather than engage in what may be a futile search for new preconditions, they suggest that scholars would do well to concentrate on several tasks: (1) clarifying how the mode of regime transition (itself conditioned by the breakdown of authoritarian rule) sets the context within which strategic interactions can take place; (2) examining how these interactions, in turn, help to determine whether political democracy will emerge and survive; and (3) analyzing what type of democracy will eventually be institutionalized. Thus, it is important to begin to distinguish between possible modes of transition to democracy. First, we can differentiate cases in which democracies are the outcome of a strategy based primarily on overt force from those in which democracies arise from compromise. This has been displayed on the horizontal axis in Figure 1. Second, we can distinguish between transitions in which incumbent ruling groups, no matter how weakened, are still ascendant in relation to mass actors and those in which mass actors


360

have gained the upper hand, even temporarily, vis-à-vis those dominant elites. This can be seen on the vertical axis in Figure 1. The cross tabulation of these distinctions produces four ideal types of democratic transition: reform, revolution, imposition, and pact. Latin America, at one time or another, has experienced all four modes of transition. To date, however no stable political democracy has resulted from regime transitions in which mass actors have gained control, even momentarily, over traditional ruling classes. Efforts at reform from below, which have been characterized by unrestricted contestation and participation, have met with subversive opposition from unsuppressed traditional elites, as the cases of Argentina (1946–1951), Guatemala (1946–1954), and Chile (1970– 1973) demonstrate.26 Revolutions generally produce stable forms of governance (Bolivia is an obvious exception), but such forms have not yet evolved into democratic patterns of fair competition, unrestricted contestation, rotation in power, and free associability, although developments in Nicaragua and Mexico may soon challenge this assertion.27

Figure 1: Modes of Transition to Democracy STRATEGIES OF TRANSITION Compromise Force Elite Ascendant PACT IMPOSITION RELATIVE ACTOR STRENGTH Mass Ascendant REFORM REVOLUTION Thus far, the most frequently encountered types of transition, and the ones which have most often resulted in the implantation of a political democracy, are “transitions from above.” Here traditional rulers remain in control, even if pressured from below, and successfully use strategies of either compromise or force—or some mix of the two—to retain at least part of their power.

Figure 2: Modes of Transition to Democracy in Latin America Compromise..............................................................................................Force PACT IMPOSITION Venezuela (1958-) Costa Rica (1948) Brazil (1947-) Colombia (1958-) Ecuador (1976-) Mexico (1988-)* Uruguay (1984-) Chile (1932–1970) Guatemala (1946–1954) El Salvador (1982)* Chile (1988-) Peru (1978-) Venezuela (1945–1948) Argentina (1983-) REFORM REVOLUTION Argentina (1946–1951), Mexico (1910–1929)* Guatemala (1984-)* Bolivia (1952-) Nicaragua (1979-)* Chile (1970–1973)**


361

* These cases cannot be considered democracies in the definition used here. They are included because they are in periods of transformation and thus illustrate possible modes of transition in the future. ** See footnote 26. Of these two modes of transition, democratization by pure imposition is the least common in Latin America—unless we incorporate cases in which force or the threat of force is applied by foreign as well as domestic actors. This is not the case for both Europe and Asia, where democratization through imposition often followed in the wake of World War II. In Figure 2, the cell labeled imposition includes Brazil and Ecuador, where the military used its dominant position to establish unilaterally the rules for civilian governance. Cases on the margin include Costa Rica (where in 1948 an opposition party militarily defeated the governing party but then participated in pact-making to lay the foundation for stable democratic rule), Venezuela (1945–48) and Peru (where the military’s control over the timing and shape of the transition was strongly influenced by a mass popular movement),28 and Chile (where the military’s unilateralism was curbed somewhat by its defeat in the 1988 plebiscite).29 Where democracies that have endured for a respectable length of time appear to cluster is in the cell defined by relatively strong elite actors who engage in strategies of compromise, as Figure 2 demonstrates. This cell includes the cases of Venezuela (1958-), Colombia (1958-), the recent redemocratization in Uruguay (1984-), and Chile (1932– 1970).30 What unites all of these diverse cases, except Chile, is the presence of foundational pacts, that is, explicit (though not always public) agreements between contending actors, which define the rules of governance on the basis of mutual guarantees for the “vital interests” of those involved. Chile appears to be an exception because there was no explicit pact or agreement among elites in 1932, when the democratic regime was simply “restored” on the basis of preexisting constitutional rules left over from the first democratic transition in 1874. While the Chilean case suggests that elite-based democracies can be established in the absence of foundational pacts, this may be more difficult in the contemporary period, which is characterized by more developed organized interests, the presence of mass politics, stronger military capabilities, and a tighter integration into the international market. Under such conditions, pactismo may prove to be essential.31 Foundational pacts are well exemplified by the case of Venezuela. Here a series of agreements negotiated by the military, economic, and party leaders rested on explicit institutional arrangements.32 The military agreed to leave power and to accept a new role as an “apolitical, obedient, and nondeliberative body” in exchange for an amnesty for abuses committed during authoritarian rule and a guaranteed improvement of the economic situation of officers. Political parties agreed to respect the electoral process and share power in a manner commensurate with the voting results. They also accepted a “prolonged political truce” aimed at depersonalizing debate and facilitating consultation and coalitions. Capitalists agreed to accept legal trade unions and collective bargaining in exchange for significant state subsidies, guarantees against expropriation or socializing property, and promises of labor peace from workers’ representatives. This arrangement changed what could have become potentially explosive issues of national debate into


362

established parameters by removing them from the electoral arena. The foundational pacts underlying some new democracies have several essential components. First, they are necessarily comprehensive and inclusive of virtually all politically significant actors. Indeed, because pacts are negotiated compromises in which contending forces agree to forego their capacity to harm each other by extending guarantees not to threaten each other’s vital interests, they are successful only when they include all significantly threatening interests. Thus, the typical foundational pact is actually a series of agreements that are interlocking and dependent upon each other; it necessarily includes an agreement between the military and civilians over the conditions for establishing civilian rule, an agreement between political parties to compete under the new rules of governance, and a “social contract” between state agencies, business associations, and trade unions regarding property rights, market arrangements, and the distribution of benefits. Second, while such pacts are both substantive (about the main tenets of policy) and procedural (about the rules of policymaking), they initially emphasize rulemaking because “bargaining about bargaining” is the first and most important stage in the process of compromise. Only after all contending forces have agreed to bargain over their differences can the power-sharing which leads to consensual governance result. This initial bargain can begin to lay the basis for mutual trust if only by building up reserves of familiarity between opposing groups. Subsequently, the very decision to enter into a pact can create a habit of pact making and an accommodative political style based on a “pact to make pacts.” Such foundational pacts must be differentiated from smaller, more partial “managerial” accords.33 These include the neofunctional arrangements frequently found in social democratic polities in Europe, for example, the annual corporatist negotiations among capital, labor, and the state in postwar Austria for setting wages and social policy, as well as the frequent mini-accords hammered out between political opponents in Latin America. Unlike foundational pacts, managerial accords are partial rather than comprehensive, exclusionary rather than inclusionary, and substantively oriented rather than rule making in content. These characteristics of comprehensiveness, inclusion, and rule making are critical in identifying the presence of a foundational pact. They help distinguish between basic agreements, like those present in Venezuela in 1958, and more transitory political deals, like the Pact of Apaneca which was forged in El Salvador in 1983 between the Christian Democratic Party and ARENA.34 Finally, these pacts serve to ensure survivability because, although they are inclusionary, they are simultaneously aimed at restricting the scope of representation in order to reassure traditional dominant classes that their vital interests will be respected. In essence, they are antidemocratic mechanisms, bargained by elites, which seek to create a deliberate socioeconomic and political contract that demobilizes emerging mass actors while delineating the extent to which all actors can participate or wield power in the future. They may accomplish this task by restricting contestation (as Colombian parties did in 1958 by agreeing to alternate in power regardless of the outcome of elections), by restricting the policy agenda itself (as Venezuelan parties did in 1958 by agreeing to implement the same economic program), or by restricting the franchise (as Chilean elites did beginning with the electoral law of 1874). Regardless of which strategic option is


363

chosen, the net effect of these options is the same: the nature and parameters of the initial democracy that results is markedly circumscribed.

Types of Democracies and Their Prospects in the Contemporary Period What are the implications of this excursus into preconditions and modes of transition for the prospects of democratization in contemporary Latin America? To begin with, the notion of unfolding processes and sequences from regime breakdown to transition to consolidation and persistence is fundamental in understanding the two concurrent realities of democratization in Latin America today. On the one hand, most of the newly emergent civilian or militarized civilian regimes—Argentina, Chile, Peru, Ecuador, Guatemala, Honduras, El Salvador, and Nicaragua—face the overwhelming problem of sheer survivability. What threatens their survival is the omnipresent specter of a military coup, a coup which may be provoked by intense partisan political disagreements, by the inability of political parties to manage the current profound economic crisis of the region, by the actions of antisystem elites, by a mass mobilization of labor, peasants, or the urban poor that escapes the control of traditional dominant classes, by the actions of a foreign power, or by threats to the vital corporate interests of the military itself. Significant uncertainty over the rules of the game still prevail in these fragile democracies. What becomes important in maintaining civilian rule is to find mechanisms—other than rigged or unpredictable elections—that can limit this uncertainty, especially by reducing incentives for civilians on the losing end to appeal to the military for salvation. This suggests that there are two critical tasks initially facing Latin American democratizers: first, to arrive at a sufficiently strong consensus about the rules of the game (including institutional formalities guaranteeing respect for certain crucial but minoritarian concerns) so that no major elite is tempted to call upon the military to protect its vital interests and, second, to begin to design conscious strategies for the establishment of qualitatively new civil-military relations appropriate to future stable civilian rule. This is probably easier to accomplish in the more developed regions of the continent, where the armed forces have learned the importance of cooperating with capitalist and managerial elites, than in the less developed ones (Bolivia, Central America, and the Caribbean), where the military still retains relatively confident notions of its ability to manage the economy and polity or is simply too corrupt to worry about such matters.35 On the other hand, other types of democracies in the region—Venezuela, Costa Rica, and, more recently, Brazil and Uruguay—are relatively consolidated in that actors are not so preoccupied by the overriding concern with survivability. Rather, the challenge that confronts most of these polities (and that will certainly confront newer democracies as preoccupation with mere survivability recedes) is providing some new and better resolution to the ancient question of cui bono. This issue of “who benefits” from democracy is singularly problematic in Latin America, where the pattern of dependent capitalist development has been especially ruthless in its historic patterns of exploitation.36 This means that the extension of citizenship and equal political rights must take place in a context of extreme inequality, which is unparalleled even in Africa or


364

Asia.37 It must also take place during la decada perdida, that is to say, in the midst of the most severe and prolonged economic crisis since the Depression.38 The relationship between the problematics of survivability and cui bono may well represent the central dilemma of democratization in Latin America. The choices taken by key political actors to ensure the survivability of a fragile democracy—the compromises they make, the agreements they enter into—will necessarily and even irrevocably affect who gains and who loses during the consolidation of a new regime. Subsequent “populist” decisions to redistribute gains without regard for losses may affect the durability of the regime itself, regardless of how consolidated it may appear to be. At the same time, decisions not to redistribute or inaction on this front may also influence regime durability because the commitment to democracy in part rests on the widely held (if sometimes inaccurate) conviction that economic benefits will be more fairly distributed or the welfare of the general population improved under this type of polity. Hence the current concern with both survivability and “who benefits” merely underlines the significance of choices made during the founding moments of democracies and highlights some potential relationships between political democracy and economic outcomes for future research. It also produces some not-so-promising scenarios for the emergence of different types of democracies. First, political democracy in Latin America may be rooted in a fundamental paradox: the very modes of transition that appear to enhance initial survivability by limiting unpredictability may preclude the future democratic self-transformation of the economy or polity further down the road. Ironically, the conditions that permit democracies to persist in the short run may constrain their potential for resolving the enormous problems of poverty and inequality that continue to characterize the continent. Indeed, it is reasonable to hypothesize that what occurs in the phase of transition or early consolidation may involve a significant trade-off between some form of political democracy, on the one hand, and equity, on the other. Thus, even as these democracies guarantee a greater respect for law and human dignity when compared to their authoritarian predecessors, they may be unable to carry out substantive reforms that address the lot of their poorest citizens. If this scenario should occur, they would become the victims of their successful consolidation, and the democratic transitions of the 1980s that survive could prove to be the “frozen” democracies of the 1990s . Second, while this may be the central dilemma of elite-ascendant processes of democratization, there may be important differences between countries like Uruguay, a pacted transition, and Brazil, a unilaterally imposed transition. Pacted democracies, whatever their defects, have been honed through compromise between at least two powerful contending elites. Thus, their institutions should reflect some flexibility for future bargaining and revision over existing rules. In Uruguay, for example, while the agreed-upon rules made it very difficult to challenge agreements between the military and the parties on the issue of amnesty for crimes committed during authoritarian rule, the left opposition, excluded from this agreement, was nevertheless able to force the convocation of a plebiscite on this major issue, which it subsequently lost. It is difficult to imagine that anything similar could occur in Brazil. Because the military exerted almost complete control over the transition, it never curtailed its own prerogatives nor fully agreed to the principle of civilian control, and it has not been compelled to adopt institutional rules


365

reflecting the need for compromise. The contrast between the cases of Uruguay and Brazil raises a hypothesis that merits investigation: to the extent that transitions are unilaterally imposed by armed forces who are not compelled to enter into compromises, they threaten to evolve into civilian governments controlled by authoritarian elements who are unlikely to push for greater participation, accountability, or equity for the majority of their citizens. Paradoxically, in other words, the heritage left by “successful” authoritarian experiences, that is, those characterized by relatively moderate levels of repression and economic success which has left the military establishment relatively intact, may prove to be the major obstacle to future democratic self-transformation.39 This danger exists, albeit to a lesser extent, in civilian-directed unilateral transitions, for example, Mexico, because the institutional rules that are imposed are likely to favor incumbents and permit less scope for contestation. Third, the attempt to assess possible consequences of various modes of transition is most problematic where strong elements of imposition, compromise, and reform are simultaneously present, that is to say, where neither incumbent elites nor newly ascendant power contenders are clearly in control and where the armed forces are relatively intact. This is currently the case in Argentina and Peru, as Figure 2 demonstrates. Given the Argentinean military’s defeat in the Falklands/ Malvinas war, the high level of mass mobilization during the transition, and the absence of pacts between civilian authority and the armed forces, on the one hand, and trade unions and employers, on the other, Argentina combines elements of several modes of transition. Such a mixed scenario, while perhaps holding out the greatest hope for political democracy and economic equity, may render a consistent strategy of any type ineffectual and thus lend to the repetition of Argentina’s persistent failure to consolidate any type of regime. The prospects for failure are even greater in Peru. Given the absence of explicit agreements between the leading political parties, the possibility of mass mobilizations in the midst of economic depression, the presence of an armed insurgency, and a unified military, Peru is currently the most fragile democracy in South America. Fourth, because political democracies generally arise from a compromise between contending organized elites that are unable to impose their will unilaterally or the unilateral action of one dominant group, usually the armed forces, this does not bode well for democratization in situations in which the armed forces are inextricably tied to the interests of a dominant (and antidemocratic) agrarian class. Guatemala and El Salvador in particular are characterized by a landowning elite whose privileged position is based on labor-repressive agriculture and on a virtual partnership with the armed forces, thereby making it unlikely that their militaries (as currently constituted) will tolerate comprehensive political competitiveness, civil liberties, or accountability. Regardless of the profound differences between these two Central American countries, the extraordinary pressure of U.S. intervention as well as international diffusion means that, at minimum, they can be expected to adhere to “electoralism,” meaning the regularized holding of elections, even as they continue to restrict the other political rights and opportunities of their citizens. This hybrid mix of electoral forms and authoritarianism, which has been dubbed “electocratic rule” by one observer,40 is likely to emerge in other developing areas wherever the spread of elections under foreign inspiration either


366

precedes or is intended to coopt strong domestic pressures for democratization. These observations can be distilled into types of democracies, which, at least initially, are largely shaped by the mode of transition in Latin America, as Figure 3 illustrates. They suggest that democratization by imposition is likely to yield conservative democracies that can not or will not address equity issues. To the extent that imposition originates from outside, however, the result is likely to be some form of electoral authoritarian rule, which can not be considered democracy at all. Pacted transitions are likely to produce corporatist or consociational democracies in which party competition is regulated to varying degrees determined, in part, by the nature of foundational bargains. Transition through reform is likely to bring about competitive democracies, whose political fragility paves the way for an eventual return to authoritarianism. Finally, revolutionary transitions tend to result in one-party dominant democracies, where competition is also regulated. These types are characterized by different mixes and varying degrees of the chief dimensions of democracy: contestation, participation, accountability, and civilian control over the military. Such predictions are discouraging, but they may be offset by more hopeful observations that affect the contingent choices of contemporary democratizers. On the one hand, the Cold War features of the international system have changed remarkably, and this may offer new opportunities for the reformist mode of transition in Latin America. The failure of two of the three cases cited in this category, Guatemala (1946– 1954) and Chile (1970–1973), was profoundly affected by U.S. intervention, motivated in large pan by the ideological identification of mass-based reforms with the spread of Soviet influence in the western hemisphere. U.S. intervention against peasant-based movements in Central America has been justified in the same manner. To the extent that the global state system loses its “bipolarity,” the credibility of such accusations becomes increasingly difficult to sustain, thus potentially creating more space for mass ascendant political movements. The fact that this mode of transition failed in the past in Latin America does not mean that it will not succeed in the future.41

Figure 3: Modes of Transition and Types of Democracy


367

On the other hand, this discussion of modes of transition and varying probabilities for survival has not presumed that democracies will benefit from superior economic performance, which is fortunate given the state of contemporary Latin American economies. Most observers assume that crises in growth, employment, foreign exchange earnings, and debt repayments necessarily bode ill for the consolidation of democratic rule, and few would question the long-run value of an increasing resource base for stability. But austerity may have some perverse advantages, at least for initial survivability. In the context of the terrible economic conditions of the 1980s, the exhaustion of Utopian ideologies and even of rival policy prescriptions has become painfully evident. Neither the extreme right nor the extreme left has a plausible alternative system to offer—to themselves or to mass publics. Though populism, driven by diffuse popular expectations and desencanto with the rewards of compromised democracy, is always a possibility—witness the experience of Peru and the recent elections in Argentina—it can not deliver the immediate rewards that have been its sustenance in the past. To the extent that this situation diminishes both the expected benefits and rewards from antisystem activity, it enhances the likelihood of democracies to endure. This suggests a possible hypothesis for future exploration. The relationship between democratization and economic performance, rather than rising or falling in tandem, may be parabolic. Conditions to strike bargains may be most favorable in the midst of protracted austerity, as well as in the midst of sustained plenty. They may be worse when the economy is going through stop-and-go cycles or being hit with sudden windfalls or scarcities. If true, this provides a ray of hope for the otherwise unpromising decade ahead. Finally, there is no a priori reason why one type of democracy can not be transformed into another, that is to say, why electoral authoritarian regimes, for example, can not evolve into conservative or competitive democracies, or corporatist democracies into more competitive ones. Given the frequency of pactismo and the gravity of the equity problem in Latin America, the latter scenario is especially important. While pacted


368

transitions establish an improvisational institutional framework of governance that may become a semipermanent barrier to change, this framework is subject to further modification in the future. Such modification may be brought about preemptively when some ruling groups, having experienced the advantages of democratic rule, become more inclined over time to seek to accommodate potential pressures from below rather than suppress them, or it may occur through the direct pressure of organized social groups.42 in either case, democratization can prove to be an ongoing process of renewal. The notion that one type of democracy may gradually evolve into a qualitatively different type suggests that the dynamics of democratic consolidation must differ in important ways from the transition if “freezing” is to be avoided. Because the overriding goal of the transition is to reach some broad social consensus about the goals of society and the acceptable means to achieve them, successful transitions are necessarily characterized by accommodation and compromise: But if this emphasis on caution becomes an overriding political norm during consolidation, democracies may find it difficult to demonstrate that they are better than their predecessors at resolving fundamental social and economic problems. Thus, consolidation, if it is to be successful, should require skills and commitments from leading actors which are qualitatively different from those exhibited during the transition. In this latter phase, these actors must demonstrate the ability to differentiate political forces rather than to draw them all into a grand coalition, the capacity to define and channel competing political projects rather than seek to keep potentially divisive divisive off the agenda, and the willingness to tackle incremental reforms, especially in the domains of the economy and civil-military relations, rather than defer them to some later date. If the cycle of regime change that has plagued Latin America is to be broken and replaced by an era of protracted democratic rule, democratizers must learn to divide as well as to unite and to raise hopes as well as to dampen expectations.

Notes This article was originally presented at the Conference on Latin America at the Threshold of the 1990s, sponsored by the Institute of Latin America of the Chinese Academy of Social Sciences and the Ford Foundation and held in Beijing on Jun 8–16,1988. The author wishes to thank Ken Erickson, Richard Fagen, Samuel Valenzuela, an anonymous reviewer, and, most especially, Philippe Schmitter. 1. These questions underlie a number of new studies on democracy. See, for example, Guillermo O’Donnell, Philippe Schmitter, and Laurence Whitehead, eds., Transitions from Authoritarian Rule, 4 vols. (Baltimore: The Johns Hopkins University Press, 1986): Paul W.Drake and Eduardo Silva, eds., Elections and Democratization in Latin America, 1980– 1985 (San Diego: Center for Iberian and Latin American Studies, University of California, 1986); Enrique A.Baloyra, Comparing New Democracies: Transition and Consolidation in Mediterranean Europe and the Southern Cone (Boulder: Westview Press, 1987): Carlos Huneeus, Para Vivir La Democracia (Santiago: Editorial Andante, 1987); and Larry Diamond, Juan J.Linz, and Seymour Martin Lipset, eds., Democracy in Developing Countries, 4 vols. (Boulder: Lynne Rienner Publishers, 1988–90).


369

2. See Dankwart A.Rustow, transitions to Democracy: Towards a Dynamic Model,” Comparative Politics, 2 (April 1970). 3. This statement requires some qualification. J.A.Schumpeter defines democracy as “that institutional arrangement for arriving at political decisions in which individuals acquire the power to decide by means of a competitive struggle for the people’s vote” in Capitalism, Socialism and Democracy (London: Geo. Allen and Unwin, 1943), p. 269. Under this definition the competition for leadership through free elections is the distinctive feature of democracy. But Schumpeter, unlike Jeane Kirkpatrick and other U.S. policymakers in the 1980s, considered civil liberties a necessary condition for the operation of democracy. Thus, it can not be assumed that he would have shared the current emphasis on the mere presence of elections, which I have elsewhere referred to as “electoralism” that is, “the faith that merely holding elections will channel political action into peaceful contests among elites and accord public legitimacy to the winners in these Contests.” See Terry Lynn Karl, “Imposing Consent? Electoralism versus Democratization in El Salvador,” in Drake and Silva, eds., p. 34. 4. For an example of this approach, see Suzanne Jonas, “Elections and Transitions: The Guatemalan and Nicaraguan Case,” in John Booth and Mitchell Seligson, eds., Elections and Democracy in Central America (Chapel Hill: University of North Carolina Press, 1989). Jonas and Stein argue against separating political democracy from socioeconomic equity and support “a broader view that meaningful ‘transitions’ to democracy’ [in Central America] involve more sweeping social change on the scale of the major bourgeois and socialist revolutions historically.” See Suzanne Jonas and Nancy Stein, “Democracy in Nicaragua,’ in Suzanne Jonas and Nancy Stein, eds., Democracy in Latin America (New York: Bergin and Garvey Publishers, 1990, p. 43. 5. In examining the problem of constructing institutions that can translate the preferences of majorities into public policy, for example, social choice theorists have demonstrated the difficulty of designing decision-making procedures that give equal weight to the preferences of all citizens and that permit the aggregation of these preferences into governmental policies without violating any of the other basic tenets of democratic theory. See, for example, William H.Riker, Liberalism versus Populism: A Confrontation between the Theory of Democracy and the Theory of Social Choice (San Francisco: W.H.Freeman and Co, 1982), and the review by Jules Coleman and John Ferejohn, “Democracy and Social Choice,” Ethics, 97 (October 1986). Theorists of democracy have long grappled with other dilemmas involving notions of social justice and equity. See, for example, Peter Bachrach, The Theory of Democratic Elitism: A Critique (Washington, D.C.: University Press of America, 1980) and Carole Pateman, Participation and Democratic Theory (Cambridge: Cambridge University Press, 1970). 6. I have drawn the first two dimensions and, to some extent, the third from Robert A. Dahl Polyarchy: Participation and Opposition (New Haven: Yale University Press, 1971). But Dahl, like other democratic theorists, does not emphasize the establishment of civilian control over the military through the limitation of military prerogatives. Indeed, this dimension often appears to be an assumed condition or even an unstated prerequisite in other definitions of democracy. Alfred Stepan, Rethinking Military Politics: Brazil and the Southern Cone (Princeton: Princeton University Press, 1988), is an important corrective in this regard. Stephan defines the military’s institutional prerogatives as “those areas where whether challenged or not, the military as an institution assumes they have an acquired right or privilege, formal or informal, to exercise effective control over its internal governance, to play a role within extra-military areas within the state apparatus, or even to structure


7.

8.

9.

10.

11.

12. 13.

370

relationships between the state and political or civil society” (p. 93). The clear determination and limitation of these areas are a measure of civilian control and, in my view, are also a measure of democratization. This formulation originally appeared in Seymour Martin Lipset, “Some Social Requisites of Democracy: Economic Development and Political Legitimacy,” American Political Science Review, 53 (March 1959). Some proponents of this view often measured the prospects for democracy by per capita gross domestic product, leading the occasional political observer to await the moment when a particular country would cross “the threshold” into democracy. This supposed threshold has varied from country to country. Spain’s Lopez Redo once predicted that his country would not become democratic until it reached a per capita income of $2,000. More recently, Mitchell Seligson has argued that Central America needs to approach a per capita income of $250 (in 1957 dollars) and a literacy rate of over 50 percent as a necessary precondition for democratization. See James M.Malloy and Mitchell A.Seligson, eds., Authoritarians and Democrats: Regime Transition in Latin America (Pittsburgh: University of Pittsburgh Press, 1987), pp 7–9. For example, Howard Wiarda, “Toward a Framework for the Study of Political Change in the Iberic-Latin Tradition: The Corporative Model,” in Howard Wiarda, ed., Corporatism and National Development in Latin America (Boulder Westview, 1981), argued that Latin America possessed “a political culture and a sociopolitical order that at its core is essentially two-class, authoritarian, traditional, elitist, patrimonial, Catholic, stratified, hierarchical and corporate.” A similar argument can be found in Richard N.Morse, “The Heritage of Latin America,” in Howard Wiarda, ed., Politics and Social Change in Latin America (Amherst University of Massachusetts Press, 1974). The notion of “civic culture,” first introduced by Gabriel Almond and Sidney Verba in The Civic Culture (Princeton: Princeton University Press, 1963), sought to analyze the relationship between the political attitudes of a population and the nature of its political system It was the forerunner of the works on Latin America cited above. This was the basic argument put forward by Leonard Binder et al., eds., Crises and Sequences in Political Development (Princeton: Princeton University Press, 1971), and by Eric Nordlinger, “Political Development, Time Sequences and Rates of Change,” in Jason L.Finkle and Robert W.Gable, eds., Political Development and Social Change, 2nd ed (New York: John Wiley, 1971). See Barrington Moore, Jr, Social Origins of Dictatorship and Democracy (Boston: Beacon Press, 1966). See John Weeks, “An Interpretation of the Central American Past,” Latin American Research Review, 21 (1986); Enrique Baloyra-Herp, “Reactionary Despotism in Central America,” Journal of Latin American Studies 15 (1983) and Jeffrey Paige, “Coffee and Politics in Central America,” in Richard Tardanico, ed, Crisis in the Caribbean Basin (Beverly Hills Sage Publications, 1987). In a more recent work, Paige seeks to differentiate his argument from that of Moore. He correctly contends that there is no collision between an industrial bourgeoisie and a landed class in either Costa Rica, El Salvador, or Nicaragua and that the agrarian aristocracy has successfully transformed itself into a modern capitalist class, both conditions that belie Moore’s argument. Nonetheless, in Guatemala and El Salvador a landed class continues to exercise domination, and the commercialization of agriculture has not replaced a labor-repressive mode of production, thus providing some important confirmation of Moore. See Jeffrey Paige, “The Social Origins of Dictatorship, Democracy and Socialist Revolution in Central America,” paper presented at the Annual Meeting of the American


371

Sociological Association San Francisco, August 8, 1989. 14. See Guillermo O’Donnell, Modernization and Bureaucratic-Authoritariansim (Berkeley: University of California, Institute for International Studies, 1973), and Fernando Henrique Cardoso, “Associated-Dependent Development Theoretical and Practical Implications,” in Alfred Stephan ed, Authoritarian Brazil (New Haven: Yale University Press, 1973), pp 142– 178. 15. See Samuel P.Huntington, “Will More Countries Become Democratic?,” Political Science Quarterly, 99 (1984). 16. Furthermore, through the church’s active promotion of “base communities,” it could even be argued that contemporary Catholicism contributes to the creation of a uniquely democratic culture by encouraging participation among previously unorganized groups of the urban and rural poor. See Philip Oxhorn, “Bringing the Base Back In The Democratization of Civil Society under the Chilean Authoritarian Regime” (Ph D diss, Harvard University, 1989). 17. For criticism of the O’Donnell hypothesis linking capital deepening to authoritarian rule, see David Collier, ed., The New Authoritarianism in Latin America (Princeton: Princeton University Press, 1979), and Karen Remmer and Gilben Merkx, “BureaucraticAuthoritarianism Revisited,” Latin American Research Review, 17 (1982). 18. Albert Hirschman has even claimed that this search can be pernicious. In his view, to lay down strict preconditions for democracy—“dynamic growth must be resumed, income distribution must be improved,…political parties must show a cooperative spirit …”—may actually encourage the deconsolidation of existing democracies. Hirschman argues that this will almost certainly obstruct constructive thinking about the ways in which democracies may be formed, survive, and even become stronger in the face of and in spite of continuing adversity. See Albert Hirschman, “Dilemmas of Democratic Consolidation in Latin America,” unpublished notes for the Sao Paulo Meeting on Democratic Consolidation in Latin America and Southern Europe, 1986. 19. See especially Guillermo O’Donnell and Philippe C Schmitter, Tentative Conclusions about Uncertain Transitions (Baltimore: The Johns Hopkins University Press, 1986), Adam Przeworski, “Some Problems in the Study of the Transition to Democracy,” in O’Donnell, Schmitter, and Whitehead, eds., vol. 3, and Adam Przeworski, “Democracy as a Contingent Outcome of Conflicts,” in Rune Slagsted and Jon Elster, eds., Constitutionalism and Democracy (New York: Cambridge University Press, 1989). 20. See Philippe Schmitter, “Democratic Consolidation of Southern Europe,” unpublished manuscript. 21. Evelyne Huber Stephens makes a similar observation in “Economic Development, Social Change and Political Contestation and Inclusion in South America,” paper prepared for the Latin American Studies Association, New Orleans, 1988. 22. See Terry Lynn Karl, The Paradox of Plenty: Oil Booms and Petro-States (Berkeley: University of California Press, forthcoming), and “Petroleum and Political Pacts: The Transition to Democracy in Venezuela,: Latin American Research Review, 22 (1986). 23. See Arturo Valenzuela and Samuel Valenzuela, “Los Origines de la Democracia: Reflexiones Teoricas sobre el Caso de Chile,” Estudios Publicos, 12 (Spring 1983). 24. This is the general thrust of Daniel Levine’ s analysis of Venezuela, which attributes the emergence of a democratic regime primarily to statecraft and the ability of political actors to compromise. See Daniel Levine, “Venezuela since 1958: The Consolidation of Democratic Politics,” in Juan Linz and Alfred Stepan eds., The Breakdown of Democratic Regimes: Latin America (Baltimore: The Johns Hopkins University Press, 1978). 25. An approach of this sort treats regime changes as critical junctures and carries an implicit


26.

27.

28.

29.

30.

372

assumption of patterns of political change characterized by gradualism punctuated by sharp discontinuities. It has a long tradition in the study of politics, but it is especially important in recent work on the “new institutionalism,” See, for example, J.G.March and J.P.Olson, “The New Institutionalism: Organizational Factors in Political Life,” American Political Science Review 78 (September 1984), 734–749, and Stephen D.Krasner. “Sovereignty An Institutional Perspective,” Comparative Political Studies 21 (April 1988), 66–94. Krasner, though emphasizing political institutions alone rather than the combination of social structures and institutions, also argues that institutions established in the past constrain present choices, that the preferences of individual actors are conditioned by institutional structures, and that historical trajectories are path-dependent adopts a similar framework. Ruth Berins Collier and David Collier, Shaping the Political Arena: Critical Junctures, the Labor Movement and Regime Dynamics in Latin America (Princeton: Princeton University Press, forthcoming), is the most ambitious effort to utilize this sort of path-dependent approach. In their comparative analysis, they examine the different trajectories that result from the initial patterns of incorporation of the labor movement into political life. Strictly speaking, the case of Chile from 1970 to 1973 is not an effort of regime transition from authoritarian rule in the sense considered here. Rather, it is better understood as an attempt to move from one type of democracy to another, that is, a move down the vertical scale of the classification scheme in Figure 1 towards a reformist democracy. There are interesting moves in this direction in the processes taking place in both Nicaragua and Mexico. Nicaragua is the first revolutionary regime on the continent to hold national elections in which a number of political parties have been able to complete. In 1984, the Traditional Liberal and Conservative parties and several small leftist parties competed with the FSLN and won almost 35 percent of the vote. In 1990, the UNO, a coalition of fourteen anti-Sandínista parties, defeated the Sandinistas, who promised to respect the mandate of the electorate. In Mexico, the PRI has begun to permit greater contestation at the municipal and regional level, but these elections are still characterized by numerous restrictions, fraud, and localized violence. There is little information on the dynamics of regime transition in Costa Rica. See Jacobo Schifter, La fase oculta de la guerra civil en Cost Rica (San Jose: EDUCA, 1979), and Fabrice Edouard Lehoucq, “Explaining the origins of Democratic Regimes: Costa Rica in Theoretical Perspective” (Ph.D.diss., Duke University, forthcoming), which applies the notion of democracy as a contingent institutional compromise to this case. On the transition in Peru, see Cynthia Sanborn, “Social Democracy and the Persistence of Populism in Peru” (Ph.D.diss., Harvard University, forthcoming). Even where the military retained control over the transition, however, it systematically engaged in a process of consultation with civilian parties. See Anita Isaacs, “The Obstacles to Democratic Consolidation in Ecuador,” paper presented to the Latin American Studies Association, San Juan, Puerto Rico, September 21–23, 1989; Francis Hagopian and Scott Mainwaring, “Democracy in Brazil: Origins, Problems and Prospects,” World Policy Journal (Summer 1987), 485–514; and Manuel Antonio Garreton, “El Plebiscito de 1988 y la transicion a la democracia” (Santiago: FLACSO, 1988). On these cases, see Charles G.Gillespie, “Uruguay’s Transition from Collegial MilitaryTechnocratic Rule,” in O’Donnell, Schmitter. and Whitehead, eds.; Jonathan Hartlyn, “Democracy in Colombia: The Politics of Violence and Accommodation,” in Diamond, Linz, and Lipset eds., vol. 4; Alexander W.Wilde, “Conversations among Gentlemen: Oligarchical Democracy in Colombia,” in Linz and Stepan, eds., The Breakdown of Democratic Regimes: Karl, “Petroleum and Political Pacts.”


373

31. I am grateful to Samuel Valenzuela for this point. See Samuel Valenzuela, Democratizacion via Reforma: La Expansion del Sufragio en Chile (Buenos Aires: Ediciones IDES, 1985). 32. The roots of these arrangements can be found in the Pacto de Punto Fijo and the Declaration de Principios y Programa Minimo de Gobierno which were signed prior to the country’s first elections by all contending presidential candidates. These agreements bound all signatories to the same basic political and economic program regardless of the electoral outcome. These pacts are described more fully in Karl, “Petroleum and Political Pacts.” 33. This distinction was originally drawn by Philippe Schmitter in a conference on “MicroFoundations of Democracy,” University of Chicago, March 1988. 34. This agreement served primarily as a mechanism for partitioning state offices and establishing other temporary forms of power-sharing. Because it excluded powerful, wellorganized forces on the left and was never aimed at establishing permanent rules of the game, it does not meet the criteria for a foundational pact. 35. I am grateful to an anonymous reviewer for this observation. 36. Most observers locate the roots of this exploitation in colonial and postcolonial landholding patterns that, slowly or abruptly, concentrated property ownership and dispossessed the majority. Specific social processes not conducive to demoralization accompanied these landholding patterns. For example, unlike the reciprocity and community at the local level, the penetration of capitalism altered traditional clientelist relations between landlords and peasants in Latin America from a two-way to a one-way affair. As Paul Harrison, Inside the Third World: The Anatomy of Poverty (London: Penguin Books, 1979), p. 105, remarks, “in Latin America the peasant has only duties, the landowner rights.” Such social relations have left little residue of notions of mutual obligation or reciprocity between the rich and the poor. 37. I am referring to indicators of inequality here, not absolute poverty. While most of southern Asia and Africa is far poorer than Latin America, their colonial past, patterns of land tenure, and relations of production are quite different. Parts of Asia that have experienced capitalist commercialization of agriculture are now beginning to approximate these same indicators of inequality, but\Asia in general has not reached the regional scale of inequality that marks Latin America. 38. One statistic eloquently demonstrates the depth of the crisis. By 1987, Latin America’s debt represented 46 percent of the region’s GNP and more than four times the value of its exports. See IDB, Economic and Social Progress in Latin America: 1988 Report (Washington: Inter-American Development Bank, 1988), p. 541. 39. The notion that especially “successful” authoritarian regimes paradoxically may pose important obstacles for democratization can be found in Anita Isaacs, “Dancing with the People: The Politics of Military Rule in Ecuador, 1972–1979: (Ph.D.diss., Oxford University, 1986), and Guillermo O’Donnell, “Challenges to Democratization in Brazil,” World Policy Journal 5 (Spring 1988), 281–300. 40. I am grateful to Charles Call for this label. 41. There are important differences here, however, between South America and the Caribbean basin. Military interventions, which have been confined to this latter region in the past, predated the Cold War and are likely to continue after its demise. As the case of Panama shows, the rationale may simply change. 42. Paul Cammack has argued that a ruling coalition might make strategic concessions in its own long-term interest to help sustain democracy, especially after having experienced the failure of militaries to act as reliable allies. See Paul Cammack, “Democratization: A Review of the Issues,” Bulletin of Latin American Research, 4(1985),39–46.There seems to be little evidence for this predicted behavior in the current period, however, and further


374

democratization through mass pressure seems to be more likely.

SECTION FOUR: Welfare States


Commentary The first two articles in this section are by scholars from Scandinavia whose long familiarity with the welfare state has enabled them to advance research beyond previous stereotypes, it may well be that we begin to understand the nature of welfare best at the precise moment that it is under attack. W.Korpi’s ‘Social Policy and Distributional Conflict in Capitalist Democracies: A Preliminary Comparative Framework’ presents a general, comparative model for the understanding of social policy strategies. What matters most, according to this model, is the basic distributional arrangements that have been made in societies as the result of their patterns of industrialisation and state formation. In this connection, Korpi finds the key variable to be the way in which working classes have been incorporated into politics in the past. Welfare is here seen very much in terms of pressure from below, and pressure that comes from a particular class. G.Esping-Anderson’s The Three Worlds of Welfare Capitalism’ also offers a general model, and it is one which is made even more sophisticated by its possession of an altogether wider definition of welfare. What matters quite as much to Esping-Anderson as policies designed to help those who suffer from the irregularities of capitalism are transfer payments which help established payments. In this connection, he distinguishes between three types of welfare regime. The Scandinavian system is universal, and has as a great achievement the ability to keep unemployment low. The German system gains popularity because it looks after the middle classes as does that of Scandinavia, but it differs from it in one key respect: it does not help women enter the labour force to anywhere near the same extent. In contrast to both of these is the Anglo-Saxon model which only helps the socially disadvantaged, and thereby leaves itself open to welfare backlash. A.Orloff’s ‘The Political Origins of America’s Belated Welfare State’ stands neatly between this section and the next, since it considers why America welfare policies were different from those of Europe. Very importantly in the context of this collection is the fact that there was a good deal of welfare provision early on, and that it was occasioned by the state—in order to care for civil war veterans. At later stages in American history, most notably in the Progressive Era before the First World War and during Roosevelt’s New Deal, there has been renewed pressure from below for welfare provision. But Orloff nicely demonstrates that pressure from below is not in itself enough, thereby adding to Korpi’s original model. Cross class alliances are needed to make welfare secure, and the legacy of a corrupt state in America meant that the middle classes were reluctant to allow the state to undertake more tasks. In addition, the American system pays more attention to law than to state policy, thereby placing innovative welfare policies at a discount.



379

54 Social Policy and Distributional Conflict in the Capitalist Democracies. A Preliminary Comparative Framework* Walter Korpi

* Source: West European Politics, 1980, vol. 3, pp. 296–316.

The purpose of this paper is to take preliminary steps towards the development of a theoretical framework for the comparison of social policies in the industrialized nations, which combine a capitalistic economic system with political democracy. It is a part of a larger research project attempting to compare social policies in these countries against the background of the structure, organisation and relative power of their main classes and the role of shifting distributions of power between classes for the patterns and forms of class conflict in these societies. The Scandinavian nations are often described as welfare states. Sweden, in particular, is often seen as the model of the welfare state. But the concept of the welfare state is a rather vague one. It is also used to refer to societies as diverse as Germany in the days of Bismarck and contemporary United States. In fact, most of the industrialized western countries are often characterized as welfare states. Generally, the average standard of living and the role of the state in safeguarding minimum standards of living in the population appear to have been taken as criteria for such characterizations.1 The under-defined concept of the welfare state, however, has not hindered research efforts to explain cross-national differences in the degree of ‘welfare stateness’. Researchers have tended to resolve the Gordian knot of the welfare state concept by applying relatively narrow operational definitions in terms of internationally available statistical indicators, often the measures of public expenditures on certain social security programs published by the ILO.2 However, also broader indicators have been used, e.g. the degree of social mobility.3 A central issue in these studies has been the relative role of parliamentary politics in accounting for differences between welfare states. They have departed from differing theoretical starting-points, most of them in mainstream social science, a few from some varieties of Marxian theory. In spite of this theoretical diversity their conclusions rather heavily point in one direction—the political composition of parliaments and governments in the western nations and especially the relative strength of social democratic parties in these nations has been of little or no significance for the extent and ways in which the welfare state has developed.4 In these studies, functional necessities in terms of the requirements of industrialism and the composition of the population rather than political conflicts appear to emerge as the main determinants of the extent and form of the welfare

Social Policy and Distributional Conflict in the Capitalist Democracies.

380

state in the western nations. Some studies, however, have questioned or qualified this interpretation.5 In this paper I will attempt to explore the ways in which social policies can vary between nations and to examine some dimensions of social policies which can be expected to reflect political influences. The concept of social policy is here used in a comprehensive way to include, in principle, all the ways in which the state enters into the distributive processes of the capitalist democracies.

Distributive Processes In order to analyze the ways in which state measures affect distributive processes in the capitalist democracies, we must have a theoretical conception of what these processes are and how they function. These distributive processes can be analyzed in terms of their outcomes, i.e. the variation in the degree of inequality which is the assumed result of the distributive process. The distributive processes can also be analyzed in terms of the conditions within which they work as well as in terms of the intervening processes that are assumed to preceed the outcomes, i.e. inequality. Within the social sciences there .are a number of partly competing, partly overlapping theoretical approaches in the analysis of distribution and inequality in western societies. These theoretical approaches appear to differ primarily in the way in which power resources are seen as influencing the degree of inequality. In mainstream social science the ‘pluralistic’ conception prevails. This means that power resources in the advanced western societies are seen as being relatively equally distributed between multiple groups or collectivities in these societies. Because of their even and stable distribution, power resources are therefore not seen as being of major relevance for inequality and changes in the patterns of inequality. Within economics the two dominant theoretical approaches, the marginal productivity theory and human capital theory, pay little attention to power. Some of the recent attempts to modify and to find alternatives to marginal productivity theory, primarily those based on the concepts of ‘job competition’ and ‘split’ or ‘dual’ labour markets accord a somewhat larger role to power.6 Within sociology the dominant structural-functional theory of stratification has been based on ideas similar to those that later were incorporated into human capital theory. Some mainstream sociological theories, however, have given power a central place in the explanation of inequality.7 Some approaches to distribution inspired by Marxian theory, especially those of the Leninist variety, assume on the contrary that power resources in capitalist societies are basically distributed in a dichotomous pattern, so that the capitalist class controls practically all resources while little of interest is left to the wage earners. Because of this highly uneven but stable distribution of power resources, also in this type of approach the distribution of power resources is seen as being of little significance for inequality. The present research project starts from a differing theoretical approach to the relationship between power resources and inequality. Instead of a relatively stable, pluralistic or dichotomous, distribution of power resources we want to explore the hypothesis that the distribution of power resources in the capitalist democracies can differ between nations and also change over time. Control over power resources is hypothesized


381

to be of central importance for inequality. Thus we hypothesize that in the capitalist democracies the control over power resources is a major factor affecting the functioning of the distributive processes within society, what the outcomes of these processes are and how distributional conflict is patterned. The major power resources in the capitalist democracies are assumed to be related to their class structures. This gives us basically two broad types of power resources, on the one hand, power resources based on the control over the means of production and, on the other hand, power resources based in organizations which can coordinate the ‘human capital’ resources of individual wage-earners into collective action, that is primarily unions and political parties. In concrete cases, however, this basic distribution is, of course, complemented and complicated by hosts of other factors.

Figure 1: Rough Sketch of Hypothesized Distributional Process in the Capitalist Democracies

In our preliminary theoretical outline, the variations in the difference between the two major types of power resources—control over the means of production and the organization of wage earners into unions and political parties—are thus assumed to be of major importance for the forms and functioning of distributive processes in the capitalist democracies including the extent and forms in which conflicts over distribution become manifest. We assume also that the interventions of the state in the distributive processes reflect variations in the distribution of these major power resources in society. Of course other factors, such as business cycles and the international situation, must be assumed to affect the actions of the state as well as the distributive processes.8 The outcomes of the distributive processes can be seen as a multi-dimensional pattern of inequality in the distribution of welfare or levels of living including income, health, education and housing. In a very rough way the basic ideas of this tentative distributional theory can be sketched as in Figure 1.

The Welfare State and Equality


382

The welfare state can be defined in a number of ways. Some of the aspects which we would like to take into consideration can be seen as conditions for the welfare state, such as a relatively well-functioning political democracy and a relatively high average standard of living. When trying to specify the ‘essence’ of the welfare state, however, it is difficult to avoid the variable of equality-inequality in terms of standards and conditions of living among citizens. Even if we focus on what appears to be a more limited aspect, the absence of poverty among the citizens, we are implying a definition in terms of equality. This is because most western governments in practice apply relative rather than absolute definitions of poverty. Poverty is no longer defined in terms of some absolute standard of living related to nutrition as Rowntree once suggested. Instead, it is defined in terms of the position of the worst-off citizens in relation to the average condition among the citizens. Among the capitalist democracies the official use of an absolute definition of poverty appears to be limited to the United States. On the theoretical level, however, American spokesmen also admit that the relative definition is to be preferred.9 A tentative definition of the welfare state would thus have to be based on the extent to which equality in basic conditions of living has been achieved among the citizens. Because of this reliance on equality such a definition is likely to be controversial. A somewhat ‘milder’ formulation can be phrased in terms of the extent to which the worstoff categories of citizens approach the average condition among all citizens.

Dimensions of Welfare State Policies How are we to characterize the policies and actions by the state which affect the distribution of welfare among the citizens? In research on social policy there is a long tradition of focusing on the major types of social insurance systems. Very valuable historical studies of the development of the major social insurance schemes in the western European countries have been carried out in connection with the so called HIWED (Historical Indicators of Western European Democracies) project.10 As mentioned above in cross-sectional studies the ILO measures of social security expenditures as proportion of GNP is an often-used indicator.11 Since the major social insurance programs appear to capture the essence of social policy, it is understandable that researchers focus on the development of these programs as well as on public expenditures on social security. However, we must realize that in comparative studies of social policy, there are serious problems connected with these types of indicators. One of these concerns the scope of the measures used. In a well-known essay Titmuss once used the concepts of ‘fiscal welfare’ and ‘occupational welfare’ to refer to state and organizational activities which are not channelled through the conventional machinery of social policy, but which have redistributional effects of much the same nature as the traditional social services.12 The intervention of the state in the redistributive processes in society are thus not limited to those explicitly directed towards persons with publicly acknowledged needs. In studying the role of public policy for the functioning of the distributional processes in society we must thus broaden the scope of our attention to include also measures other than the central social insurance programs.


383

In comparative studies of welfare state politics one must also keep in mind the possibility that the frontiers of distributional conflict may change over time. Thus if we focus on a special set of public measures, for example the major social insurance program, we must be aware of what one might call ‘saturation effects’. At one particular period in a particular country, the set of social insurance programs studied can be in the centre of the political struggles over distribution. Once these programs have been achieved, however, distributional conflict can move to new frontiers. A focus on the specific set of programs need thus no longer be a good ‘dependent variable’ in the sense that it captures the changes that can be going on in the distributional processes in the country. The study by Alber indicating that the political composition of governments was of relevance for the development of the major social insurance programs in western Europe up to the period between the World Wars but not in the post-war period, may reflect the possibility that in the latter period distributional conflict has moved to other frontiers.13 Public expenditure on the major social insurance programs may thus constitute a decreasing part of all public interventions in the distributive processes. Unfortunately earlier studies give only a few leads when we attempt to locate the dimensions in terms of which we can analyze public interventions into the distributive processes. Yet researchers in these areas have undoubtedly had some systems, dimensions or typologies in their minds when writing on social policy. Thus, for instance, Titmuss assures us: ‘Socialist social policies are, in my view, totally different in their purposes, philosophy and attitudes to people from Conservative social policies. They are (or should be) pre-eminently about equality, freedom and social integration.’14 The closest Titmuss appears to have come to a systematic discussion of different dimensions or typologies of social policy, however, is in his posthumously published lecture notes where he differentiates between three types of social policies: institutional, marginal and economic growth types.15 In attempting to characterize various public interventions in the distributive processes of the capitalist democracies we can choose differing points of departure. One possibility, close at hand, appears to be an examination of different sectors of public policy. Another possibility is to characterize state intervention in terms of the stages in the distributive processes where they enter We can also try to develop more comprehensive typologies of social policies.

Sectors of Public Policy State interventions into the distributive process are structured in terms of policy areas which generally are institutionalized through governmental departments and agencies. Such policy areas provide a natural starting point or, at least, a checklist for increasing our awareness of possible variations in state interventions. They thus remind us that in addition to social insurance and other welfare programs there are also public policies related to housing education, health, the labour market, work, taxation, the economy, production, consumption etc. However, although such institutionalized policy divisions in various ways are related to theoretical issues, they appear insufficient if we are searching for more basic dimensions of state interventions in the distributive processes.


384

Stages of the Distributive Process One aspect of importance in this context is the stage of the distributional processes at which state interventions occur. This aspect partly captures differences between institutionalized areas of public policy, something which can be illustrated by the processes distributing buying power. In exchange for their participation on the labour market, employees receive wages or salaries. The buying power which is the end product of this participation is affected by state interventions at several stages. The wages going to employees may have been modified by state interventions, for instance, in terms of minimum-wage laws or legislation affecting the bargaining power of unions. The conversion of wages to income is affected by the levels of participation in the labour force and by employment This indicates that labour market policies and general economic policies can be of central importance in determining the distribution of income. Of course, income is also affected by returns from capital and so on State interventions are of importance also in the conversion of income to buying power. This conversion is thus affected by taxation and inflation which are related to fiscal and economic policies. But it is of course also affected by public transfers, the traditional focus of studies in social policy. To a large extent too, these interventions can be characterized in terms of their timing in the distributive process. Public goods and services available free or at nominal rates for all citizens can be said to enter at an early stage in this part of the conversion process. Other types of pricing policy, concerning foodstuffs and housing, for example, have a similar status. General transfer programs available to citizens in terms of largely universal criteria, such as age, also come in at a relatively early stage of the distributive process. Social insurance programs, which come into effect as soon as the earnings capacity of the individual is decreased, in turn, enter at an earlier stage than means tested programs, which come into effect only when an unacceptable decrease in buying power has occurred. The above discussion is summarized in Figure 2.

Figure 2: State Interventions into Different Stages of the Distribution of Buying Power.


385

Marginal and Institutional Policy Models As indicated above, two polar types of social policy models have been hinted at in the social policy literature, one describing a ‘marginal’ type of social policy, the other


386

describing what we might call an ‘institutional’ type of social policy.16 These policy models, however, have not been spelled out in any detail. Titmuss talks about a ‘residual welfare model of social policy’, which is based on the assumption that private markets and the family constitute the natural channels through which the needs of citizens are to be met. Social policy is to enter only when these natural mechanisms do not function and then only as a temporary substitute. According to the ‘institutional redistributive model of social policy’, however, social policy is an integrated part of society and offers citizens public services outside the market on criteria of need. In this latter model, social policy thus has an important function for redistributing resources in society. The above ideal types of social policy can be seen as the combinations of extreme variable values on a multi-dimensional scale according to which we can hope to describe several relevant aspects of social policies in the western countries. Tentatively the following sub-dimensions can be assumed to be implied by these polar types (Table 1).

Table 1: Preliminary Outline of Subdimensions of Marginal and Institutional Ideal Types Social policy model Subdimension Marginal Institutional Proportion of GNP for social purposes Small Large Proportion of population affected Small Large Importance of full employment and labour force Small Large participation programs Small Large Importance of programs preventing needs Selective Universal Dominant types of programs Standards of benefits MinimumNormal Fees Taxation Dominant type of financing Low High Progressivity of taxation Importance of social control Large Small Importance of private organizations Large Small We can expect that social policy has a quantitatively larger role in the institutional than in the marginal policy models. This is indicated by the proportion of GNP going to social expenditures as well as by the proportion of the population being affected by these policies. An institutional type of social policy can be assumed to enter and to affect the distributive processes at earlier stages than the marginal types of policies. This means that we can expect policies directed towards increasing labour force participation and decreasing unemployment to be more important in the institutional than in the marginal policy models. Generally we can expect that in comparison with the marginal model of social policy, the institutional types of social policies will attempt to prevent needs from arising rather than being limited to alleviating needs which already have become manifest. Further we would expect universalistic measures, directed towards large sectors of the population, to be important in the institutional model of social policy. Selective policies directed towards subgroups of the population with specific needs will be relatively more


387

important in the marginal model. In the marginal model the goal of most policies can be expected to be to provide for a minimum standard of living, whereas in the institutional model we would expect to find a greater stress on achieving a normal or average standard of living. It is possible that fees and actuarial principles play a greater role for financing programs in the marginal model, whereas taxation can be assumed to be more important in the institutional model. Generally we would expect the progressivity of taxation to be higher in the institutional model than in the marginal model. Furthermore, social control, indicated for instance by the extent of individually means-tested programs and aid in kind, is assumed to be more important in the marginal than in the institutional model. Private welfare organizations may play a greater role in marginal than in institutional policy contexts.

Redistribution of Social Policy As indicated above, Titmuss saw redistribution as one of the key goals of his preferred social policy model, which he even labelled the ‘institutional redistributive model of social policy’. The discussion above indicates, however, that the issue of redistribution in social policy is a rather complicated one. The problem here is how wide a concept of social policy we are using. If we limit ourselves to traditional income maintenance programs it is quite possible that a set of such programs constructed according to the marginal model and thus keyed to alleviating the needs of the worst-off citizens may have greater redistributive effects per unit of money spent than institutional types of programs. In this context we can note that in the United States, which appears to have something approaching a ‘marginal’ type of social policy, income maintenance programs have often been evaluated in terms of their redistributive effects, that is, to what extent the programs are ‘target efficient’ in concentrating expenditures on the subgroup defined as poor.17 Institutional types of social policy programs, however, do often not have redistributive effects in the above limited sense. By providing for normal standards and income compensation through largely universal programs, the redistributive effects of these programs may often appear small. However, they may decrease inequality and have important redistributive effects, for instance, by providing for equal access of all citizens to health care and through the system of taxation. They may also have important effects on distribution by contributing to full employment. Variation in the level of employment has been found to strongly affect the distribution of income in western societies.18 Coalition and Formation of the ‘Welfare Backlash’ The debate on ‘the welfare backlash’ touches on several aspects of the dimensions of social policy and brings into focus the issue of rationality in political actions. To the extent that we view political actions as rational behaviour, an assumption which I generally find very fruitful, the welfare backlash poses something of a problem. We know that in most capitalist democracies the distribution of income is negatively skewed within the population, that is, the mean income is considerably higher than the median


388

income. Thus the majority of the citizens are limited to sharing less than half of the total income. In such a situation we might expect coalitions to be formed, where the majority of the citizens combine to redistribute income at the expense of the minority, which receives an unproportional share of the total income. The phenomenon of welfare backlash indicates that this is not what is taking place. Several writers on the welfare backlash appear to imply that citizens do not want to pay high taxes, irrespective of what they receive in return for taxes paid. Wilensky indicates that the welfare backlash has been avoided only in countries with centralistic, corporative patterns of government where taxes are not highly visible.19 He does not consider alternative explanations to the absence of welfare backlash in terms of rational choice by the citizens and does not attempt to investigate other possible causes of the extremely heterogeneous array of somehow deviating or ‘protest’ behaviour among citizens which he terms ‘welfare backlash’. It is possible, however, to try to understand variations in the extent to which social policy is accepted by the citizens in terms of rational choice related to the consequences of these policies. The way in which social policy works can be assumed to be of great importance for the extent to which the consequences of these policies are accepted. Of importance in this context is what types of coalitions among the electorate that different types of social policies tend to generate. A marginal type of social policy, predominant for example in the United States, explicitly or implicitly draws a ‘poverty line’ in the population and thus separates the poor and relatively small minority from the better off majority of the population. The marginal type of social policy programs are directed primarily to those below the poverty line. Thus a rational base is not given for coalition formation between those above and those below the poverty line. The poverty line, in effect, splits the working class and tends to generate coalitions between the better-off workers and the middle class against the lower sections of the working class. Marginalistic social policies thus create a large constituency for a welfare backlash. In fact the ‘welfare backlash’ becomes rational political activity for the majority of the citizens (Figure 3). Certain types of policy strategies, for instance the negative income tax, can be expected to generate this type of coalition formation, where the poor minority is opposed by the majority of the population.

Figure 3: Size of Welfare Backlash Constituencies Created by Marginal and Institutional Social Policy Strategies.


389

An institutional type of social policy, however, can be expected to generate very different patterns of coalitions in the electorate. Since universal programs tend to dominate and a large proportion of the population is affected by the programs, we can expect most households to benefit from these policies in one way or another. Thus an institutional type of social policy leaves a much smaller constituency for a potential welfare backlash than the marginal type of policy. It tends to encourage coalition formation between the working class and the middle class in support for continued welfare state policies. The poor need not therefore stand alone. It is probable that the great acceptance which social policies, in spite of very high levels of taxation, have enjoyed in some countries, to a


390

large extent reflects the consequences of these policies and the patterns of coalitions which they have tended to generate.

Power Distribution and Distributional Conflict To give some bases for a discussion of the feasibility and potential fruitfulness of this study, we will now consider some preliminary empirical data of relevance in this context. In view of the often-made assumptions about the ‘convergence’ of the industrialized nations towards similar social and political structures20 one of the first issues that must be raised here is whether there still are significant differences between the distribution of power resources between their main classes. Of basic interest in this context is the extent to which the working classes in these countries are organized and the extent to which they have achieved control over the government. Two important indicators of the extent of working-class organization and power resources in the capitalist democracies are the proportion of the labour force that is unionized and the proportion of the electorate which supports the parties of the left (here defined as the traditional social democratic parties and those to their left). The extent to which the working class has achieved control over the government during the postwar period has at least two partially different aspects. We can thus measure the extent of working-class control over the political executive by weighting the proportion of government members from the left parties with the proportion of parliamentary seats held by these parties and with the length of time during which the government was in power. Another aspect of relevance is the degree of stability and continuity in the presence of working-class based parties in the government. Oscillating political majorities found, for instance, in Britain or New Zealand, can be expected to generate different types of longterm conflict strategies than situations of more or less permanent participation in coalition governments, and of course also situations where working-class based parties have longterm and stable control over the government. Together with Michael Shalev, I have assembled empirical data describing the above aspects of working-class organization and control in the capitalist democracies during the post-war period.21 For the period between 1946 to 1976 we find very great differences between the countries in these respects in Table 2 (*For description of variables, see Appendix 1). It appears that on the bases of the empirical data these countries can be grouped into five categories: (1) Sweden, Austria and Norway form a group of countries where the working classes have a high degree of mobilization and have had a relatively stable control over the political executive. (2) Another category of countries where the working class is relatively highly mobilized but where its control over the government welfare backlash has been much more unstable and sporadic is made up of Denmark, New Zealand, Britain

Table 2: Patterns of Working Class Mobilization and Political Control in 18 Industrial Capitalist Societies, 1946–76.* Left votes


391

as Proportion percentage Working Weighted time with le Percent of class cabinet representati Pattern Country unionization electorate mobilization share in cabine High Sweden 71 43High High High mobilization, stable Austria 55 45High Medium High control Norway 46 41High High High High Denmark 49 39High Medium Medium mobilization, 42 41High Medium Medium occasional New Zealand control United 44 35High Medium Medium Kingdom Belgium 47 32High Medium Medium MediumAustralia high mobilization,Finland

50

Low control France Italy Japan Low Ireland mobilization, exclusion Canada United States Low-medium Germany mobilization,Netherlands Switzerland partial participation

44High

Medium Low

3

37Medium

Medium Medium

25

32Medium

Low

Low

23

34Medium

Low

Medium

27 36

28Medium 9Low

Low Low

Low Medium

26 27

11Low 1Low

Low Low

Low Low

35 30 23

31Medium 31Medium 18Low

Low Low Low

Medium Medium High

and Belgium. (3) Countries where the working class is relatively highly mobilized but where it has been largely excluded from or only weakly represented in the governments. France, Finland and Italy with labour movements split between social democrats and communists belong to this category as do Japan and Australia. (4) Ireland, the United States and Canada, where the working class has a relatively low degree of mobilization and has been almost entirely excluded from the government. Finally, (5) is made up to the three central European nations which are split by the Catholic-Protestant line of cleavage: the Netherlands, West Germany and Switzerland. The mobilization of the


392

working class is about average in West Germany and in the Netherlands but is relatively low in Switzerland. In all three countries social democratic parties have been included in the governments for relatively long periods, often as minority representatives in coalition governments. Shalev and I have shown that these five groups of countries differ greatly in terms of changes over the present century in one important indicator of distributional conflict, the level of industrial conflict.22 Contrary to prevailing theories on the convergence of industrialized nations in terms of social and political structures and patterns of conflict with the development of industrialism, during the postwar period we find increasing divergence among these countries in terms of the level of industrial conflict. In countries where the working class is strongly mobilized and has had a stable control over the government (Sweden, Norway and Austria), industrial conflict has decreased very markedly and has in fact almost ‘withered away’. Where the working class has been highly mobilized but where its control over the government has been only sporadic and unstable (Denmark, New Zealand, Britain and Belgium), the level of industrial conflict has not undergone important changes but continues to be relatively high. In countries where the working class has a low control over the government but is relatively strongly mobilized (Australia, Finland, France, Italy and Japan) the level of industrial conflict has increased very markedly during the post-war period. Where the working class has a low level of mobilization and has been more or less excluded from the governments (Ireland, United States and Canada), industrial conflict continues to have a very large volume. Whereas in other countries the length of disputes has decreased markedly over this century, in these countries strikes continue to be relatively protracted and generate very high relative levels of man-days idle as a result of industrial disputes. In the fifth category of countries, finally, where the working class has a medium or low level of mobilization but has participated in coalition or minority governments for long periods, the level of industrial disputes has either been relatively low since World War I (Netherlands or Switzerland) or has decreased after World War II (West Germany) Our interpretation of these long-term changes in relationships between, on the one hand, working-class mobilization and control over the government and, on the other hand, the patterns of industrial conflict, is a political one. Very crudely condensed it runs as follows. We argue that where the working class has been strongly mobilized and has achieved a relatively stable control over the government, a crucial decrease in the difference between the power resources of the main classes in the capitalist democracies has occurred. This decrease in the difference in power resources improves the relative position of the still weaker working class and opens up new alternatives for action for this class. A strong and stable control over the government makes it possible for the working class to use public policy to intervene in the distributional processes in society. It is thus no longer limited to fight for its share of production on the labour market. Such a decrease in the difference in power resources between the main classes in the capitalist democracies can be expected to have consequences for the long-term strategies of conflict between these classes. Industrial conflict can be expected to decrease and political struggles related to public interventions into the distributive processes in society to become the central arena for distributional conflicts. In countries where such a


393

decrease in the difference in power resources between the main classes has not occurred, industrial conflict is expected to retain its central role as the arena for distributional conflict, and the level and form of industrial conflict is assumed to depend on the relative strength of the working class.

Inequality, Poverty and Redistribution The project of which this paper is a part is still in its early stages. In order to illustrate some of the problems and possibilities associated with its empirical parts we will here take a preliminary look at data referring to different general aspects of social policy in the countries concerned. Income distribution is an important outcome of the distributive processes in the capitalist democracies. The best data on income distributions for comparative purposes available now probably are those published by Sawyer, for certain OECD countries.23 Of special relevance here are two of his measures based on post-tax income, standardized for size of households. One is the traditional Giniindex, which tends to be sensitive primarily to differences in the central parts of the income distribution. The other is the variance of logs, which is more sensitive to variations in the proportion of income going to the poorer parts of the population. OECD has assembled data on the proportions of the population living in relative poverty in different countries.24 Both national definitions of relative poverty as well as a standardized measure of relative poverty are given.25 Finally we will consider the index of redistributive effects of government budgets suggested by Hewitt.26 This index is based on the ILO-measure of the proportion of public expenditures going to social security. On the assumption that direct taxes are more progressive than indirect taxes, this proportion is weighted by the proportion of total taxation which is based on direct taxes. Although complete data for all countries are missing, a look at available figures indicates that our hypothesis of the consequences of differences in working-class mobilization and control over the government for the distributional processes in these countries cannot be refuted in any simple manner (Table 3).27 Sweden and Norway with highly mobilized working classes and long-term social democratic control over the government appear to have less income inequality than most of the other countries. They also have relatively small proportions of poor persons and relatively high redistributive effects of the government budgets. Among countries with a highly mobilized working class with only sporadic government control, government budgets appear to be relatively highly redistributive. In this category of countries complete data are available only for Britain which, however, appears to have fairly low levels of income inequality and relative poverty. Among the categories of countries where working class parties have been more or less excluded from participation in the government, inequality of income distributions is relatively high. France and the United States have especially high income inequalities but also Canada and Australia fall above the average. The proportions living in relative poverty are also relatively high in these countries and their government budgets do not appear to be markedly redistributive. Data for Italy are unfortunately missing. Japan


394

appears to have a relatively low income inequality.28 In the category of countries where the working class has a medium or low mobilization but where social democratic parties have a tradition of participating in governments, several interesting observations can be made. Thus the Netherlands has the lowest value on our measures of income inequality and a relatively redistributive government budget. West Germany has an above average inequality. However, the lowest deciles of the German income distribution appears to receive a relatively large proportion of income, something which is reflected in the lower value of the variance of logs as well as in the low proportion of population living in relative poverty. In contrast to Switzerland, the government budgets in West Germany also appear to be relatively redistributive.

Unemployment and Public Expenditures This preliminary and partial overview of data of relevance for the describing differences in social policies between the capitalist democracies can be complemented with some figures referring to employment and public expenditures. The level of unemployment in a country depends on many factors, of which the politically determined ones are only one part. Generally, however, we would expect that the choices made by governments in the inflation-unemployment dilemma, which many western governments have perceived themselves as facing during the post-war period, have been affected by the political composition of the government. Social democratic governments can be predicted to have opted for lower levels of unemployment than centrist or conservative governments, which can be expected to give a priority to fighting inflation. Before the period of ‘stagflation’ in the 1970s, the level of unemployment was high in countries like the United States, Canada, Ireland, Belgium and Denmark (Table 4).29 It was low, however, in West Germany, New Zealand, Switzerland, Japan, the Netherlands, Sweden and Norway. Up to the mid-1970s, there thus appears to be a slight tendency toward lower levels of unemployment in countries with strong labour movements and social democratic control over the governments. But this relationship is a weak one and indicates that the level of unemployment has a complicated background. Thus Switzerland, New Zealand, West Germany and Japan have surprisingly low levels of reported unemployment whereas, for

Table 3: Inequality of Post-Tax Income, Proportions Living in Relative Poverty and Redistributive Effects of Government Budgets by Level of Working Class Mobilization and Political Control over the Government in Capitalist Democracies Income per cent poor inequality Pattern Country Variance Gini OECD National Redistributive effects of logs. index def. def. of government budget 1962–64 High Sweden .060 .271 3.5 9.5


mobilization, Austria stable Norway control Denmark High mobilization, New Zealand periodic Britain Belgium AverageAustralia Finland high mobilization, France low control Italy Japan Low Ireland mobilization, Canada political United exclusion States LowWest medium mobilization, Germany partial Netherlands participation Switzerland

.072

301

395

9.3 6.9

5.0

6.5 6.8 .080 .327

7.5

.109 .354

8.0

13.2 14.4 8.2

.136 .417

16.0

15–20

4.2 5.2 4.2

.105 .348

11.0

24.0 15.1

2.2 2.2 3.6

.111 .369

13.0

11.9

4.5

.087 .386 .047 .264

3.0

.071 .336

6.0

8.3 8.8 5.1

Table 4: Unemployment, Government Expenditures for Employment Support, Public Expenditures and Social Security Expenditures in Different Categories of Capitalist Democracies Public Social security Expenditures Expenditures as expenditures as for per cent in per cent in employment GNP GNP, Pattern Country Percent 1975–78 % of 1959– GNP 1950 1975 1962 1972 78 High Sweden 1.8 1.61 37.5 51.0 6.0 9.3 mobilization,Austria (1.7) 0.02 25.0 40.3 14.1 15.3 Norway (2.0) 0.55 25.5 46.5 5.1 9.8 stable control High Denmark (3.9) 19.4 47.5 6.5 9.9 mobilization,New (0.2) 0.20 7.6 6.5 Zealand occasional Britain 3.3 0.36 30.4 46.1 5.7 7.7 (2.6) 26.3 44.9 11.7 14.1 control Belgium MediumAustralia 2.1 4.7 4.0


high Ireland mobilization,France low control Italy Japan Low Ireland mobilization,Canada exclusion United states LowWest medium Germany mobilization,Netherlands partial Switzerland participation

(2.1) 2.5 3.8 1.4 5.7 5.6 5.4

0.22

26.9 28.4 27.8

0.05 0.17 0.26 0.33

37.2 42.4 43.1 2.1

26.8 27.4

1.2

0.60

(1.8) (0.1)

396

9.9 12.4 10.4

41.2 36.2

6.7 11.8 7.5 2.8 5.3 5.4 5.5

30.8

45.6

11.9

12.4

27.0 20.8

54.3 27.4

8.6

14.2

6.4 7.3 7.4

instance, Britain, Denmark and Belgium have higher levels than we might have expected.30 During the period of ‘stagflation’ in the latter half of the 1970s, government expenditures in support of employment programs were relatively high in Sweden, Norway and West Germany but low in Austria and Japan. Unemployment appears to have increased markedly in West Germany and the Netherlands whereas it has remained relatively low in Sweden. The proportion of the GNP used for public expenditures reflects a variety of factors. In 1950 this proportion was high in Sweden, West Germany and Britain but low in Switzerland and Denmark. Up to 1975 this proportion had increased markedly in all the countries under consideration, on average from about one fourth to about one half of GNP. The increase has been highest in the Netherlands, Denmark and Norway and lowest in Switzerland, the United States and Finland. The proportion of public expenditures going to social security as defined by the ILO31 does not show clear differences between our countries. It tends to be low in countries with a politically excluded or weak working class, for instance, in Japan, Ireland, United States, Canada and Australia. However, it is low also in Sweden and Norway but high in Australia, Belgium, France and West Germany. The above preliminary overview indicates that there are rather great complications in the use and interpretation of overall, crude statistics easily available for describing the policies in our different countries.

Conclusions Research on the forms, causes and consequences of social policy and the development of the welfare state should be explicitly related to models of the distributive processes in the capitalist democracies. In the social sciences there are presently many partly overlapping, partly competing theoretical approaches in explanations of distribution and inequality. On the one hand, we have the mainstream social science theories, which predict a


397

convergence in structures and policies of the industrialized countries as a result of the development of industrialization. On the other hand, we have certain varieties of Marxian theory which tend to view all capitalist societies as basically similar and tend to regard variations in working-class power within these nations as insignificant. The central hypothesis of the present project, which is still in its early stages, goes contrary to both the above approaches in contemporary social science. The project is based on the hypothesis that the distribution of power resources between the main classes in the capitalist democracies is of central importance for processes of distribution and the pattern of distributional conflict in these countries. We thus expect that the distribution of power resources tends to affect the conditions for the distributive processes in society, the way and extent to which the state enters into these processes, the types of policies that are used, the arenas in which distributional conflicts tend to become manifested as well as the outcomes of these processes in terms of various aspects of inequality. The preliminary analysis in this paper indicates that such an approach may be a fruitful one.

Appendix 1 Union membership has been ‘standardized’ by the size of the non-agricultural labour force. We have data on votes, legislative seats, and cabinet seats for all ‘left’ parties, defined as social democratic parties and parnies to their left. In our dichotomy we have thus, for instance, excluded the US Democratic party from the left parties. Also rightwing splinter groups from the main social democratic parties are excluded, for instance the Italian Social Democratic party. The dichotomization of what is essentially a continuous variable implies that considerable variation remains within each pole of the dichotomy. The available information was converted into annual time series, where necessary by distributing subdivisions of years on a proportionate basis. Figures on cabinet participation were obtained from a data file at the Department of Political Science, University of Minnesota. Our figures refer to the proportion of the partisan members of each cabinet who represented left parties. The ‘weighted’ cabinet measure is the product of this proportional cabinet score and a score for proportional legislative representation. Data on valid votes, size of electorate, and legislative seats are nearly always from Mackie, Thomas T. and Rose, Richard, The International Almanac of Electoral History (London: Macmillan, 1974) and the annual updates by these authors in the European Journal of Political Research.

Categories used in Table 2 Working class Mobilization is based on the combined rank order of per cent unionization and left proportion of the electorate. Weighted Cabinet Share: Low = .00–.10; Medium=.11–.30; High=.31–.50. Time with Left Representation in Cabinet: Low=0–24 per cent; Medium=25–74 per cent; High=75–100 per cent. Splits within the labour movement are those described in standard sources on comparative labour movements. Party splits are defined in terms of the average share of the total left vote held by the


398

largest party in the period 1946–.76. A minor split indicates a share of 90 per cent or higher. A major split implies a share of about 50 per32 Notes Professor of Social Policy, the Swedish Institute for Social Research and university of Stockholm. This paper constitutes one of the interim products of an ongoing research project on comparative social and distributional policy in the western nations. It is supported by the German Marshall Fund of the United States, the Bank of Sweden Tercentenary Foundation and the Swedish Delegation for Social Research. I wish to thank participants in the ‘welfare state’ workshop at the joint sessions of the European Consortium for Political Research in Florence, 1980 and in particular Professor Rudolf Klein for valuable comments on the manuscript. 1. See for example, Briggs, Asa, ‘The welfare state in historical perspective’, European Journal of Sociology, vol. 2, no. 2, (1961) pp 221–58. 2. E.g. ILO, The Cost of social security (Geneva: ILO, 1972). 3. E.g.Parkin, Frank, Class, Inequality and the Political Order (London: MacGibbon & Kee, 1971). 4. Cf.e.g.Jackman, Richard, Politics and social equality: A comparative analysis (New York: Wiley, 1975), Miliband, Ralph, The State in Capitalist Society (London: Weidenfold and Nicholson, 1969), Parkin, op., cit., Scase, Richard, Social Democracy in Capitalist Society (London: Rowman and Littlefield, 1977), Therborn, Goran, ‘Sweden before and after social democracy: A first overview’, Acta Sociologica, vol. 21, (1978) Supplement, 37–58, and Wilensky, Harold, The Welfare state and equality: Structural and ideological roots of public expenditures (Berkeley: University of California Press, 1975). 5. Cf.Hewitt, Christopher, ‘The effect of political democracy and social democracy on equality in industrial societies: A cross-national comparison’, American Sociological Review, vol. 42 (June 1977), pp. 450–64, Stephens, John, The Transition from Capitalism to Socialism (London: Macmillan, 1979) and Alber, Jens, ‘The growth of social insurance in Western Europe: Has social democracy made a difference?’ Paper prepared for the XIth World Congress of the International Political Science Association in Moscow, August 1979. 6. Thurow, Lester, C.Generating inequality: Mechanisms of distribution in the US economy (New York: Basic Books, 1975) and Doeringer, Peter, B. and Piore, Michael, J. International Labour Markets and Manpower Analysis (Lexington, Mass.: D.C.Heath, 1971). 7. E.g.Lenski, Gerhard, E., Power and Privilege (New York: McGraw Hill, 1966). 8. For a discussion of this approach see Korpi, Walter, The Working Class in Welfare Capitalism, Work Unions and Politics in Sweden (London: Routledge and Kegan Paul, 1978), Chap. 2. 9. Cf.Korpi, Walter, ‘Approaches to the study of poverty in the United States. Critical comments from a European perspective’. In Perspectives on Poverty, The National Academy of Sciences (In press). 10. Cf.Flora, Peter, ‘On the development of the Western European welfare states’. IPSA Congress Paper, Edinburgh, 1976. 11. E.g.Wilensky, op., cit., 12. Titmuss, Richard, M., Essays on ‘the Welfare State’ (London: Unwin, 1958), pp. 34–55.


399

13. Alber, op. cit. 14. Titmuss, Richard, M., Commitment to Welfare (London: Unwin, 1968). p. 116. 15. Titmuss, Richard, M., Social policy (London: Allen & Unwin, 1974), pp. 30–1. See also Wilensky, Harold and Lebeaux, Ch. Industrial and social welfare (New York: Russell Sage Foundation, 1958), pp. 138–47. 16. Ibid. 17. Korpi, ‘Approaches to the study of poverty in the United States’. 18. For Swedish data, see Spant, Roland, Den svenska inkomstfordelningens utveckling (Uppsala: Studia Oeconomica Uppsalaensia, no. 4,1976). 19. Wilensky, Harold, The ‘new corporatism’: centralization and the welfare state (Beverly Hills: Sage Publications, 1976). 20. E.g.Kerr, Clark et al. Industrialism and Industrial Man (Cambridge, Mass.: Harvard University Press, 1960). 21. Korpi, Walter and Shalev, Michael, ‘Strikes, power and politics in western nations, 1900– 76’, Political Power and Social Theory, vol. 1, (no. 1,1980) pp. 299–332 and ‘Strikes industrial relations and class conflict in capitalist societies’, British Journal of Sociology, vol. 30 (June 1979) pp. 164–87. 22. Ibid., 23. Sawyer, Malcolm, Income distribution in the OECD countries (Paris: OECD, 1976). 24. OECD, Public expenditure on income maintenance programs (Paris, 1976). 25. The standardised measure of relative poverty varies according to size of households. Thus, for instance, for a one-person household it is put at two-thirds of the average per capita income, whereas for a three-person household it is 125 per cent of the per capita income. 26. Hewitt, op. cit. 27. Sources of data for Table 3 are Sawyer, op. cit., (income unequality), OECD, op. cit., (poverty) and Hewitt, op., cit., (redistribute effects). 28. A Japanese college has indicated that the data used by Sawyer may over-represent middlesize income earners. 29. In Table 4, unemployment measures are adjusted to US standards except for figures given within parentheses, which are national definitions given by the OECD. Other sources for this table are for public expenditures, Kohl, Jürgen, Trends and Problems in Postwar Public Expenditure Development in Western Europe and North America’, chap. 9 in Peter Flora and Arnold J.Heidenheimer (eds.), The Development of Welfare States in Europe and America (New Brunswick: Transaction Press, 1980) and OECD op., cit., and for expenditures for unemployment ‘Employment and manpower policies as part of the adjustment process (Note by the secretariat)’, MAS (78)13 (1978). 30. Hibbs, Douglas A. ‘Political parties and macroeconomic policy’, American Political Science Review, vol. 71 (December 1977) pp. 1467–82, and ‘Communication’, American Political Science Review, vol. 73 (March 1979) pp. 185–90 has suggested that unemployment is affected by the composition of governments. His sample of countries, however, excludes several countries which do not easily fit this hypothesis. 31. The ILO measure of social security expenditures includes public expenditures on medical care, cash sickness benefits, employment injury benefits under social insurance, unemployment insurance, statutory old-age pensions and family allowances. 32. See Korpi and Shalev ‘Strikes, power and politics in the western nations, 1900–1976’.


400

55 The Three Political Economies of the Welfare State* Gosta Esping-Andersen

* Source: Canadian Review of Sociology and Anthropology, 1989, vol. 26, pp. 10–36.

Le très long débat sur la nature et les causes de l’état providence n’a pas apporté de réponses définitives à l’une ou 1’autre de ces questions. Get article a trois buts: 1/réntégrer le débat dans la tradition intellectuelle de l’économie politique pour mieux faire ressortir les principaux problèmes théoriques; 2/spécifier les caractéristiques principales de l’état providence, les méthodes traditionnelles de mesure de l’état providence en termes de dépenses n’éant plus satisfaisantes, 3/ ‘sociologiser’ 1’étude de l’état providence. La plupart des études sur le sujet ont été basées sur une conception linéaire du monde: plus ou moins de puissance, d’industrialisation ou de dépenses. Cet article considère que les états providences sont d’abord des groupes de régimes-types et que leur développement doit être expliqué de manière interactive. The protracted debate on the welfare state has failed to produce conclusive answers as to either the nature or causes of welfare state development. This article has three aims: 1/to reintegrate the debate into the intellectual tradition of political economy. This serves to put into sharper focus the principal theoretical questions involved; 2/ to specify what are the salient characteristics of welfare states. The conventional ways of measuring welfare states in terms of their expenditures will no longer do; 3/ to ‘sociologize’ the study of welfare states. Most studies have assumed a world of linearity: more or less power, industrialization or spending. This article insists that we understand welfare states as clusters of regime-types, and that their development must be explained interactively.

The Legacy of Classical Political Economy Contemporary welfare state debates have been guided by two questions. First, does social citizenship diminish the salience of class? Or, in other words, can the welfare state fundamentally transform capitalist society? Second, what are the causal forces behind welfare state development? These questions are not recent. Indeed, they were formulated by the 19th Century political economists 100 years before any welfare state can rightfully be said to have come into existence. The classical political economists—whether of liberal, conservative or Marxist persuasion—were preoccupied with the relationship between capitalism and welfare. Their answers obviously diverged, but their analyses were unequivocally directed to the relationship between market (and property) and the state (democracy). The question they asked was largely normative: what is the optimal

The Three Political Economies of the Welfare State

401

division of responsibility between market and state? Contemporary neo-liberalism echoes the contributions of classical liberal political economy. To Adam Smith, the market was the superior means for the abolition of class, inequality and privilege. Aside from a necessary minimum, state intervention would likely stifle the equalizing process of competitive exchange, create monopolies, protectionism and inefficiency: the state upholds class, the market can potentially undo class society (Smith, 1961:11, especially pp. 232–6).1 Liberal political economists were not necessarily of one mind when it came to policy advocacy. Nassau Senior and later Manchester liberals emphasized the laissez-faire element of Smith, rejecting any form of social protection outside the cash nexus. J.S.Mill and the ‘reformed liberals,’ in turn, were willing to let markets be regulated by a modicum of political regulation. Yet, they were all agreed that the road to equality and prosperity should be paved with a maximum of free markets and a minimum of state interference. This enthusiastic embrace of market capitalism may now appear unjustified. But, we must take into account that the state which confronted these early political economists was tinged with legacies of absolutist privileges, mercantilist protectionisms, and pervasive corruption. They were attacking systems of governance which repressed the ideals of both freedom and enterprise. Hence, theirs was revolutionary theory, and from this vantage point, we can understand why Adam Smith sometimes reads like Karl Marx.2 Democracy was an Achilles heel to many liberals. Their ideals of freedom and democratic participation were grounded in a world of small property owners; not of growing property-less masses who held in their sheer numbers the possibility of seizing state power. The liberals feared the principle of universal suffrage, for it would likely politicize the distributional struggle, pervert the market and fuel inefficiencies. Many liberals discovered that democracy would contradict the market. Both conservative and Marxist political economists understood this contradiction, but proposed of course opposite solutions. The most coherent conservative critique of laissez faire came from the German historical school; in particular from Friedrich List, Adolph Wagner and Gustav Schmoller. They refused to believe that capitalist efficiency was best assured by the pure commodity status of workers in the raw cash nexus of the market. Instead, conservative political economy believed that patriarchical neo-absolutism could provide the kind of legal, political and social framework that would assure a capitalism without class struggle. One prominent conservative school promoted a ‘Monarchical Welfare State’ that would, at once, provide for social welfare, class harmony, loyalty, and productivity. It was discipline, not competition, that would guarantee efficiency. The state (or church) was the institution best equipped to harmonize conflicting interests.3 Conservative political economy emerged in reaction to the French Revolution and the Paris Commune. It was avowedly nationalist, anti-revolutionary, and sought to arrest the democratic impulse. It feared social levelling and favored a society that retained both hierarchy and class. It held that class conflicts were not natural; that democratic mass participation, the dissolution of recognized rank and status boundaries were threats to social harmony.


402

The key to Marxian political economy, of course, was its rejection of the liberal claim that markets guarantee equality. Capitalist accummulation as Dobb (1946) put it, disowns people of property with the end-result being ever deeper class divisions. Here, the state’s role is not neutrally benevolent, nor is it a fountain of emancipation; it exists to defend property rights and the authority of capital. To Marxism this is the foundation of class dominance. The central question, not only for Marxism but for the entire contemporary debate on the welfare state, is whether and under what conditions the class divisions and social inequalities produced by capitalism can be undone by parliamentary democracy. The liberals feared that democracy would produce socialism and they were consequently not especially eager to extend it. The socialists, in contrast, suspected that parliamentarism would be little more than an empty shell or, as Lenin suggested, a mere ‘talking shop’ (Jessop, 1982). This line of analysis, echoed in much of contemporary Marxism, leads to the conclusion that social reforms emerge in response to the exigencies of capitalist reproduction, not to the emancipatory desires of the working classes.4 Among socialists, a more positive analysis of parliamentarism came to prevail after the extension of full political citizenship. The theoretically most sophisticated contributions came from Austro-Marxists such as Adler and Bauer, and from German social democrats, especially Eduard Heimann. Heimann’s (1929) starting point was that conservative reforms may have been motivated by desires to repress labor mobilization, but that their very presence nonetheless alters the balance of class power: the social wage lessens the worker’s dependence on the market and employers. The social wage is thus also a potential power resource that defines the frontier between capitalism and socialism. It introduces an alien element into the capitalist political economy. This intellectual position has enjoyed quite a renaissance in recent Marxism (Offe, 1985; Bowles and Gintis, 1986). The social democratic model, as outlined above, did not necessarily abandon the orthodox assumption that fundamental equality requires economic socialization. Yet, historical experience soon demonstrated that socialization was a goal that could not be pursued realistically through parliamentarism.5 Social democracy’s embrace of parliamentary reformism as its dominant strategy for equality and socialism was premised on two arguments. The first was that workers require social resources, health and education to participate effectively in a democratized economy. The second argument was that social policy is not only emancipatory, but it is also economically efficient (Myrdal and Myrdal, 1936). Following Marx on this point, the strategy therefore promotes the onward march of capitalist productive forces. But, the beauty of the strategy was that social policy would also assure social democratic power mobilization. By eradicating poverty, unemployment and complete wage dependency, the welfare state increases political capacities and diminishes the social divisions that are barriers to political unity among workers. The social democratic model, then, puts forward one of the leading hypotheses of contemporary welfare state debate: the argument that parliamentary class mobilization is a means for the realization of socialist ideals of equality, justice, freedom and solidarity.


403

The Political Economy of the Welfare State Our political economy forebears defined the analytic basis of much recent scholarship. They isolated the key variables of class, state, market and democracy; and they formulated the basic propositions about citizenship and class, efficiency and equality, capitalism and socialism. Contemporary social science distinguishes itself from classical political economy on two scientifically vital fronts. First, it defines itself as a positive science and shies away from normative prescription (Robbins, 1976). Second, classical political economists had little interest in historical variability; they saw their efforts as leading towards a system of universal laws. Although contemporary political economy sometimes still clings to the belief in absolute truths, the comparative and historical method that, today, underpins almost all good political economy is one that reveals variation and permeability. Despite these differences, most recent scholarship has as its focal point the stateeconomy relationship defined by 19th Century political economists. And, given its enormous growth, it is understandable that the welfare state has become a major test case for contending theories of political economy. Below, we shall review the contributions of comparative research on the development of welfare states in advanced capitalist countries. It will be argued that most scholarship has been misdirected, mainly because it became detached from its theoretical foundations. We must therefore recast both the methodology and concepts of political economy in order to adequately study the welfare state. This will constitute the focus of the final section of this paper. Two types of approaches have dominated in the explanation of welfare states; one, a systemic (or structuralist) theory; the other, an institutional or actor-oriented explanation.

The Systems/Structuralist Approach System or structuralist theory seeks to capture the logic of development holistically. It will easily focus on the functional requisites for the reproduction of society and economy, it will be inclined to emphasize cross-national similarities rather than differences. One variant begins with a theory of the industrial society, and argues that industrialization makes social policy both necessary and possible. It makes welfare states necessary because pre-industrial modes of social reproduction, such as the family, the church, noblesse oblige, and guild solidarity are destroyed by the forces attached to modernization—social mobility, urbanization, individualism, and market dependence. The crux of the matter is that the market is no adequate substitute because it caters only to those who are able to perform in it. Hence, the ‘welfare function’ is appropriated by the the nation state. The welfare state is also made possible by the rise of modern bureaucracy as a rational, universalist and efficient form of organization. It is a means for managing collective goods, but also a center of power in its own right, and will thus be inclined to promote its own growth.


404

This kind of reasoning has informed the so-called ‘logic of industrialism’ perspective, according to which the welfare state will emerge as the modern industrial economy destroys traditional forms of social security (Flora and Alber, 1981; Pryor, 1969). But, the thesis has difficulties explaining why government social policy only emerged 50 or even 100 years after traditional community was effectively destroyed. The basic response draws on Wagner’s Law (1962; 1883) and on Marshall (1920), namely that a certain level of economic development, and thus surplus, is needed in order to permit the diversion of scarce resources from productive use (investments) to welfare (Wilensky and Lebeaux, 1958). In this sense, the perspective follows in the footsteps of the old liberals. Social redistribution endangers efficiency, and only at a certain economic level will a negativesum trade-off be avoidable (Okun, 1975). The new structural Marxism offers a surprisingly parallel analysis. It breaks with its classical forebears’ strongly action-centred theory. Like the industrialism thesis, its analytical starting point is not the problems of markets, but the logic of a mode of production. Capital accumulation creates contradictions that social reform can alleviate (O’Connor, 1973). This tradition of Marxism, like its ‘logic of industrialism’ counterpart, fails to see much relevance in actors in the promotion of welfare states. The point is that the state, as such, is positioned in such a way that it will serve the collective needs of capital. The theory is thus premised on two crucial assumptions: first, that power is structural and second, that the state is ‘relatively’ autonomous from class directives (Poulantzas, 1973; Block, 1977; for a recent critical assessment of this literature, see Therborn, 1986; and Skocpol and Amenta, 1986). The ‘logic of capitalism’ perspective invites difficult questions. If, as Przeworski (1980) has argued, working class consent is assured on the basis of material hegemony, that is self-willed subordination to the system, it is difficult to see why up to 40 per cent of the national product must be allocated to the legitimation activities of a welfare state. A second problem is to derive state activities from a ‘mode of production’ analysis. Eastern Europe may perhaps not qualify as socialist, but neither is it capitalist. Yet, there we find ‘welfare states,’ too. Perhaps accumulation has functional requirements in whichever way it proceeds? (Skocpol and Amenta, 1986; Bell, 1978).

The Institutional Approach The classical political economists made it clear why democratic institutions should influence welfare state development. The liberals feared that full democracy might jeopardize markets and inaugurate socialism. Freedom, in their view, necessitated a defence of markets against political intrusion. In practice, this is what the laissez-faire state sought to accomplish. But it was this divorce of politics and economy which fuelled much of the institutionalist analyses. Best represented by Polanyi (1944), but also by a number of anti-democratic exponents of the historical school, the institutional approach insists that any effort to isolate the economy from social and political institutions will destroy human society. The economy must be embedded in social communities in order for it to survive. Thus, Polanyi sees social policy as one necessary precondition for the reintegration of the social economy.


405

An interesting recent variant of institutional alignment theory is the argument that welfare states emerge more readily in small, open economies that are particularly vulnerable to international markets. As Katzenstein (1985) and Cameron (1978) show, there is a greater inclination to regulate class distributional conflicts through government and interest concertation when both business and labor are captive to forces beyond domestic control. The impact of democracy on welfare states has been argued ever since J.S. Mill and de Tocqueville. The argument is typically phrased without reference to any particular social agent or class. It is, in this sense, that it is institutional. In its classical formulation, the thesis was simply that majorities will favor social distribution to compensate for market weakness or market risks. If wage earners are likely to demand a social wage, so are capitalists (or farmers) apt to demand protection in the form of tariffs, monopoly or subsidies. Democracy is an institution that cannot resist majoritarian demands. In its modern formulations, the democracy thesis has many variants. One identifies stages of nationbuilding in which full citizenship incorporation requires social rights (Marshall, 1950; Bendix, 1964; Rokkan, 1970). A second variant, developed by both pluralist and public choice theory, argues that democracy will nurture intense party competition around the median voter that will, in turn, fuel rising public expenditures. Tufte (1978), for example, argues that major extensions of public intervention will occur around elections as a means of voter mobilization. The democratic-institutionalist approach faces considerable empirical problems (Skocpol and Amenta, 1986). According to the thesis, a democratic polity is the basic precondition for welfare state emergence, and welfare states are more likely to develop the more democratic rights are extended. Yet, the thesis confronts not only the historical oddity that the first major welfare state initiatives occurred prior to democracy, but also that they were often motivated by desires to arrest its realization. This was certainly the case in France under Napoleon II, in Germany under Bismarck, and in Austria under Taaffe. Conversely, welfare state development was most retarded where democracy arrived early, such as in the United States, Australia, and Switzerland. This apparent contradiction can be explained, but only with reference to social classes and social structure: nations with early democracy were overwhelmingly agrarian and dominated by small property owners who used their electoral powers to reduce, not raise, taxes (Dich, 1973). In contrast, ruling classes in authoritarian polities are better positioned to impose high taxes on an unwilling populace.

Social Class as a Political Agent We have noted that the case for a class mobilization thesis flows from social democratic political economy. It differs from structuralist and institutional analyses by its emphasis on the social classes as the main agents of change, and its argument that the balance of class power determines distributional outcomes. To emphasize active class mobilization does not necessarily deny the importance of structured or hegemonic power (Korpi, 1983). But it is held that parliaments are, in principle, effective institutions for the translation of mobilized power into desired policies and reforms. Accordingly,


406

parliamentary politics are capable of overriding hegemony, and may be made to serve interests that are antagonistic to capital. Further, the class mobilization theory assumes that welfare states do more than simply alleviate the current ills of the system; a ‘social democratic’ welfare state will, in its own right, establish critical power resources for wage earners and, thus, strengthen labor movements. As Heimann (1929) originally held, social rights push back the frontiers of capitalist power and prerogatives. The question of why the welfare state itself is a power resource is vital for the theory’s applicability. The answer is that wage earners in the market are inherently atomized and stratified, compelled to compete, insecure and dependent on decisions and forces beyond their control. This limits their capacity for collective solidarity and mobilization. The social rights, income security, equalization and eradication of poverty that a universalistic welfare state pursues, are necessary preconditions for the strength and unity that collective power mobilization demands (Esping-Andersen, 1985a). The single most difficult problem for this thesis is to specify the conditions for power mobilization. Power depends on the resources that flow from the unity of electoral numbers and from collective bargaining. Power mobilization, in turn, depends on levels of trade union organization, vote shares, parliamentary and cabinet seats held by left, or labor, parties. But how long a period of sustained power mobilization is required in order to produce decisive effects? If power is measured over a brief time span (5–10 years), we risk the fallacy of a ‘Blum’/ Mitterrand’ effect: a brief spell of leftist power that proves ineffectual because the left is ousted again before having had a chance to act. There are several valid objections to the class mobilization thesis. Three, in particular, are quite fundamental. One is that in advanced capitalist nations, the locus of decision making and power may shift from parliaments to neo-corporatist institutions of interest intermediation (Shonfield, 1965; Schmitter and Lembruch, 1979). A second criticism is that the capacity of labor parties to influence welfare state development is circumscribed by the structure of rightist party power. Castles (1978; 1982) has argued that the degree of unity among the rightist parties is more important than is the activated power of the left. Other authors have emphasized the fact that denominational (usually social Catholic) parties in countries such as Holland, Italy, and Germany mobilize large sections of the working classes and pursue welfare state programs not drastically at variance with their socialist competitors (Schmidt, 1982; Wilensky, 1981). The class mobilization thesis has, rightly, been criticized for its Swedocentrism, i.e., its inclination to define the process of power mobilization too much on the basis of the rather extraordinary Swedish experience (Shalev, 1984). These objections address a basic fallacy in the theory’s assumptions about class formation: we cannot assume that socialism is the natural basis for wage-earner mobilization. Indeed, the conditions under which workers become socialists are still not adequately documented. Historically, the natural organizational bases of worker mobilization were pre-capitalist communities, especially the guilds, but also the Church, ethnicity or language. A ready-made reference to false consciousness will not do to explain why Dutch, Italian or American workers continue to mobilize around nonsocialist principles. The dominance of socialism in the Swedish working class is as much a puzzle as is the dominance of confessionalism in the Dutch. The third and, perhaps, most fundamental objection has to do with the model’s linear


407

view of power. It is problematic to hold that a numerical increase in votes, unionization or seats will translate into more welfare statism. First, for socialist as for other parties, the magical’50 per cent’ threshold for parliamentary majorities seems practically unsurmountable (Przeworski, 1985). Second, if socialist parties represent working classes in the traditional sense, it is clear that they will never succeed in their project. In very few cases has the traditional working class been numerically a majority; and its role is rapidly becoming marginal.6 Probably the most promising way to resolve the combined linearity and working class minority problem lies in recent applications of Barrington Moore’s path-breaking class coalition thesis to the transformation of the modern state (Weir and Skocpol, 1985; Gourevitch, 1986; Esping-Andersen, 1985a; Esping-Andersen and Friedland, 1982). Thus, the origins of the Keynsian full employment commitment and the social democratic welfare state edifice have been traced to the capacity of (variably) strong working class movements to forge a political alliance with farmers’ organizations; additionally, it is arguable that sustained social democracy has come to depend on the formation of a new working class-white collar coalition. The class coalitional approach has additional virtues. Two nations, such as Austria and Sweden, may score similarly on working class mobilization variables and yet produce highly unequal policy results. This can be explained by differences in the two countries’ historical coalition formation: the breakthrough of Swedish social democratic hegemony stems from its capacity to forge the famous ‘red-green’ alliance; the comparative disadvantage of the Austrian socialists rests in the ‘ghetto’ status assigned to them by virtue of the rural classes being captured by a conservative coalition (Esping-Andersen and Korpi, 1984). In sum, we have to think in terms of social relations, not just social categories. Whereas structural-functionalist explanations identify convergent welfare state outcomes, and class mobilization paradigms see large, but linearly distributed, differences, an interactive model such as the coalitions approach directs attention to distinct welfare state regimes.

What is the Welfare State? Every theoretical paradigm must somehow define the welfare state. How do we know when and if a welfare state responds functionally to the needs of industrialism, or to capitalist reproduction and legitimacy? And how do we identify a welfare state that corresponds to the demands that a mobilized working class might have? We cannot test contending arguments unless we have a commonly shared conception of the phenomenon to be explained. A remarkable attribute of the entire literature is its lack of much genuine interest in the welfare state as such. Welfare state studies have been motivated by theoretical concerns with other phenomena, such as power, industrialization or capitalist contradictions; the welfare state itself has generally received scant conceptual attention. If welfare states differ, how do they differ? And when, indeed, is a state a welfare state? This turns attention straight back to the original question: what is the welfare state?


408

A common textbook definition is that it involves state responsibility for securing some basic modicum of welfare for its citizens. Such a definition skirts the issue of whether social policies are emancipatory or not; whether they help system legitimatation or not; whether they contradict or aid the market process; and what, indeed, is meant by ‘basic’? Would it not be more appropriate to require of a welfare state that it satisfies more than our basic or minimal welfare needs? The first generation of comparative studies started with this type of conceptualization. They assumed, without much reflection, that the level of social expenditure adequately reflects a state’s commitment to welfare. The theoretical intent was not really to arrive at an understanding of the welfare state, but rather to test the validity of contending theoretical models in political economy. By scoring nations with respect to urbanization, level of economic growth, and the share of aged in the demographic structure, it was believed that the essential features of industrial modernization were adequatly captured. Alternatively, by scoring nations on left party strength, or working class power mobilization (with complex weighted scores of trade unionism, electoral strength and cabinet power), others sought to identify the impact of working class mobilization as formulated in the social democratic model. The findings of the first generation comparativists are extremely difficult to evaluate. No convincing case can be made for any particular theory. The shortage of nations for comparisons statistically restricts the number of variables that can be tested simultaneously. Thus, when Cutright (1965) or Wilensky (1975) finds that economic level, with its demographic and bureaucratic correlates, explains most welfare state variations in ‘rich countries,’ relevant measures of working class mobilization or economic openness are not included. A conclusion in favor of a ‘logic of industrialism’ view is therefore in doubt. And, when Hewitt (1977), Stephens (1979), Korpi (1983), Myles (1984) and Esping-Andersen (1985b) find strong evidence in favor of a working class mobilization thesis, or when Schmidt (1982; 1983) finds support for a neocorporatist, and Cameron (1978) for an economic openness argument it is without fully testing against the strongest alternative explanation.7 Most of these studies claim to explain the welfare state. Yet, their focus on spending may be irrelevant or, at best, misleading. Expenditures are epiphenomenal to the theoretical substance of welfare states. Moreover, the linear scoring approach (more or less power, democracy or spending) contradicts the sociological notion that power, democracy, or welfare are relational and structured phenomena. By scoring welfare states on spending, we assume that all spending counts equally. But, some welfare states, the Austrian for example, spend a large share on benefits to privileged civil servants. This is normally not what we would consider a commitment to social citizenship and solidarity. Others spend disproportionally on means-tested social assistance. Few contemporary analysts would agree that a reformed poor relief tradition qualifies as a welfare state commitment. Some nations spend enormous sums on fiscal welfare in the form of tax privileges to private insurance plans that mainly benefit the middle classes. But these tax expenditures do not show up on expenditure accounts. In Britain, total social expenditure has grown during the Thatcher period; yet, this is almost exclusively a function of very high unemployment. Low expenditures on some programs may signify a welfare state more seriously committed to full employment.


409

Therborn (1983) is right when he holds that we must begin with a conception of state structure. What are the criteria with which we should judge whether, and when, a state is a welfare state? There are three approaches to this question. Therborn’s proposal is to begin with the historical transformation of state activities. Minimally, in a genuine welfare state the majority of its daily routine activities must be devoted to servicing the welfare needs of households. This criterion has far-reaching consequences. If we simply measure routine activity in terms of spending and personnel, the result is that no state can be regarded as a real welfare state until the 1970’s! And, some that we normally label as welfare states will still not qualify because the majority of their routine activities concern defence, law and order, administration and the like (Therborn, 1983). Social scientists have been too quick to accept nations’ self-proclaimed welfare state status. They have also been too quick to conclude that the presence of the battery of typical social programs signifies the birth of a welfare state. The second conceptual approach derives from Richard Titmuss’ (1958) classical distinction between residual and institutional welfare states. The former assumes that state responsibility begins only when the family or the market fails; its commitment is limited to marginal groups in society. The latter model addresses the entire population, is universalistic, and implants an institutionalized commitment to welfare. It will, in principle, extend welfare commitments to all areas of distribution vital for societal welfare. This approach has fertilized a variety of new developments in comparative welfare state research (Myles, 1984; Korpi, 1980; Esping-Andersen and Korpi, 1984; 1986; Esping-Andersen, 1985b; 1987). And it has forced researchers to move away from the black box of expenditures and towards the content of welfare states: targeted versus universalistic programs, the conditions of eligibility, the quality of benefits and services and, perhaps most importantly, the extent to which employment and working life are encompassed in the state’s extension of citizen rights. This shift to welfare state typologies makes simple linear welfare state rankings difficult to sustain. We might in fact be comparing categorically different types of states. The third approach is to select theoretically the criteria on which to judge types of welfare states. This can be done by measuring actual welfare states against some abstract model and then by scoring programs, or entire welfare states, accordingly (Day, 1978; Myles, 1984). The weakness of this approach is that it is ahistorical, and does not necessarily capture the ideals or designs that historical actors sought to realize in the struggles over the welfare state. If our aim is to test causal theories that involve actors, we should begin with the demands that were actually promoted by those actors that we deem critical in the history of welfare state development. It is difficult to imagine that anyone struggled for spending per se.

A Respecification of the Welfare State8 Few can disagree with T.H.Marshall’s (1950) proposition that social citizenship constitutes the core idea of a welfare state. What, then, are the key principles involved in social citizenship? In our view, they must involve first and foremost the granting of social rights. This mainly entails a de-commodification of the status of individuals vis-à-vis the


410

market. Secondly, social citizenship involves social stratification; one’s status as a citizen will compete with, or even replace, one’s class position. Thirdly, the welfare state must be understood in terms of the interface between the market, the family, and the state. These principles need to be fleshed out prior to any theoretical specification of the welfare state. Rights and De-Commodification As commodities in the market, workers depend for their welfare entirely on the cashnexus. The question of social rights is thus one of de-commodification, that is of granting alternative means of welfare to that of the market. De-commodification may refer either to the service rendered, or to the status of a person, but in both cases it signifies the degree to which distribution is detached from the market mechanism. This means that the mere presence of social assistance or insurance may not necessarily bring about significant de-commodification if they do not substantially emancipate individuals from market dependence. Means-tested poor relief will possibly offer a security blanket of last resort. But if benefits are low and attached with social stigma, the relief system will compel all but the most desperate to participate in the market. This was precisely the intent of the 19th Century poor laws. Similarly, most of the early social insurance programs were deliberately designed to maximize labor market performance (Ogus, 1979). Benefits required long contribution periods and were tailored to prior work effort. In either case, the motive was to avert work-disincentive effects. There is no doubt that de-commodification has been a hugely contested issue in welfare state development. For labor, it has always been a priority. When workers are completely market dependent, they are difficult to mobilize for solidaristic action. Since their resources mirror market inequalities, divisions emerge between the ‘ins’ and the ‘outs,’ making labor movement formation difficult. De-commodification strengthens the worker and weakens the absolute authority of the employer. It is for exactly this reason that employers always opposed de-commodification. De-commodified rights are differentially developed in contemporary welfare states. In social assistance dominated welfare states rights are not so much attached to work performance as to demonstrable need. Needs-tests and typically meagre benefits, however, serve to curtail the de-commodifying effect. Thus, in nations where this model is dominant (mainly in the Anglo-Saxon countries), the result is actually to strengthen the market since all but those who fail in the market will be encouraged to contract private sector welfare. A second dominant model espouses compulsory state social insurance with fairly strong entitlements. Yet, again, this may not automatically secure substantial decommodification, since this hinges very much on the fabric of eligibility and benefit rules. Germany was the pioneer of social insurance, but over most of the century can hardly be said to have brought about much in the way of de-commodification through its social programs. Benefits have depended almost entirely on contributions and, thus, work and employment. In fact, before the Second World War, average pensions in the German insurance system for workers were lower than prevailing poverty assistance rates (Myles, 1984). The consequence, as with the social assistance model, was that most workers


411

would choose to remain at work rather than retire. In other words, it is not the mere presence of a social right, but the corresponding rules and preconditions that dictate the extent to which welfare programs offer genuine alternatives to market dependence. The third dominant model of welfare, namely the Beveridge-type citizens benefit, may, at first glance, appear the most de-commodifying. It offers a basic, equal benefit to all irrespective of prior earnings, contributions or performance. It may indeed be a more solidaristic system, but not necessarily de-commodifying since, only rarely, have such schemes been able to offer benefits of such a standard that they provide recipients with a genuine option to that of working. De-commodifying welfare states are, in practice, of very recent date. A minimalist definition must entail that citizens can freely, and without potential losses of job, income or general welfare, opt out of work under conditions when they, themselves, consider it necessary for reasons of health, family, age or even educational self-improvement; when, in short, they deem it necessary for participating adequately in the social community. With this definition in mind, we would, for example, require of a sickness insurance that individuals be secured benefits equal to normal earnings, the right to absence with minimal proof of medical impairment, and for the duration that the individual deems necessary. These conditions, it is worth noting, are those usually enjoyed by academics, civil servants and higher echelon white collar employees. Similar requirements would be made of pensions, maternity leave, parental leave, educational leave and unemployment insurance. Some nations have moved towards this level of de-commodification but only recently and, in many cases, with significant exemptions. Thus, in almost all nations benefits were upgraded to equal normal wages in the late 1960s and early 1970s. But, in some countries, for example, prompt medical certification in case of illness is still required; in others, entitlements depend on long waiting periods of up to two weeks; and, in still others, the duration of entitlements is very short (in the United States, for example, unemployment benefit duration is maximally six months, compared to 30 in Denmark). Overall, the Scandinavian welfare states tend to be the most de-commodifying; the Anglo-Saxon the least. The Welfare State as a System of Stratification Despite the emphasis given to it in both classical political economy and in T.H. Marshall’s pioneering work, the relationship between citizenship and social class remains severely neglected, both theoretically and empirically. Generally speaking, the issue has either been assumed away (it has been taken for granted that the welfare state creates a more egalitarian society). Or it has been approached narrowly in terms of income distribution or in terms of whether education promotes upward social mobility. A more basic question, it seems, is what kind of stratification system is promoted by social policy. The welfare state is not just a mechanism that intervenes in, and possibly corrects, the structure of inequality; it is in its own right a system of stratification. It orders actively and directly social relations. Comparatively and historically, we can easily identify alternative systems of stratification embedded in welfare states. The poor relief tradition, and its contemporary


412

means-tested social assistance offshoot, was conspicuously designed for purposes of stratification. By punishing and stigmatizing recipients, it promotes severe social dualisms, especially within the ranks of the working classes. It comes as no suprise that this model of welfare has been a chief target of labor movement attacks. The social insurance model promoted by conservative reformers such as Bismarck and von Taaffe was also explicitly a form of class politics. It sought, in fact, to achieve two simultaneous stratification results. The first was to consolidate divisions among wage earners by legislating distinct programs for different class and status groups, each with its own conspicuously unique set of rights and privileges designed to accentuate the individual’s appropriate station in life. The second objective was to tie the loyalties of the individual directly to the monarchy, or central state authority. This was Bismarck’s motive when he promoted a direct state supplement to the pension benefit. This statecorporativist model was pursued mainly in nations such as Germany, Austria, Italy and France and often resulted in a labyrinth of status-specific insurance funds (in France and Italy, for example, there exist more than 100 status-distinct pension schemes). Of special importance in this corporatist tradition was the establishment of particularly privileged welfare provisions for the civil service (‘Beamten’). In part, this was a means of rewarding loyalty to the state and in part, a way of demarcating this group’s uniquely exalted social status. We should, however, be careful to note that the corporatist statusdifferentiated model springs mainly from the old guild tradition. The neo-absolutist autocrats, such as Bismarck, saw in this tradition a means to combat the rising labor movements. The labor movements were as hostile to the corporatist model as they were to poor relief, in both cases for obvious reasons. Yet, the alternatives first espoused by labor were no less problematic from the point of view of uniting the workers as one solidaristic class. Almost invariably, the model that labor first pursued was that of the self-organized friendly societies or equivalent union- or party-sponsored fraternal welfare plan. This is not surprising. Workers were obviously suspicious of reforms sponsored by a hostile state, and saw their own organizations not only as bases of class mobilization, but also as embryos of an alternative world of solidarity and justice, as a microcosm of the socialist haven to come. Nonetheless, these microsocialist societies often became problematic class ghettos that divided rather than united workers. Membership was typically restricted to the strongest strata of the working class and the weakest—who needed protection most—were most likely outside. In brief, the fraternal society model contradicted the goal of working class mobilization. The socialist ghetto approach was an additional obstacle when socialist parties found themselves forming governments and having to pass the social reforms they so long had demanded. For reasons of political coalition building and broader solidarity, their welfare model had to be recast as welfare for the ‘people.’ Hence, the socialists came to espouse the principle of universalism and, borrowing from the liberals, typically designed on the lines of the democratic flat-rate, general revenue financed, Beveridge model. As an alternative to means-tested assistance and corporatist social insurance, the universalistic system promotes status equality. All citizens are endowed with similar rights, irrespective of class or market position. In this sense, this system is meant to cultivate cross-class solidarity, a solidarity of the nation. But, the solidarity of flat-rate


413

universalism presumes an historically peculiar class structure; one in which the vast majority of the population are the little people’ for whom a modest, albeit egalitarian, benefit may be considered adequate. Where this no longer obtains, as occurs with growing working class prosperity and the rise of the new middle classes, flat-rate universalism inadvertently promotes dualism because the better off turn to private insurance and to fringe-benefit bargaining to supplement modest equality with what they have decided are accustomed standards of welfare. Where this process unfolds (as in Canada or the United Kingdom), the result is that the wonderfully egalitarian spirit of universalism turns into a dualism similar to that of the social assistance state: the poor rely on the state, and the remainder on the market. It is not only the universalist, but in fact all historical welfare state models which have faced the dilemma of class structural change. But, the response to prosperity and middle class growth has been varied and so, therefore, has been the stratificational outcome. The corporatist insurance tradition was, in a sense, best equipped to manage new and loftier welfare state expectations since the existing system could technically be upgraded quite easily to distribute more adequate benefits. Adenauer’s 1957 pension reform in Germany was a pioneer in this respect. Its avowed purpose was to restore status differences that had eroded due to the old insurance system’s incapacity to provide benefits tailored to expectations. This it did simply by moving from contribution- to earnings-graduated benefits without altering the framework of status-distinctiveness. In nations with either a social assistance or a universalistic Beveridge-type system, the option was whether to allow the market or the state to furnish adequacy and satisfy middle class aspirations. Two alternative models emerged from this political choice. The one typical of Great Britain and most of the Anglo-Saxon world was to preserve an essentially modest universalism in the state and allow the market to reign for the growing social strata demanding superior welfare. Due to the political power of such groups, the dualism that emerges is not merely one between state and market, but also between forms of welfare state transfers: in these nations, one of the fastest growing components of public expenditure is tax-subsidies for so-called ‘private’ welfare plans. And the typical political effect is eroding middle class support for what is less and less a universalistic public sector transfer system. Yet another alternative has been to seek a synthesis of universalism and adequacy outside of the market. This road has been followed in the countries where, by mandating or legislation, the state includes the new middle classes by erecting a luxurious secondtier, universally inclusive, earnings related insurance scheme on top of the flat-rate egalitarian one. Notable examples are Sweden and Norway. By guaranteeing benefits tailored to expectations, this solution re-introduces benefit inequalities, but effectively blocks off the market. It thus succeeds in retaining universalism and, therefore, also the degree of political consensus required to preserve broad and solidaristic support for the high taxes that such a welfare state model demands. Welfare State Regimes Welfare states vary considerably with respect to their principles of rights and stratification. This results in qualitatively different arrangements between state, market


414

and the family. The welfare state variations, we find, are therefore not linearly distributed, but clustered by regime-types. In one cluster, we find the ‘liberal’ welfare state, in which means-tested assistance, modest universal transfers, or modest social insurance plans predominate. These cater mainly to a clientele of low income, usually working class, state dependents. It is a model in which, implicitly or explicitly, the progress of social reform has been severely circumscribed by traditional, liberal work-ethic norms; one where the limits of welfare equal the marginal propensity to demand welfare instead of work. Entitlement rules are therefore strict and often associated with stigma; benefits are typically modest. In turn, the state encourages the market, either passively by guaranteeing only a minimum, or actively by subsidizing private welfare schemes. The consequence is that this welfare state regime minimizes de-commodification effects, effectively contains the realm of social rights, and erects a stratification order that blends a relative equality of poverty among state welfare recipients, market-differentiated welfare among the majorities, and a class-political dualism between the two. The archetypical examples of this model are the United States, Canada, and Australia. Nations that approximate the model are Denmark, Switzerland, and Great Britain. A second regime-cluster is composed of nations such as Austria, France, Germany and Italy. Here, the historical corporatist-statist legacy was upgraded to cater to the new ‘post-industrial’ class structure. In these ‘corporatist’ welfare states, the liberal obsession with market efficiency and commodification was never pre-eminent and, as such, the granting of social rights was hardly ever a seriously contested issue. What predominated was the preservation of status differentials; rights, therefore, were attached to class and status. This corporativism was subsumed under a state edifice perfectly ready to displace the market as a provider of welfare; hence, private insurance and occupational fringe benefits play a truly marginal role in this model. On the other hand, the state’s emphasis on upholding status differences means that its redistributive effects are negligible. But, the corporativist regimes are also typically shaped by the Church, and therefore influenced by a strong commitment to the preservation of traditional family patterns. Social insurance typically excludes non-working wives, and family benefits encourage motherhood. Day care, and similar family services, are conspicuously underdeveloped, and the ‘subsidiarity principle’ serves to emphasize that the state will only interfere when the family’s capacity to service its members is exhausted. An illustrative example is German unemployment assistance. Once a person has exhausted his/her entitlement to normal unemployment insurance, eligibility for continued assistance depends on whether one’s family commands the financial capacity to aid the unfortunate; this obtains for persons of any age. The third, and clearly smallest, regime-cluster is composed of those countries in which the principles of universalism and de-commodifying social rights were extended also to the new middle classes. We may call it the ‘social democratic’ regime-type since, in these nations, social democracy clearly was the dominant force behind social reform. Norway and Sweden are the clearest cases, but we should also consider Denmark and Finland. Rather than tolerate a dualism between state and market, between working class and middle class, the social democrats pursued a welfare state that would promote an equality of the highest standards, rather than an equality of minimal needs as was pursued


415

elsewhere. This implied, first, that services and benefits be upgraded to levels commensurable to even the most discriminate tastes of the new middle classes; and, secondly, that equality be furnished by guaranteeing workers full participation in the quality of rights enjoyed by the better off. This formula translates into a mix of highly de-commodifying and universalistic programs that, nonetheless, are tailored to differentiated expectations. Thus, manual workers come to enjoy rights identical to those of salaried white collar employees or civil servants; all strata and classes are incorporated under one universal insurance system; yet, benefits are graduated according to accustomed earnings. This model crowds out the market and, consequently, inculcates an essentially universal solidarity behind the welfare state. All benefit, all are dependent, and all will presumably feel obliged to pay. The social democratic regime’s policy of emancipation addresses both the market and the traditional family. In contrast to the corporatist-subsidiarity model, the principle is not to wait until the family’s capacity to aid is exhausted, but to pre-emptively socialize the costs of familihood. The ideal is not to maximize dependence on the family, but capacities for individual independence. In this sense, the model is a peculiar fusion of liberalism and socialism. The result is a welfare state that grants transfers directly to the children, and takes direct caring responsibilities for children, the aged and the helpless. It is, accordingly, committed to a heavy social service burden, not only to service family needs, but also to permit women to choose work rather than the household. Perhaps the most salient characteristic of the social democratic regime is its fusion of welfare and work. It is, at once, a welfare state genuinly committed to a full employment guarantee, and a welfare state entirely dependent on its attainment. On the one side, it is a model in which the right to work has equal status to the right of income protection. On the other side, the enormous costs of maintaining a solidaristic, universalistic and decommodifying welfare system means that it must minimize social problems and maximize revenue income. This is obviously best done with most people working, and the fewest possible living off social transfers. While it is empirically clear that welfare states cluster, we must recognize that no single case is pure. The social democratic regimes of Scandinavia blend crucial socialist and liberal elements. The Danish and Swedish unemployment insurance schemes, for example, are still essentially voluntarist. Denmark’s labor movement has been chronically incapable of pursuing full employment policies due in part to trade union resistance to active manpower policies. And in both Denmark and Finland, the market has been allowed to play a decisive role in pensions. Neither are the liberal regimes pure. The American social security system is redistributive, compulsory and far from actuarial. At least in its early formulation, the New Deal was as social democratic as was contemporary Scandinavian social democracy. In contrast, the Australian welfare state would appear exceedingly close to the bourgeoisliberal ideal-type, but much of its edifice has been the co-responsibility of Australian labor. And, finally, the European corporatist regimes have received both liberal and social democratic impulses. Social insurance schemes have been substantially destratified and unified in Austria, Germany, France and Italy. Their extremely corporativist character has thus been reduced. Notwithstanding the lack of purity, if our essential criteria for defining welfare states


416

have to do with the quality of social rights, social stratification, and the relationship between state, market and family, the world is composed of distinct regime-clusters. Comparing welfare states on scales of more or less or, indeed, better or worse, will yield highly misleading results.

The Causes of Welfare State Regimes If welfare states cluster into three distinct regime types, we are confronted with a substantially more complex task of identifying the causes of welfare state differences. What is the explanatory power of industrialization, economic growth, capitalism, or working class political power in accounting for regime types? A first superficial answer would be: very little. The nations we study are all more or less similar with regard to all but the working class mobilization variable. And we find very powerful labor movements and parties in each of the three clusters. A theory of welfare state developments must clearly reconsider its causal assumptions if we wish to explain clusters. The hope to find one single powerful causal motor must be abandoned; the task is to identify salient interaction effects. Based on the preceding arguments, three factors in particular should be of importance: the nature of (especially working-) class mobilization; class-political coalition structures; and the historical legacy of regime institutionalization. As we have noted, there is absolutely no compelling reason to believe that workers will automatically and naturally forge a socialist class identity; nor is it plausible that their mobilization will look especially Swedish. The actual historical formation of working class collectivities will diverge, and so also will their aims and political capacities. Fundamental differences appear both in trade unionism and party development. A key element in trade unionism is the mix of craft and industrial unions. The former is prone to particularism and corporativism; the latter is inclined to articulate broader, more universal objectives. This blend decisively affects the scope for labor party action and also the nature of political demands. Thus, the dominance of the AFL in pre-war United States was a major impediment to social policy development. Likewise, the heavily craftoriented Danish labor movement, compared to its Norwegian and Swedish counterparts, blocked social democracy’s aspirations for an active labor market policy for full employment. In the United States, craft unions believed that negotiating occupational benefits was a superior strategy, given their privileged market position. In Denmark, craft unions jealously guarded their monopoly on training and labor mobility. Conversely, centralized industrial unionism will tend to present a more unified and consolidated working class clientele to the labor party, making policy consensus easier, and power mobilization more effective. It is clear that a working class mobilization thesis must pay attention to union structure. Equally decisive is political or denominational union fragmentation. In many nations, for example Finland, France and Italy, trade unions are divided between socialist and communist parties; white collar unions are politically unaffiliated or divide their affiliation among several parties. Denominational trade unionism has been a powerful feature in Holland, Italy and other nations. Since trade unionism is such a centrally important basis for party mobilization, such fragmentation will weaken the left and thus


417

benefit the non-socialist parties’ chances of power. In addition, fragmentation may entail that welfare state demands will be directed to many parties at once. The result may be less party conflict over social policy, but it may also mean a plurality of competing welfare state principles. For example, the subsidiarity principle of Christian workers will conflict with the socialists’ concern for the emancipation of women. The structure of trade unionism may, or may not, be reflected in labor party formation. But, under what conditions are we likely to expect certain welfare state outcomes from specific party configurations? There are many factors that conspire to make it virtually impossible to assume that any labor, or left, party will ever be capable, single-handedly, of structuring a welfare state. Denominational or other divisions aside, it will be only under extraordinary historical circumstances that a labor party alone will command a parliamentary majority long enough to impose its will. We have noted that the traditional working class has, nowhere, ever been an electoral majority. It follows that a theory of class mobilization must look beyond the major leftist party. It is an historical fact that welfare state construction has depended on political coalition building. The structure of class coalitions is much more decisive than are the power resources of any single class. The emergence of alternative class coalitions is, in part, determined by class formation. In the earlier phases of industrialization, the rural classes usually constituted the single largest electorate. If social democrats wanted political majorities, it was here that they were forced to look for allies. Therefore, it was ironically the rural economy that was decisive for the future of socialism. Where the rural economy was dominated by small, capital intensive family farmers, the potential for an alliance was greater than where it rested on large pools of cheap labor. And, where farmers were politically articulate and well-organized (as in Scandinavia), the capacity to negotiate political deals was vastly superior. The role of the farmers in coalition formation and, hence, in welfare state development is clear. In the Nordic countries, the conditions obtained for a broad red-green alliance for a full-employment welfare state in return for farm price subsidies. This was especially true in Norway and Sweden, where farming was highly precarious and dependent on state aid. In the United States, the New Deal was premised on a similar coalition (forged by the Democratic party) but with the important difference that the labor intensive South blocked a truly universalistic social security system, and opposed further welfare state developments. In contrast, the rural economy of Continental Europe was very inhospitable to red-green coalitions. Often, as in Germany and Italy, much of agriculture was labor intensive and labor unions and left parties were seen as a threat. In addition, the conservative forces on the continent had succeeded in incorporating farmers into ‘reactionary’ alliances, helping to consolidate the political isolation of labor. Political dominance was, until after World War II, largely a question of rural class politics. The construction of welfare states in this period was, therefore, dictated by which force captured the farmers. The absence of a red-green alliance does not necessarily imply that no welfare state reforms were possible. On the contrary, it implies which political force came to dominate their design. Great Britain is an exception to this general rule, because the political significance of the rural classes eroded before the turn of the century. In this way, Britain’s coalition logic showed at an early date the dilemma that faced most other nations later, namely that the new white collar middle classes


418

constitute the linchpin for political majorities. The consolidation of welfare states after World War II came to depend fundamentally on the political alliances of the new middle classes. For social democracy, the challenge was to synthesize working class and white collar demands without sacrificing the commitment to solidarity. Since the new middle classes have, historically, enjoyed a relatively privileged position in the market, they have also been quite successful in meeting their welfare demands outside the state or, as civil servants, by privileged state welfare. Their employment security has traditionally been such that full employment has been a peripheral concern. Finally, any program for drastic income equalization is likely to be met with great hostility among a middle class clientele. On these grounds, it would appear that the rise of the new middle classes would abort the social democratic project and strengthen a liberal welfare state formula. The political position of the new middle classes has, indeed, been decisive for welfare state consolidation. Their role in shaping the three welfare state regimes described earlier is clear. The Scandinavian model relied almost entirely on social democracy’s capacity to incorporate them in a new kind of welfare state: one that provided benefits tailored to the tastes and expectations of the middle classes, but nonetheless retained universalism of rights. Indeed, by expanding social services and public employment, the welfare state participated directly in manufacturing a middle class instrumentally devoted to social democracy. In contrast, the Anglo-Saxon nations retained the residual welfare state model precisely because the new middle classes were not wooed from the market into the state. In class terms, the consequence is dualism. The welfare state caters essentially to the working class, and to the poor. Private insurance and occupational fringe benefits cater to the middle classes. Given the electoral importance of the latter, it is quite logical that further extensions of welfare state activities are resisted. Indeed, the most powerful thrust in these countries is an accent on fiscal welfare; i.e., on tax expenditures and deductions for private sector welfare plans. The third, Continental European, welfare state regime has also been patterned by the new middle classes, but in a different way. The cause is historical. Developed by conservative political forces, these regimes institutionalized a middle class loyalty to the preservation of both occupationally segregated social insurance programs and, ultimately, to the political forces that brought them into being. Adenauer’s great pension reform in 1957 was explicitly designed to resurrect middle class loyalties.

Conclusion We have here presented an alternative to a simple class mobilization theory of welfare state development. It is motivated by the analytical necessity of shifting from a linear to an interactive approach with regard to both welfare states and their causes. If we wish to study welfare states, we must begin with a set of criteria that define their role in society. This role is certainly not to spend or tax; nor is it necessarily that of creating equality. We have presented a framework for comparing welfare states that takes into consideration the principles for which the historical actors willingly have struggled and mobilized. And,


419

when we focus on the principles embedded in welfare states, we discover distinct regime clusters, not merely variations of ‘more’ or ‘less’ around a common denominator. The salient forces that explain the crystallization of regime differences are interactive. They involve, first, the pattern of working class political formation and, second, the structuration of political coalitions with the historical shift from a rural economy to a middle class society. The question of political coalition formation is decisive. Third, past reforms have contributed decisively to the institutionalization of class preferences and political behavior. In the corporatist regimes, hierarchical statusdistinctive social insurance cemented middle-class loyalty to a peculiar type of welfare state. In the liberal regimes, the middle classes became institutionally wedded to the market. And, in Scandinavia, the fortunes of social democracy after the war were closely tied to the establishment of a middle class welfare state that benefits both its traditional working class clientele and the new white collar strata. In part, the Scandinavian social democrats were able to do so because the private welfare market was relatively undeveloped and, in part, because they were capable of building a welfare state with features of sufficient luxury to satisfy the tastes of a more discriminating public. This also explains the extraordinarily high cost of Scandinavian welfare states. But, a theory that seeks to explain welfare state growth should also be able to understand its retrenchment or decline. It is typically believed that welfare state backlash movements, tax revolts, and roll-backs are ignited when social expenditure burdens become too heavy. Paradoxically, the opposite is true. Anti-welfare state sentiments over the past decade have generally been weakest where welfare spending has been heaviest, and vice-versa. Why? The risks of welfare state backlash depend not on spending, but on the class character of welfare states. Middle class welfare states, be they social democratic (as in Scandinavia) or corporatist (as in Germany), forge middle class loyalties. In contrast, liberal, residualist welfare states found in the United States, Canada and, increasingly, Britain depend on the loyalties of a numerically weak, and often politically residual social stratum. In this sense, the class coalitions in which the three welfare states were founded, explain not only their past evolution but also their future prospects.

Notes This is part of an ongoing project on welfare states and labor markets, funded by the Research Council of the IUE and European Community. I wish to thank Thomas Cusack Steven Lukes, John Myles, Fritz von Nordheim Nielsen, and the participants in my IUE seminar on Political Economy, for their constructive comments and criticisms on an earlier version of this paper. 1. Adam Smith is often cited but rarely read. A closer inspection of his writings reveals a degree of nuance and a battery of reservations that substantially qualify a delirious enthusiasm for the blessings of capitalism. 2. In the Wealth of Nations (Smith, 1961:236), he comments on states that uphold the privilege and security of the propertied as follows: ‘…civil government, so far as it is instituted for the security of property, is in reality instituted for the defence of the rich against the poor, or of


420

those who have some property against those who have none at all.’ 3. This tradition is virtually unknown to Anglo-Saxon readers, since so little has been translated into English. A key text which greatly influenced public debate and later social legislation was Adolph Wagner’s Rede Ueber die Soziale Frage (1872). For an English language overview of this tradition of political economy, see Schumpeter (1954), and especially Bower (1947). From the Catholic tradition, the fundamental texts are the two Papal Encyclicals, Rerum Novarum (1891) and Quadrogesimo Anno (1931). The social Catholic political economy’s main advocacy is a social organization where a strong family is integrated in cross-class corporations, aided by the state in terms of the subsidiarity principle. For a recent discussion, see Richter (1987). Like the liberals, the conservative political economists also have their contemporary echoes, although substantially fewer in number. A revival occurred with Fascism’s concept of the Corporative (‘Standische’) state of Ottmar Spann in Germany. The subsidiarity principle still guides much of German Christian Democratic politics (see Richter, op. cit.). 4. Chief proponents of this analysis are the German ‘state derivation’ school (Muller and Neususs, 1973); Offe (1972); O’Connor (1973); Gough (1979); and also the work of Poulantzas (1973). As Skocpol and Amenta (1986) note in their excellent overview, the approach is far from one-dimensional. Thus, Offe, O’Connor and Gough identify the function of social reforms as being also concessions to mass demands and as potentially contradictory. Historically, socialist opposition to parliamentary reforms was principled less by theory than by reality. August Bebel, the great leader of German social democracy, rejected Bismarck’s pioneering social legislation, not because he did not favor social protection, but because of the blatantly anti-socialist and divisionary motives behind Bismarck’s reforms. 5. This realization came from two types of experiences. One, typified by Swedish socialism in the 1920s, was the discovery that not even the working class base showed much enthusiasm for socialization. In fact, when the Swedish socialists established a special commission to prepare plans for socialization, it concluded after 10 years of exploration that it would be practically quite impossible to undertake. A second kind of experience, typified by the Norwegian socialists and Blum’s Popular Front government in 1936, was the discovery that radical proposals could easily be sabotaged by the capitalists’ capacity to withhold investments and export their capital abroad. 6. This is obviously not a problem for the parliamentary class hypothesis alone; structural Marxism faces the same problem of specifying the class character of the new middle classes. If such a specification fails to demonstrate that it constitutes a new working class, both varieties of Marxist theory face severe (although not identical) problems. 7. This literature has been reviewed in great detail by a number of authors. See, for example, Wilensky et al. (1985). For excellent and more critical evaluations, see Uusitalo (1984), Shalev (1983) and Skocpol and Amenta (1986). 8. This section derives much of its material from earlier writings (see, especially EspingAndersen, 1985a; 1987).

References Bell, D. 1978 The Cultural Contradictions of Capitalism. New York: Basic Books. Bendix, R. 1964 Nation-Building and Citizenship. New York: John Wiley and Sons. Block, F. 1977 ‘The Ruling Class does not Rule.’ Socialist Review 7 (May-June).


421

Bower, R.H. 1947 German Theories of the Corporate State. New York: Russel and Russel. Bowles, S., and H.Gintis 1986 Democracy and Capitalism New York: Basic Books. Brandes, S. 1910 American Welfare Capitalism, 1880–1940. Chicago: University of Chicago Press. Cameron, D. 1978 ‘The Expansion of the Public Economy: A Comparative Analysis.’ American Political Science Review 4. Castles, F. 1978 The Social Democratic Image of Society. London: Routledge and Kegan Paul Castles, F. (ed.) 1982 The Impact of Parties, London: Sage. Cutright, P. 1965 ‘Political Structure, Economic Development, and National Social Security Programs.’ American Journal of Sociology 70. Day, L. 1978 ‘Government Pensions for the Aged in 19 Industrialized Countries.’ In R. Tomasson (ed.), Comparative Studies in Sociology. Greenwich, Conn: JAI Press. Dich, J. 1973 Den Herskende Klasse. Copenhagen: Borgen. Dobb, M. 1946 Studies in the Development of Capitalism. London: Routledge and Kegan Paul. Downs, A. 1957 An Economic Theory of Democracy. New York: Harper and Row. Esping-Andersen, G. 1985a Politics against Markets. Princeton: Princeton University Press. Andersen, G. 1985b ‘Power and Distributional Regimes.’ Politics and Society 14. Andersen, G. 1987 ‘Citizenship and Socialism: De-commodification and Solidarity in the Welfare State.’ In G.Esping-Andersen, M.Rein, and L.Rainwater (eds.), Stagnation and Renewal. Armonk, NY: M.E.Sharpe. Esping-Andersen, G., and R.Friedland 1982 ‘Class Coalitions in the Making of West European Economies.’ Political Power and Social Theory 3. Esping-Andersen, G., and W.Korpi 1984 ‘Social Policy as Class Politics in Postwar Capitalism.’ In J.Goldthorpe (ed.), Order and Conflict in Contemporary Capitalism. Oxford: Oxford University Press. Esping-Andersen, G., and W.Korpi 1986 ‘From Poor Relief to Institutional Welfare States.’ In R.Erikson et al. (eds.), The Scandinavian Model. Armonk, NY: M.E.Sharpe. Evans, E. 1978 Social Policy, 1830–1914. London: Routledge and Kegan Paul. Flora, P., and J.Alber (eds.) 1981 ‘Modernization, Democratization and the Development of Welfare States in Europe.’ In P.Flora and A.Heidenheimer (eds.), The Development of Welfare States in Europe and America. London: Transaction Books. Flora, P., and A.Heidenheimer 1981 The Development of Welfare States in Europe and America. London: Transaction Books. Gough, I. 1979 The Political Economy of the Welfare State. London: Macmillan. Gourevitch, P. 1986 Politics in Hard Times. Ithaca, NY: Cornell University Press.


422

Heimann, E. 1929 Soziale Theorie der Kapitalismus. Frankfurt: Suhrkamp (1980 reprint). Hewitt, C. 1977 ‘The Effect of Political Democracy and Social Democracy on Equality in Industrial Societies.’ American Sociological Review 42. Jessop, B. 1982 The Capitalist State. Oxford: Martin Robertson. Katzenstein, P. 1985 Small States in World Markets. Ithaca, NY: Cornell University Press. Korpi, W. 1980 ‘Social Policy and Distributional Democracies.’ West European Politics 3. Korpi, W. 1983 The Democratic Class Struggle, London: Routledge and Kegan Paul. Marshall, A. 1920 Principles of Economics (8th Edition). London: Macmillan. Marshall, T.H. 1950 Citizenship and Social Class. Cambridge: Cambridge University Press. Muller, W., and C.Neususs 1973 ‘The Illusion of State Socialism and the Contradiction between Wage Labor and Capital.’ Telos 25 (Fall). Myles, J 1984 Old Age in the Welfare State. Boston: Little Brown. Myrdal, A., and G.Myrdal 1936 Kris i Befolkningsfraagan. Stockholm: Tiden. O’Connor, J. 1973 The Fiscal Crisis of the State. New York: St.Martin’s Press. Offe, C. 1972 ‘Advanced Capitalism and the Welfare State.’ Politics and Society 4. Offe, C. 1984 Contradictions of the Welfare State. London: Hutchinson. Offe, C. 1985 Disorganized Capitalism. Cambridge, Mass: MIT Press. Ogus, A. 1979 ‘Social Insurance, Legal Development and Legal History.’ In H.F.Zacher (ed.), Bedingungen für die Entstehung von Sozialversicherung. Berlin: Duncker and Humboldt. Okun, A. 1975 Equality and Efficiency: The Big Trade-Off. Washington, DC: Brookings Institute. Olson, M. 1982 The Rise and Decline of Nations. New Haven, Conn.: Yale University Press. Polanyi, K. 1944 The Great Transformation. New York: Rhinehart. Poulantzas, N. 1973 Political Power and Social Classes. London: New Left Books. Pryor, F. 1969 Public Expenditures in Communist and Capitalist Nations. London: Allen and Unwin. Przeworski, A. 1980 ‘Material Bases of Consent: Politics and Economics in a Hegemonic System.’ Political Power and Social Theory 1. Przeworski, A. 1985 Capitalism and Social Democracy. Cambridge: Cambridge University Press Richter, E. 1987 ‘Subsidiaritat und Neoconservatismus.’ Politische Vierteljahresschrift 28. Rimlinger, G. 1971 Welfare and Industrialization in Europe, America and Russia. New York: John Wiley.


423

Robbins, L. 1976 Political Economy: Past and Present. London: Macmillan. Rokkan, S. 1970 Citizens, Elections, Parties. Oslo, Universitetsforlage. Schmidt, M. 1982 ‘The Role of Parties in Shaping Macro-economic Policies.’ In F.Castles (ed.), The Impact of Parties. London: Sage. Schmidt, M. 1983 ‘The Welfare State and the Economy in Periods of Economic Crisis.’ European Journal of Political Research 11. Schmitter, P., and G.Lembruch 1979 Trends toward Corporatist Intermediation. London: Sage. Schumpeter, J. 1944 Capitalism, Socialism and Democracy. London: Allen and Unwin. Schumpeter, J. 1954 History of Economic Analysis. New York: Oxford University Press. Shalev, M. 1984 ‘The Social Democratic Model and Beyond.’ Comparative Social Research 6. Shonfield, A. 1965 Modern Capitalism. Oxford: Oxford University Press. Skocpol, T, and E.Amenta 1986 ‘States and Social Policies.’ Annual Review of Sociology 12. Skocpol, T., and J.Ikenberry 1983 ‘The Political Formation of the American Welfare State in Historical and Comparative Perspective.’ Comparative Social Research 6. Smith, A. 1961 The Wealth of Nations. Edited by E.Cannan. London: Methuen. Stephens, J. 1979 The Transition from Capitalism to Socialism. London: Macmillan. Therborn, G. 1983 ‘When, How and Why Does a Welfare State Become a Welfare State?’ Paper presented at the ECPR Workshops, Freiburg (March). Therborn, G. 1986 ‘Karl Marx Returning. The Welfare State and Neo-Marxist, Corporatist and Statist Theories.’ International Political Science Review 7. Titmuss, R. 1958 Essays on the Welfare State. London: Allen and Unwin. Tufte, E. 1978 Political Control of the Economy. Princeton: Princeton University Press. Uusitalo, H. 1984 ‘Comparative Research on the Determinants of the Welfare State: The State of the Art.’ European Journal of Political Research 12. Wagner, A. 1872 Rede Ueber die Soziale Frage. Berlin: Wiegandt und Grieben. Wagner, A. 1962 Finanzwissenschaft (1883), reproduced partly in R.A.Musgrave, and A. Peacock (eds.), Classics in the Theory of Public Finance. London: Macmillan. Weir, M, and T.Skocpol 1985 ‘State Structures and the Possibilities for “Keynesean” Responses to the Great Depression in Sweden, Britain, and the United States.’ In P. Evans et al. (eds.), Bringing the State Back In. New York: Cambridge University Press. Wilensky, H. 1975 The Welfare State and Equality. Berkeley, California: University of California Press 1981 ‘Leftism, Catholicism, and Democratic Corporatism.’ In P.Flora and A.Heidenheimer (eds.), op. cit. Wilensky, H.. and C.Lebeaux 1958 Industrial Society and Social Welfare. New York: Russel Sage. Wilensky, H. et al. 1985 Comparative Social Policy: Theory, Methods, Findings. Berkeley: International Studies. Research Series, 62.


424

56 The Political Origins of America’s Belated Welfare State* Ann Shola Orloff

* Source: M.Weir, A.Orloff and T.Skocpol, eds., The Politics of Social Policy in the United States, 1988, ch. 1, pp. 37–80.

Against the backdrop of European welfare states, the American system of public social provision seems incomplete and belated. Restricted coverage, lack of generosity to most beneficiaries, and a lack of systematic connections among different social programs—all mark the modern U.S. approach to public social provision as unusually underdeveloped, especially when one considers the relative wealth of the nation’s economy. In addition, the initiatives that established the programs constituting the American approximation to a welfare state were noticeably out of phase with developments elsewhere in the industrialized West. For some time, historians and social scientists interested in the timing of policy developments have noted only that the United States has been a “welfare laggard” relative to Europe. Policy breakthroughs leading to a nationwide system of social provision did not occur here until the passage of the Social Security Act of 1935, which established a national old-age insurance program, a federal-state unemployment insurance system, and federal subsidies for approved state programs of assistance for the elderly poor and dependent children. In contrast, European countries launched core social programs about a generation earlier, in the period between 1883 and World War I. Unlike most other advanced industrial countries, moreover, the United States has never established national health insurance. Merely characterizing the United States as a “welfare laggard,” however, overlooks some critical facts of American history, facts that doom from the outset any attempt to fit U.S. developments into the ideal-typical, European-centered pattern of welfare state development, even with allowances for delays and incompleteness. For it is not the case that America has a version of a European welfare state that was simply slightly tardy in arriving, or that U.S. social benefits will eventually “mature” into a European-style system. Not only is the contemporary U.S. system of public social provision crossnationally unusual, but also the trajectory along which the United States traveled to arrive at this system has been strikingly different from what students of European social policies have led us to believe is the inevitable course of welfare state development. The aim of this chapter is to describe and to explain the development of American social policies up through the passage of the Social Security Act of 1935. Because it is customary to treat U.S. national welfare programs as beginning in the 1930s, many people know little of what preceded this period. Thus, I will begin with a fresh overview of the overall trajectory of American social policy developments.

The Political Origins of America’s Belated Welfare State

425

An Overview of American Social Policy Histories of American social provision often presume that the federal government had no role in providing welfare benefits until the 1930s, when President Franklin Roosevelt introduced social security programs as part of his New Deal. Until then, the story goes, the needs of those Americans unable to care for themselves through participation in the labor market were addressed only, if at all, by the state and—especially—local levels of government. This account is superficially accurate but potentially quite misleading. For there was one rather spectacular exception to local predominance in the welfare field in the late nineteenth and early twentieth centuries. This was the Civil War pension system, remarkable in its own right and also consequential for later social policy developments in the United States. Originally, the federal government paid pensions only to veterans who had been disabled in the battles of the Civil War and to the dependents of soldiers killed in the war, as one might expect of a military pension system. But then, in the decades following Appomattox, the Civil War pensions were changed into de facto old-age and disability pensions that provided coverage for some one million elderly Americans, reaching about one half of all elderly, native-born men in the North around the turn of the century. While analysts today tend to overlook the social welfare function of these pensions, it was well recognized by contemporaries. In 1917, prominent social reformer Edward Devine called Civil War pensions “a main national provision for old age.”1 And University of Chicago sociologist Charles Henderson, a leading advocate of social insurance, noted in 1909 that “the military pension system has acted in great measure as a workingmen’s pension system.”2 Isaac Rubinow, another leader in the U.S. social insurance movement, wrote of the Civil War pensions in 1913: After all, it is idle to speak of a popular system of old-age pensions as a radical departure from American traditions, when our pension roll numbers several hundred thousand more names than that of Great Britain. It is preposterous to claim that the cost of such a pension would be excessive, when the cost of our pensions is over $160,000,000, or more than three times as great as that of the British system…. We are clearly dealing here with an economic measure which aims to solve the problem of dependent old age and widowhood.3 Indeed, in the period between the 1880s and World War I, while reformers, labor leaders, and politicians throughout the West initially were discussing the possibilities for adopting modern social insurance and old-age pensions to protect the “respectable” working class—and especially the “worthy aged” among them—from the indignities of the traditional poor laws, many elderly working and middle-class Americans were actually already so protected. Logically then, we might expect that U.S. reformers around the turn of the century would have been interested in building upon the Civil War pension experience to extend old-age benefits to all elderly Americans. During the Progressive Era of about 1900 to


426

1919, Americans were active participants in transnational debates over social insurance and pension proposals. Even so, only a few Progressive reformers advocated the extension of the Civil War pension system into a modern, universal system of old-age protection. These pensions were allowed to pass from existence with the dying of the Civil War cohorts, and there was no federal-level public replacement until the enactment of contributory old-age insurance plus assistance for the elderly poor under the 1935 Social Security Act. During the Progressive period, most states did adopt laws establishing workers’ compensation and so-called widows’ or mothers’ pensions—the forerunner of today’s Aid to Families with Dependent Children (AFDC).4 This period of reform also witnessed the passage of many laws regulating a wide range of industrial conditions. Yet the more expensive and administratively demanding social spending programs that formed the core of nascent systems of social protection in Europe and some parts of the British Commonwealth—old-age pensions and insurance, health and unemployment insurance— were politically unsuccessful in the United States between the 1880s and World War I. And in the wake of the war, America decisively rejected new public social protections and embraced instead the ideal of “welfare capitalism.”5 Only in the midst of a political crisis triggered by the Depression of the 1930s did Americans—belatedly, from the European perspective—initiate nationwide public social protections. Yet the delay was not without consequences for the character of the set of policies established by the 1935 Social Security Act, the “charter legislation” for the U.S. version of the modern welfare state. Two telling features of the 1935 legislation were the omission of national health insurance and the institutionalization of existing state-level differences in benefits and coverage for most programs. The Social Security Act did lead to the establishment of nationwide welfare programs, although they demonstrated varying levels of federal control. There was a fully national, compulsory system of contributory old-age insurance for those who worked in covered industries. Also established was a federal-state contributory insurance system for the unemployed in which the states were induced, but not mandated, to participate—and for which the states set benefit levels and eligibility requirements. In addition, there were optional programs of noncontributory, means-tested social assistance that remained under state-level administration, with costs to be shared by the states and the federal government. These public assistance programs, established at the discretion of the states, could aid dependent children in single-parent families and needy elderly people who did not qualify for insurance benefits. Significantly, the framers of the American social security system chose not to follow the European pattern of adding to the contributory insurance programs government subsidies financed from general revenues; nor did they establish national benefit programs financed entirely from the federal government’s revenues. How are we to make sense of the historical development of the modern system of U.S. public social provision—the trajectory of its development and the character of the Social Security Act which established almost all of the programs that compose it? Three important dimensions of analysis have emerged from previous research on the development of the welfare state; taken together, they form the elements of an institutional-political process approach. At the most fundamental analytic level—the development of institutions—we attend to the ways in which U.S. state formation shaped


427

the American political universe, within which alliances were formed and policies were formulated. In particular, we focus on the sequence of bureaucratization and democratization, two fundamental processes that transformed the political structure of America in the late nineteenth and early twentieth centuries. This changing political structure formed the context in which modern social policies were debated and met their initial legislative fate in the United States. Then, we focus on the political process, which comprises the second and third elements of this approach: the character of the political alliances that helped to launch (and delay) the American version of a welfare state and the feedback effects of earlier social policies on subsequent politics and policy debates. Before developing an explanation for the American pattern of social-welfare provision, let us first see how these analytic elements central to the institutional-political process approach help to explain the emergence of modern welfare states more generally.

Explaining the Emergence of Modern Welfare States In the last years of the nineteenth and the first years of the twentieth century, social reformers, labor leaders, and political elites across Europe and North America were actively debating social policy issues. Given the new social and political conditions created by industrialization, urbanization, the rise of capitalism, and the political incorporation of the working class, their questions centered on whether the deterrent poor law system should be replaced (at least for some groups in the population) with contributory social insurance or noncontributory pension programs. In addition, in both Europe and North America, popular movements, often based on labor organizations, pressed for new social protections, particularly for publicly funded old-age pensions that would not require tax contributions from workers. Yet neither the concern of elites about the “social question” nor the actual initiation of new welfare programs in Europe were responses simply to popular demands or to social problems newly created by industrialization, urbanization, and concomitant changes in the family. Recent comparative research on the origins of welfare states in Europe suggests, instead, that political and bureaucratic elites instituted social insurance and pension programs, utilizing existing or readily created state administrative organizations, as part of efforts aimed at the “anticipatory political incorporation of the industrial working class.”6 The issue of working-class participation in the polity—on whose terms and through which organizations—was of great concern to European elites and middle classes in this period, as extensions of the suffrage to non-propertied groups were occurring across that continent. This response suggests, in turn, that some element of political incentive, flowing from a threat to political control or from an opportunity to gain organizational or electoral advantage, especially in periods of electoral competitiveness or when new voters are entering the polity, must be operating in order to stimulate elite interest and coalition-building in the social welfare field. Indeed, a review of analyses of the initiation of social insurance and pensions suggests that the political force responsible for the introduction of new public protections was that of a cross-class coalition for new public social spending. The support of reformist elites and new middle-class groups as well as the working classes was a necessary condition for the political success of the new


428

programs in this early period. European reformers within and outside the state suggested social insurance and pensions as a means for respectable members of the working classes to avoid the cruelties of the traditional poor law. These measures were also to serve as a means for governments and propertied classes to head off the threat to the social order and to their political hegemony posed by leaving no recourse to the lower classes in times of need but the poor law system. The poor law policy itself served as the starting point for debates about what should replace it. The broader analytic implications of this point are worth noting. Policy debates are regularly informed by ideas about how best to correct the perceived imperfections of past policy, rather than simply how best to respond to social conditions as such. This means that the goals and demands of politically active groups cannot be gauged simply from their current social positions or solely from ideological and value preferences. Meaningful reactions to existing policy—a part of what we have referred to in the Introduction (M.Weir et al., eds., The Politics of Social Policy in the United States, 1988) as policy feedback, and what Hugh Heclo has called “political learning” about the “policy inheritance”—color the very interests and goals that groups or politicians define for themselves in public policy struggles.7 To implement the proposed welfare programs, policy reformers looked to state administrative organizations. In late nineteenth-century Europe, these agencies either had been recently rid of patronage practices and been professionalized through the passage of effective civil service statutes or had longstanding traditions of bureaucratic autonomy. The European experience suggests that processes of stale formation had important implications for the development of systems of modern, public social provision. First, for the new social insurance and pension programs to succeed, the state administration had to have the capacity to plan and administer relatively complex programs. Groups of influential public officials could then play key roles in formulating new social policies with existing administrative resources, pressing them on political executives and working out compromises with organized interest groups. This pattern of policy development was especially noticeable in the case of contributory social insurance. Because these programs, in addition to providing benefits, taxed workers and involved the state in activities previously within the domain of working-class voluntary associations, they often were less popular initially than noncontributory old-age pensions. Popular movements frequently arose to demand noncontributory pensions—which would not tax workers or threaten the autonomy of their organizations. The substantial popular support for state social spending initiatives could sometimes be utilized by leaders of political coalitions to overcome resistance to social insurances programs, especially if well-placed civil servants had laid the groundwork for those programs. Second, elite (and, to a lesser extent, popular) support for social policy initiatives was conditional on there being a suitable instrument for administering the new programs efficiently and honestly. After all, definitions of what is feasible or desirable in politics depend in part on the capacities and qualities that political actors attribute to state organizations and to the officials who operate them. Just as there is political learning about past policies, a kind of political learning about the government itself also occurs. Thus, the appeal of any given policy will depend to some extent on how well groups think it could be officially implemented as well as on how it may affect the fortunes of


429

groups struggling over the control of official organizations. When state organizations adequate to the administrative tasks at hand are believed to be available, elites are more likely to respond to political incentives arising from popular mobilization by building or joining a cross-class alliance to support new social spending for income protection. Finally, the process of state formation affects the operating modes of the very political organizations—especially parties and state administrative organs—through which public social policies can be collectively formulated and socially supported. In particular, the sequence in which bureaucratization and electoral democratization occur is critical in this regard, as the work of political scientist Martin Shefter on patronage and political parties would suggest.8 State bureaucratization preceded electoral democratization in most European absolute monarchies; in such instances, when political parties emerged and sought popular support, they could not simply offer the spoils of office as an inducement to voters and party activists, for access to jobs in civil administration was controlled by established bureaucratic elites. Thus, parties were forced to rely upon programmatic appeals, based on ideology or promises about how state power might be used for policies advocated by, or potentially appealing to, organized groups of constituents. At the same time, state administrations in these polities could (further) rationalize and professionalize their operations free from the partisan political concerns that dominated the calculations of elected officials. In contrast, in countries where mass electoral democratization preceded state bureaucratization, as it did in the United States, the civil administration was not protected from partisan use, and parties could use government jobs and resources for patronage. Under this set of conditions, parties would tend to rely on patronage rather than on programmatic or ideological appeals to mobilize their constituencies and reward activists. The electoral calculus of party politicians would dominate the operations of state organizations, rendering such organizations less able to use bureaucratic resources to plan autonomous state interventions into civil society. By the mid-nineteenth century, all Western states were coming under pressures to become more efficient and professionalized. Industrialization, urbanization, and the need to compete internationally, both economically and militarily, generated these imperatives. In those countries where patronage still held sway over state administrations, there were struggles over how to overcome the “political corruption” of patronage in order to create an “efficient”—that is, nonpartisan, predictable, and expert—civil service. If civil service reform succeeded before full democratization—as it did in Great Britain, for example— political parties were encouraged, as more people were granted the franchise, to change their mode of operation and to reorient their electoral appeals from the patronage system toward more programmatic appeals and constituency-based organization. The civil administration could then be changed in ways that enhanced its capacity for intervention in civil society. However, if patronage was established in a fully (or almost fully) democratic polity—as it was in the United States—it was quite difficult to uproot afterwards. Mass electorates, and the party politicians appealing to them, had a continuing interest in using government as a source of patronage, and civil service reformers had to wage difficult, eventually only partially successful, struggles to overcome democratized “political corruption” in government and party politics. Thus, to fully understand the forces that gave shape to the contemporary American social welfare system, we need to begin with the sociopolitical processes transforming


430

the American state and party system in the nineteenth century. Critical to the entire trajectory of American social policy is the fact that the United States was the world’s first mass democracy. The extension of suffrage to all white males in America had occurred by the 1830s, well before the democratization of the electorate in European countries. More importantly, this happened before the development of any political constituency interested in building an autonomous, professionalized, and nonpartisan state organization. Without such a “coalition for bureaucratic autonomy” (to use Shefter’s term), positions in the state administration were used by party leaders as resources for mobilizing the mass electorate and maintaining party organizations. This was the essence of the well-known spoils system, which dominated U.S. politics throughout the nineteenth century. Reinforcing the effects of early democratization and the dominance of patronage parties over the civil administration was the relatively protected geopolitical position of the United States. Bureaucratic state-making was not stimulated by warmaking, as in the early modern European cases so well analyzed by Charles Tilly.9 Patronage Democracy and the Civil War Pension System In industrializing America after the Civil War, parties mobilized the electorate through popularly rooted political machines whose very lifeblood was patronage. Ideological and programmatic appeals were not prominent, but politically discretionary distributional programs that provided jobs, services, or other goods, such as Civil War pensions, were quite popular with politicians and their constituencies.10 These distributional policies were of critical importance to elected officials, for it was in the late nineteenth century that American politics were at their most competitive outside the South. In the North and Midwest, Democrats and Republicans faced each other with nearly equivalent popular electoral support, and they regularly replaced each other in office. Electoral participation was at an all-time high, and given the nearly equal division of support, a few hundred votes—however secured—in the most politically competitive states, such as Indiana, Ohio, Illinois, and New York, could mean the difference between winning and losing an election.11 The Civil War pension system was an excellent example of the sort of policies generated by the operations of patronage democracy. It allowed politicians, especially Republicans, to channel to many individuals in numerous (non-Southern) communities pensions financed out of the surplus revenues coming from the constantly readjusted tariffs they sponsored to benefit various industries and areas.12 Given the extensive mobilization of voters and the extremely competitive electoral conditions of late nineteenth-century patronage democracy, it is not surprising that during the 1880s and 1890s, the pension system was changed from a provision for compensation of combat injuries and war deaths to the functional equivalent of an old-age and disability pension system for a politically important segment of the U.S. electorate. In the Northern and Midwestern states, veterans of the Civil War constituted fully 12 to 15 percent of the electorate, making the “soldier vote a prize of great worth,” a “prize” that increased in value with the growth of the Grand Army of the Republic (GAR), the veterans’ lobbying group.13 Figure 1 helps to show the effects of the transformation and expansion of the post-Civil


431

War pension system by charting the trends in disbursements, number of pensioners, and private pension bills between the late 1860s and 1920. Benefits under the original 1862 law were extended only to soldiers actually injured in combat or to the dependents of those disabled or killed. As one might expect, the number of beneficiaries and total expenditures were falling off in the late 1870s. Subsequently, however, under the pressure of the intense electoral competition of the times, both Republican and Democratic Congressmen supported legislative liberalizations, the most important in 1879 and 1890, that effectively transformed the character of the pension system.14 The 1879 Arrears Act allowed soldiers who “discovered” Civil War-related disabilities to sign up and receive in one lump sum all of the pension payments they would have been eligible for had they been receiving benefits since the 1860s! Then the 1890 Dependent Pension Act severed altogether the link between combat-related injuries and benefits under the pension system. Any veteran who had served ninety days in the Union Army, whether or not he saw combat or was injured in the Civil War, could sign up for a pension if at some point he became unable to perform manual labor. In practice after the 1890 legislation, old age alone became a sufficient disability, and in 1906 the law was further amended to state explicitly that “the age of sixty-two years and over shall be considered a permanent specific disability within the meaning of the pension laws.”15 By this time, about 90 percent of the surviving Union veterans were pensioners. In essence America had a de facto old-age and disability pension system for those who had served in the Civil War and a survivors’ allowance for many of their dependents.16

Figure 1: The Expansion of Civil War Pensions, 1866–1917


432

Source: William Glasson, Federal Military Pensions in the United States (New York: Oxford University Press, 1918, pp. 273, 280. Members of Congress were kept quite busy by the operations of the pension system in the 1880s and 1890s. Not only were there the recurrent legal liberalizations of the terms of eligibility for pensions, but, in addition, congressmen and senators devoted a large amount of time to helping their constituents establish their eligibility for pensions through personal intervention with the U.S. Pension Bureau and by sponsoring thousands of special private pension bills tailored to allow specific individuals to collect benefits even when they did not qualify under the very liberal legislated qualifications.17 The U.S. Pension Bureau itself was the “most uncompromisingly political branch of the late nineteenth century federal bureaucracy,” where there was wide discretion in the processing of claims due to a huge backlog.18 Under such circumstances, control of the civil administration was an important political resource. The partisan appointees in the Pension Bureau, most notably the Republicans, utilized this resource particularly at election time, when pension agents would be sent into the field to sign up pensioners— for their benefits and for the Grand Old Party!19 Given that a considerable proportion of men then in their twenties and thirties served in the Civil War regiments, and given that these men were in their sixties by 1890 to 1910, it is not surprising that at least one-half of all elderly, native-born men in the North, as well as many old and young widows, were receiving what were in effect federal old-age and survivors’ pensions during this period. Of course, the post-Civil War pension system


433

was uneven in its coverage of the American population, with the distribution of pensions favoring native-born men and pre-Civil War immigrants living in the North and Midwest. A majority of the elderly white Southern participants in the Civil War were veterans of the Confederate Army and therefore not eligible for federal benefits, although most of the Southern states began to give (relatively meager) state pensions to disabled or impoverished Confederate veterans in the 1890s.20 Also excluded were blacks and all post-Civil War immigrants; this meant that relatively few unskilled workers were among the system’s beneficiaries. It seems clear that benefits under the Civil War pension system flowed primarily to members of the middle class and the upper strata of the working class, rather than to the neediest Americans. Yet in Europe it was precisely the “respectable” and better-off members of the working classes who were of most concern to those reformers proposing social insurance and pensions. The problem of the pauperization of these “worthy” (and often newly enfranchised) people through the operation of traditional poor laws, which left no alternative to destitution but degrading, insufficient, and disfranchising poor relief, was in fact a major stimulant to the consideration of social policy reform in Europe. In the United States, not every Civil War pensioner would have been a pauper without the aid of a federal pension, but as Charles Henderson wrote in his important 1909 work, Industrial Insurance in the United States, “Many of the old men and women who, in Europe, would be in almshouses, are found in the United States living upon pensions with their children or in homes to which paupers are not sent and they feel themselves to be the honored guests of the nation for which they gave the last full measure of devotion.”21 Although Civil War pensions were a form of public aid expanded to meet the needs of some fortunate older Americans, not all who were elderly, disabled, or impoverished around the turn of the century received Civil War pensions. For some unemployed people, there were occasionally positions in work-relief projects funded by municipalities during business downturns, when the plight of the jobless was most visible.22 But for the non-able-bodied poor, the only alternative to destitution was charity. Charity, whether dispensed by local public poor law authorities or private philanthropic agencies, was at best inadequate. At worst, it involved disfranchisement and residence in a semi-penal workhouse. Significantly, public poor relief was offered as an alternative to, rather than as a right of, citizenship.23 The financing and administration of the poor laws throughout nineteenth-century Europe and North America had several commonalities—local administration was the rule, and poor relief everywhere was meager and degrading.24 The distinguishing feature of late nineteenth-century U.S. poor relief administration was the predominance of patronage practices. A wide variety of governmental forms existed in American localities, but whatever the form, the patronage organizations of Democrats and Republicans dominated city, town, and county governments.25 The partisan control of American local governments meant—as did the dominance of patronage politicians in the federal government—that administration was nonprofessional and “permeated…to the core” by party politics.26 Control of poor relief provided politicians with opportunities for patronage appointments and for exercising partisan preference in the awarding of contracts for building and supplying poorhouses, asylums, and other welfare institutions.27 In addition, because local officials had discretion in granting outdoor relief


434

and work relief, electoral considerations could enter easily into the determination of eligibility for aid.28 Describing the use of public works as a relief measure in New York City in the 1890s, sociologist Leah Feder noted that “partisan politics entered into its expenditure; workers referred by social agencies were dropped more quickly than those sent by politicians…carelessness, extravagance, and misappropriation in administration were rife…[although] the fund undoubtedly provided an important resource in relieving unusual distress.”29 Public charity was sometimes supplemented by the gifts of party bosses, who on occasion distributed food and fuel to the poor. Typically, they paid for this largesse out of the “private” funds they garnered through graft or “mating” (the practice of levying assessments on the salaries of public employees).30 Late nineteenth-century patronage democracy was effective in mobilizing broad popular support, but it also produced serious opposition. Populists and labor groups denounced the links between politicians and the “robber barons”—the capitalists dominating the economy through their control of railroads, manufacturing, and banking, newly national in scope. At the same time, university-educated professionals criticized especially the “corruption” and “inefficiency” of the personnel practices of the spoils system. They were dissatisfied as well with the kinds of policies fostered by patronage democracy, which, they charged, lacked any justification in terms of the “public interest.” Instead, noted critics, the development of American public policy was above all tied to the partisan political exigencies of winning elections. These exigencies did not at all preclude—in fact, probably stimulated—out-and-out corruption and fraud. Even the normal functioning of the patronage system was coming under fire as “political corruption.” For example, the chairman of the committee on the merit system in public institutions of the National Conference of Charities and Correction (NCCC), America’s leading association of charity workers and social reformers in the late 1800s, forcefully challenged the rationale and practices of the spoils system, calling that system “treasonable robbery, although not treason in the eyes of the law.”31 The Civil War pension system was targeted as a particularly egregious example of the problems to which patronage democracy led. By the 1880s and 1890s, the pension system’s partisan uses were clear to many in the broad middle-class public, despite the attempts of elected officials to legitimize the repeated expansions of the system through the use of patriotic rhetoric and waving the “bloody shirt.” As one observer noted in an investigatory article appearing in an 1884 issue of the reform-oriented magazine Century, “It is safe to assert that most of the legislation adopted since the war closed, to pay money on account of service in the Union armies, has had for its real motive not justice nor generosity, but a desire to cultivate the ‘soldier vote’ for party purposes.”32 For the remainder of the nineteenth century and in the Progressive Era of the twentieth, these “abuses” of the pension system and other “perversions of democracy” associated with mass patronage politics provided a continuing stimulus for political reform, as well as a powerful symbol of all that was considered to be amiss with existing political practices. The initial proponents of ending patronage through civil service reform in the United States were “Mugwumps,” upper- and upper-middle-class reformers located in the Northeast, especially New York and Massachusetts. Like their European counterparts of the late nineteenth century, the Mugwumps wanted public administration taken out of patronage politics, so that expertise and predictability could prevail.33 At first, however,


435

the Mugwumps’ reform proposals made only limited headway, for American party politicians had secure roots in the fully democratized, tautly mobilized mass white-male electorate. As Stephen Skowronek has pointed out in his study of American statebuilding, “to build a merit system in American government, government officers would have to move against resources and procedures vital to their power and position.”34 Through the last quarter of the nineteenth century, the conflict between politicians and the various good government reformers resulted in overall defeat for the Mugwumps. Following the extremely close presidential election of 1880 and the patronage-related assassination of President James Garfield, the Republican administration of President Chester Alan Arthur and the Republican-controlled Congress in early 1883 approved the establishment of a Civil Service Commission under the Pendleton Act, largely as a concession to the Mugwumps to avoid losing their votes.35 Analysts agree that the impact of the law was quite limited; Skowronek notes that it dried up “selected pockets of patronage so as to improve efficiency and, at the same time, serve strategic party goals…; [it] supplemented the dominant patronage relationship rather than supplanting it.”36 In fact, because of the growth in federal employment, a larger number of jobs were available for patronage by the end of the century than had been before the enactment of the Pendleton Act, even with the growth in the proportion of classified (i.e., civil service) jobs. Thus, significant civil service reform was delayed until the Progressive Era, when electoral competitiveness declined somewhat, although even then, reformers never fully succeeded in uprooting patronage. Nevertheless, the crisis of the patronage system formed the backdrop to the policy debates of the early twentieth century. The Progressive Era: Debate and Defeat of Proposed Social-Spending Measures Around the turn of the twentieth century, Americans, along with people in almost all the industrializing nations of the West, took part in debates over what the state should do in the face of the increasingly well-publicized problems of income insecurity. But the political legacies of nineteenth-century patronage democracy were far from conducive to the establishment of modern social insurance and pensions in Progressive-Era America. The incomplete success of civil service reform in the late nineteenth century had left the U.S. state with relatively underdeveloped capacities for regulating an industrial-capitalist economy and administering social programs designed to deal with its casualties. In addition, Progressive reformers were in the midst of battles against democratized “political corruption”; these overshadowed debates about social policies. At this point in American history, modern social-spending programs were neither governmentally feasible nor politically acceptable. Workers’ compensation legislation was the first modern welfare program to receive consideration in America, as it was across Europe, and it was substantially successful, achieving passage in thirty-eight states by 1919.37 After the passage of workers’ compensation laws, many European reformers had moved toward dealing with the situation of the aged poor, whose plight had helped to stimulate popular movements for pensions. As pensions were established, reformers and government officials were concerned to add a contributory element to programs designed to deal with


436

unemployment, sickness, and dependency, and the political support for pensions was utilized in the service of social insurance. For example, in Britain, the problems of the aged served as a unifying focus for the political activities of reformers and some labor leaders, helping to cement a cross-class alliance in favor of new social-spending programs.38 In contrast, in the United States, the problems associated with the aged— poverty, but also mismanaged Civil War pensions—served to drive apart groups who might otherwise have cooperated politically to push for a modern system of social protection. Thus, the trajectory of events in the United States was quite different from that in Europe, and the successful establishment of workers’ compensation in many states was not followed by similar successes in instituting old-age pensions or social insurance. Although they hesitated to press for new social-spending measures, Progressive reformers were not anti-statist in their orientation. On the contrary, they saw the state as having the potential to be an ethical agency, and this vision helped to inspire the movement for civil service reform. Therefore, it is important to be very specific about exactly what kind of cross-class coalitions were not possible in Progressive-Era America. Although middle-class reformers and organized labor were sometimes at odds over the issue of political “reform”—especially in Eastern cities where political machines had made inroads into working class constituencies—they did manage to cooperate on campaigns for workmen’s compensation, labor standards legislation, and mothers’ pensions. In general, middle-class reformers supported the expansion of the regulatory capacity of the government, which in fact kept pace with European developments.39 What can account for the American pattern of social reform legislation in the years between the turn of the century and World War I? Why were social-spending initiatives so weakly supported by elites and ultimately unsuccessful? And why did other welfare innovations, such as workers’ compensation and labor regulations, fare better? Conventional historical portraits show Americans as culturally and ideologically opposed to new state welfare programs. The failure to enact new social insurance and pension programs is explained by invoking the strength of traditional U.S. liberalism among both educated elites and the population at large.40 In fact, similar ideas were employed in support of the reform proposals in both Europe and America, as early social scientists demonstrated empirically the social causes of poverty and income insecurity, and politically active innovators grappled with the problem of integrating industrial workers into the polity.41 In liberal-democratic countries such as Britain and the United States, support for new policy proposals came from a reworking of traditional liberal ideas away from pure self-help and distrust of state intervention toward “new liberal” or “progressive” conceptions.42 Those inspired by “new liberal” ideas came to see that industrial society made people interdependent. Many noted that individual liberty could be thwarted as much by a lack of life’s necessities as by the presence of governmental restrictions.43 In modern society, argued new liberals, government could be an indispensable support for individual liberty, providing security against socially caused misfortunes and regulating competition to allow for individual initiatives. Thus, new state welfare activities could be—and were, on both sides of the Atlantic—advocated without violating the traditional liberal aim of enhancing individual freedom.44 What is more, turn-of-the-century American popular support for new welfare measures has been underestimated in conventional historical accounts. Consider the voting in the


437

presidential election of 1912. The Progressive Party, led by former President Theodore Roosevelt, garnered the second-highest vote total, gaining four million votes to Democratic candidate Woodrow Wilson’s six million, and edging out incumbent Republican President William Taft. The Progressives achieved this impressive vote total with a platform that endorsed “the protection of home life against the hazards of sickness, irregular employment, and old age through the adoption of a system of social insurance adapted to American use.”45 More precise and convincing evidence comes from Massachusetts, where eight cities held popular referenda on old age pensions in 1915 and 1916. Pensions were approved by Bay State voters by a margin of more than four to one, belying the notion that American popular values precluded this policy innovation.46 Organized labor in the United States prior to the New Deal also has been described as uniformly hostile to social welfare programs.47 It is true that some national leaders of the American Federation of Labor (AFL)—most notably AFL President Samuel Gompers— opposed contributory social insurance, which would tax workers. Yet in the pre-New Deal era, most legislative activity on labor and welfare matters took place at the state and local levels, and it was here that labor unions and federations of labor exercised most of their political influence.48 In the more industrialized states, where organized labor was relatively more weighty than in the United States as a whole, unions were often supportive of both pensions and contributory social insurance. During the Progressive Era, state labor federations endorsed a number of health and unemployment insurance bills, along with workers’ compensation, mothers’ pensions, and old-age pensions. The recession of 1914–15 induced reformers and labor organizations alike to support steps toward implementing unemployment insurance in California, Massachusetts, and New York.49 By 1918, nine state labor federations had endorsed unemployment benefits proposals.50 Likewise, the labor federations from the industrialized states of Ohio, New Jersey, Pennsylvania, and Missouri, in addition to those from California, Massachusetts, and New York, endorsed health insurance proposals between 1915 and 1919.51 Interestingly, events in Britain (and elsewhere in Europe) reveal that it was not working-class groups who were responsible for the introduction of contributory social insurance. Instead, it was politicians and ministers who persuaded initially reluctant leaders of unions and other interest groups to go along with the new programs by negotiating about specific methods of implementing taxes and payments.52 Since U.S. state and local unions were at least as willing as British labor groups to support social insurance programs, it is quite likely that comparable co-optive efforts directed at them by state-level elected or appointed officials would have succeeded. In the United States, government leaders did not make such efforts—and it was surely this lack, rather than lack of demands and political support from the unions, that explains why the social insurances failed in America. In the case of old-age pensions, it is even clearer that the failure of this form of social provision in Progressive-Era America was not due to the failure of U.S. unions to support such measures. Leaders at all levels of the labor movement generally favored noncontributory old-age pensions. Even Samuel Gompers favored noncontributory oldage pensions, stating in 1916 that pensions “carry with them the conviction of their selfevident necessity and justice,”53 Moreover, national conventions of the AFL passed resolutions in favor of pensions in 1908, 1911, 1912, and 1913.54 Indeed, the first


438

pension proposal introduced in the U.S. Congress came from former United Mine Workers leader William B.Wilson in 1909, and it was soon endorsed by the AFL and by many state and local labor leaders.55 The Wilson bill borrowed from the positive symbolism of Civil War pensions by proposing to deal with poverty among the workingclass aged through the creation of an “Old Home Guard,” in which all Americans aged sixty-five or over were invited to enlist if their annual income fell below $240; their sole duty would be to report to the War Department on the state of patriotism in their communities, for which they would be paid $120 per year.56 During the Progressive Era, it was not working-class groups, but many social reformers—and the middle-class and professional strata from which they came and to whom they oriented their arguments—who were reluctant to accept, let alone champion, social-spending measures such as old-age pensions. These sorts of programs had been advocated by a cross-class alliance in Britain and elsewhere in Europe, but in the United States, such a coalition was not politically possible. This can be seen in the way that U.S. reformers dealt with the issue of old-age poverty—the issue that had brought together a cross-class coalition for social spending in Britain. The first serious official investigation of old-age poverty occurred in Massachusetts in the wake of the introduction of several old-age pension bills in the state legislature from 1903 to 1906. The Massachusetts Commission on Old Age Pensions, Annuities, and Insurance was appointed in 1907 and issued its report in 1910. This investigatory commission, staffed predominantly with professionals and upper-class Bostonians, gathered data on the elderly poor in Massachusetts, considered various pension proposals, and reported that public pensions were neither necessary nor morally desirable. Given that Massachusetts had long been a pioneer in social and labor legislation, serving as a gateway to the United States for British social policy innovations and as an example to other states, the setback in the Bay State was critical to developments everywhere in America. Indeed, the report dealt a virtual death blow to what had previously been a promising movement toward old-age pensions in that state and elsewhere.57 A labor representative, Arthur M.Huddell, expressed his disagreement with the commission’s recommendations and questioned their arguments about the preferability of contributory pensions over noncontributory pensions, citing Civil War pensions as a favorable precedent: “The pension to the veterans of the Civil War has built up the American family [and] the old veteran and his widow are made comfortable in their old age…and have a feeling of independence that old people should have.”58 But the willingness of Huddell and other reformers from labor’s ranks to endorse noncontributory pensions for the “veterans of industry” was not complemented by a similar attitude on the part of elites in Massachusetts or in the rest of the United States. Throughout Progressive-Era America, old-age pensions were downplayed even by the most socially progressive elites, such as the members of the American Association for Labor Legislation (AALL). In 1913–14—just as the AFL was solidifying its commitment to old-age pensions—the AALL made the decision to promote health insurance rather than old-age pensions as the “next great step” after workers’ compensation in the “inevitable” progress toward a comprehensive program of modern social protections in America.59 Following the massive unemployment in the recession of 1913–15, the AALL also advocated a model bill for unemployment insurance, along with other measures to


439

help the jobless.60 Yet leaders of the AALL decided to forego a campaign for old-age pensions after their successes in promoting workers’ compensation and industrial safety laws. Such a campaign assuredly would have appealed to organized labor and to broader working-class electorate—as pensions did in Britain, for example. A campaign for contributory social insurance was less able to appeal to all segments of the labor movement, particularly since there had been no momentum for new government socialspending programs built up from prior reform activities on behalf of pensions. Yet pensions were unappealing to many of the leaders and members of the AALL itself as well as to the broader middle-class political public. Why did U.S. elites and middle-class people generally oppose new state socialspending initiatives? And why were the capacities and character of the U.S. state not conducive to the introduction of modern welfare measures? To understand these facts, we must turn to the institutional context formed by the particular patterns of American statebuilding, and to the ramifications this had for coalition-building and for the character of parties and the civil administration. In Progressive-Era America, civil service reform made some headway, though less at the national level than at the state and local levels.61 The demand for reforms in government broadened from the very elite ranks of Mugwumpry to include the growing ranks of the educated, professionalizing middle class and, in many places, farmers and organized labor as well.62 In the newly noncompetitive political situation ushered in by the realigning election of 1896, there was increased freedom for officials in the executive branch, especially the president himself, to move against patronage practices, and to establish bureaucratic organizations useful in augmenting executive power.63 Yet the legacies of nineteenth-century patronage democracy and the crisis it precipitated in the Progressive Era created a relatively unfavorable context for the enactment of social reforms such as old-age pensions and social insurance. At the most basic level, the civil administration of the early twentieth-century American state was quite weak, given the lack of an established state bureaucracy and the dispersion of authority inherent in U.S. federalism and division of powers. Civil service reform, which might have enhanced the capacities of American government for certain types of interventions in civil life, had still not achieved much progress by the early 1900s. In contrast to the situation in many European countries, there were in the United States no strategically placed officials to pave the way for new social welfare programs. Thus, reforms in this period did not have their source within the underdeveloped American bureaucracy. Instead, they were supported by broad coalitions of interest groups and pressed upon state legislatures.64 American political parties did not tend to be the vehicles of reform, either. They were not—as were many of their European counterparts—programmatic parties looking for new policies to attract newly enfranchised working-class voters. Rather, they were patronage parties with an already mobilized working-class following whose mode of operation was under attack as “corruption” by many reformers. The clear challenge for party leaders in the Progressive Era was to find ways to appeal to middle-class reformers and their organizations while not alienating their traditional, patronage-oriented supporters. New social-spending measures tended not to meet that challenge, for elites and middle-class groups were still not convinced that officials in control of the state


440

administration had been sufficiently cleansed of “corruption” to be entrusted with such activities. The possibility of continuing “political corruption” in new government interventions worried elites and middle-class groups. The struggle against the patronage system was far from complete, and the debate over new social welfare spending programs was additionally entangled by issues of administrative probity and competence. By contrast, in most European countries, where bureaucratization was effectively implemented prior to mass extension of the suffrage and the emergence of debates over workingmen’s insurance, such entanglements were avoided. Thus, even as proposals for new social insurance or pension schemes were being debated, Civil War pensions continued to be a symbol of all that was wrong with mass patronage democracy and public policy. In 1911 and 1912, Charles Francis Adams, a well-known journalist and reformer, carried out an investigation of the abuses of the military pension system which was published in the reform magazine World’s Work.65 The series’ title when reprinted as a book in 1912— The Civil War Pension Lack-of-System, a four-thousand million dollar record of legislative incompetence tending to political corruption—summarizes well the reformer’s view of the link between patronage democracy and pensions. Although worries about political corruption undermined cross-class coalitions for social spending, especially on old-age pensions, some important social-welfare innovations did occur in many states during the Progressive Era. Labor regulations were strengthened, especially for women and children, and mothers’ pension legislation, which provided means-tested allowances to widows outside of the poor law, was also widely successful. Workers’ compensation laws, too, were enacted by numbers of state legislatures.66 Why was it possible for these to be enacted in a political climate so hostile to public initiatives that involved social spending? Regulatory activities, unlike socialspending programs, could address the new needs of industrial society—and appeal to both working- and middle-class voters—without adding to the potential for political “corruption” or overtaxing U.S. administrative capacities. The two new welfare programs that were successful in Progressive-Era America, workmen’s compensation and mothers’ pensions, represented the reworking of government functions already being carried out in the courts, and they did not significantly increase government spending. In Massachusetts and Wisconsin, reformers were able to overcome business opposition to enact legislation establishing savings bank and government life insurance, programs that were not perceived to have a high potential for abuse.67 Progressive social reformism quickly lost its momentum with the entry of the United States into World War I. However, with the creation of centralized federal administrative agencies to mobilize American resources for the war effort, many reform-minded Americans expected that the new agencies could be reoriented to peacetime activities in the aftermath of the war. Of particular importance to reformers were the United States Employment Service (USES) and war risk insurance. The USES, set up as an independent agency in 1918, was operating over eight hundred field offices at its peak of operations in 1919. These offices provided some job referrals and also collected statistics on unemployment. Members of the AALL and others interested in problems of unemployment and labor market organization hoped that the USES would come to serve as the foundation for the administration of unemployment insurance, as had been the case


441

in Britain, where the world’s first unemployment insurance system (initiated in 1911) had soon followed the establishment of national labor exchanges (1909).68 War risk insurance (established by law in 1917), explicitly designed to avoid the pitfalls associated with the post-Civil War pension system,69 was also the cause for hopeful projections on the part of progressive reformers. Samuel McCune Lindsay, in his 1919 presidential address to the AALL, described soldiers’ insurance as “so striking a forward step that it may almost be said to atone for our previous backwardness,” and he called on “those who believe in social insurance…to see that our next step shall be to hold on to this gain” and to extend it to the civilian population.70 Yet reformers’ hopes for extending wartime organizations and programs were dashed in the aftermath of the war. In Europe, the effect of national mobilization for war was to stimulate political trends favoring social insurance, but in the United States, the opposite happened. Congress—giving political expression to the opposition of state, local, and private interests to the extraordinary federal powers developed during the war—quickly dismantled the emergency war organizations, including the USES, after the Armistice. When Congress slashed the appropriations for USES in 1919, the agency was forced to disband its field offices and to depend for unemployment statistics on voluntary submissions of information from business, states, and localities.71 Without sufficient funding and support for the USES, there was little immediate chance for the development of an unemployment insurance program. And in the postwar reaction against federal government intervention in civil society, public old-age, disability, and survivors’ insurance for civilians was not to be built up from war risk insurance. In essence, the postwar United States opted not to institute the modern public social provision considered during the Progressive Era, or to build upon and extend wartime government activities such as the USES or war risk insurance. Yet the Progressive Era attempts to build new realms of public activity free from patronage were not without implications for the future of social policy. Two structural effects were the enhancement of the importance of sub-national governments and the intensification of administrative fragmentation inherent in a federal system. Partially because civil service reform had made most headway at the sub-national level, states were the locus of administration for successful labor and welfare innovations in the Progressive Era itself and into the 1920s, and interests in such state-level programs became institutionalized. When social reform again emerged on the policy agenda in the wake of the Great Depression, policymakers would have to contend with this Progressive state-building legacy as they designed new programs. In the postwar years, reformers and political leaders, as well as many average Americans, believed that businessmen could and would look after the American economy and Americans’ economic well-being. To the extent that the problems of income insecurity associated with industrial capitalism were given any attention at all, they were to be addressed through “welfare capitalism”—programs run by employers for their workers—or through private and local charity. Welfare capitalist schemes reached only a tiny proportion of the work force, but they were important in the 1920s as a cultural ideal in opposition to public social efforts.72 The conservative Republican presidents of the 1920s were not at all interested in building a professionalized bureaucracy to administer social programs; rather, they embraced the vision of an anti-bureaucratic “associative


442

state” developed by Herbert Hoover, who served as secretary of commerce for Presidents Warren Harding and Calvin Coolidge before becoming chief executive himself in 1928. Hoover worked in his capacity as a public servant to encourage research, planning, and cooperation on the part of private groups; these activities would help citizens, under the “enlightened” leadership of corporate capitalists, to “meet the needs of industrial democracy without the interference of government bureaucrats.”73 A striking example of the “associative state” approach to social problems was the 1921 President’s Conference on Unemployment, organized by Hoover to gain backing from the invited politicians, labor leaders, and social policy experts (including some from the AALL) for plans— already largely formulated—to deal with the severe recession of that year. Harding and Hoover ruled out any kind of governmental action from the beginning, and the conference was left to recommend better coordination of private and local charitable efforts.74 Social reformers recognized that the 1920s were a remarkably unpropitious time for advocating public solutions to the problems of industrial capitalism.75 Many—most importantly the members of the AALL—accommodated themselves to the new political climate by downplaying state initiatives and emphasizing the so-called “preventive approach” associated with the work of University of Wisconsin economist John R.Commons.76 Commons and his disciples called for labor and social insurance legislation that would give individual businessmen tangible financial incentives for increasing workplace safety, stabilizing employment, and otherwise improving workers’ and social welfare. Expert, coordinated administration of labor regulations at the state level was also an important part of the Commons approach to industrial relations.77 It was in his home state of Wisconsin, with its unusually strong network of ties between state government and state research university, that this type of administration was best developed. The Wisconsin State Industrial Commission was staffed by labor economists schooled in the Commons philosophy. The commission engaged in research and was empowered by the legislature to enforce and adjust all the state’s industrial regulations.78 In the absence of other reform activity, the AALL became increasingly oriented to the experiences and policy ideas of the Wisconsin State Industrial Commission, where— almost alone among state government agencies in the 1920s—social reform-minded “experts” (albeit of the “preventive” stripe) were still in a position of political influence. Only in the state of Wisconsin was there continuing activity around proposed unemployment insurance legislation through the 1920s. In 1921, an unemployment insurance bill based on the preventive approach and including employers’ incentives came close to passing in the state legislature. The reformers continued their campaign until a similar bill finally succeeded in 1932—the first in the United States.79 Outside Wisconsin, popular political activity for welfare reform in the 1920s consisted almost exclusively of state-level campaigns for noncontributory old-age pensions led by the Fraternal Order of Eagles, a predominantly working- and lower-middle-class, white fraternal order. At times, the Eagles worked in uneasy alliance with professional, middleclass reformers of the AALL or the newly formed American Association for Old Age Security (later the American Association for Social Security); a handful of progressive Republican and Northern Democratic politicians supported the pension cause as well.80 But by 1928, the only fruits of the extensive campaigning for new public spending for


443

old-age protection were six state-level pension laws, all of them “county-optional,” meaning that counties were allowed, but not mandated, to pay pensions to some of their aged residents. As it turned out, not many counties exercised their option, and only about one thousand elderly people were receiving these pensions in 1928, fewer than those still collecting Civil War pensions!81 From Welfare Capitalism to the Welfare State It was in the years of the Great Depression that the United States initiated nationwide programs of social protection through the passage of the Social Security Act of 1935. The economic crisis transformed the American political world as it had existed in the 1920s and set into motion political changes that allowed for the success of the New Deal socialwelfare legislation. Nationally, the ideological and policy inheritance represented by welfare capitalism was utterly discredited. And American citizens, politically quiescent throughout the 1920s, became politically engaged on a mass scale. The social insurance and assistance legislation passed during the Depression era bears the imprint of these factors, but it was also shaped by institutional arrangements and policy feedbacks that were themselves the legacy of the struggles for civil service reform and social welfare innovation of the late-nineteenth and early-twentieth centuries

The Role of Popular Pressures in New Deal Social-Welfare Initiatives By the time of the 1932 election, close to four years of the Great Depression had shattered the ideas that businessmen could guarantee the economic and social well-being of Americans, and that limited private and local charity and corporate welfare programs could serve as a substitute for public social provision in times of economic downturn. President Hoover had responded to demonstrably increased need and to rising demands for federal relief within the framework of his voluntarist philosophy, in which the state’s welfare role was limited to encouragement of private philanthropic efforts.82 But with the persistence of the Depression, and the blatant failure of voluntary efforts to alleviate need, the demands for federal action to cope with the economic crisis mounted. Popular and expert interest again turned toward public policy solutions to economic and social problems. Franklin Roosevelt’s mandate was hardly definitive in terms of what national policies he would pursue to achieve economic recovery or to end economic insecurity; the 1932 platform of the Democrats pledged only old-age and unemployment insurance at the state level.83 Yet Roosevelt was known as a reformer—when he served as governor of New York, he encouraged the initiation of public programs for old-age protection—and though he made no pledge as to what he would do beyond “bold experimentation,” his campaign promised to involve the federal government actively in dealing with the problems caused by the economic crisis.84 Frances Perkins, FDR’s close advisor and labor secretary, later wrote that it was “basic in the appeal for votes that suffering would be relieved immediately” through federal action, a stance completely different from Hoover’s.85 In this way, popular support for a new political orientation toward problems of economic


444

security, as expressed by the 1932 election, was critical in providing an opening for the Roosevelt administration or congressional Democrats to initiate new public social programs. The democratic upsurge in American politics did more than sweep Roosevelt into office, of course. There was increased mass political activity throughout the decade—in addition to an increase in voting among new-stock urban dwellers, there were marches of the unemployed, sit-down strikes, and farmers’ protests.86 Of critical importance to New Deal social policy developments were the social movements that arose demanding extended government welfare activities and new public social spending: Huey Long’s “Share Our Wealth” movement, Father Coughlin’s National Union for Social Justice, and Dr.Francis Townsend’s old-age pension movement.87 Indeed, an explanation commonly offered for the New Deal social-welfare breakthroughs is that they resulted from popular pressures.88 It is certainly true that such pressures played a critical role in the formation of the American system of social insurance and welfare, but it is not the case that the Social Security Act was a direct product of mass movements. How, then, did popular demands for new state welfare activity affect the course of policy development? During the 1930s, the mass movements dedicated to expanded public social protection continued to press for state action—indeed, they made such action a political necessity. Without popular pressure, and the electoral incentive it represented to congressmen, the policy initiatives taken by the Roosevelt administration probably would not have been successful. Secretary of Labor Frances Perkins, a key policymaker for social security, saw popular political activity as creating a unique political opening for the enactment of social insurance programs.89 She later recalled, in reference to the passage of old-age protections in the New Deal, that “without the Townsend Plan it is possible that the Old Age Insurance system would not have received the attention which it did at the hands of Congress.”90 Indeed, given the character of the American state structure in the 1930s, executive policy initiatives depended for their success on mass pressure being exerted on Congress. The divided authority of the American state meant that extremely broad coalitions had to come together to overcome the many potential vetoes and to achieve coordination among the various branches and levels of the government; this was especially the case in the New Deal period, before the emergence of the bureaucracy as a substantial factor in policymaking.91 Yet popular demands were not directly transmitted into the circles of policy formation. U.S. political parties were still far from being programmatic or committed to coherent social policies, and they were not organized to channel the policy proposals of groups to the elected and appointed officials who actually made policy, as did many European parties.92 In fact, the very strength of mass movements pressing for the extension of state welfare activities had a somewhat paradoxical effect: it reinforced the fears of Roosevelt and his advisors that “unwise” legislation might be enacted, and thereby intensified their cautiousness on matters of social policy and their determination to exclude radical voices from policy discussions.93 This cautious approach in response to mass pressure is perhaps best illustrated in the case of old-age protection. The Townsendites were the largest and politically best connected of the various organizations pressing for new state welfare activities. The movement consisted of thousands of elderly Americans, many of them middle-class,


445

organized into “Townsend Clubs.” The Townsend movement was especially influential in Congress, both because of the elderly’s high propensity to vote and because of the organizational structure of the movement, with local clubs in almost every congressional district.94 In contrast to the usual maximum monthly payments of about $30 under the state old-age assistance programs that had proliferated after 1929,95 the Townsendites advocated a rather remarkable $200 monthly pension for all Americans aged sixty or more, given on the condition that they cease working and spend the pension payment within thirty days.96 The Townsend Plan and others like it, though clearly popular with many Americans, were completely dismissed by the Roosevelt administration officials drafting the Social Security Act’s old-age pension and insurance provisions.97 Roosevelt insisted to his advisors that any program instituted be on a sound actuarial basis—as opposed to the “unsound” Townsend plan and other popular “panaceas.” This meant some sort of contributory scheme, as opposed to one drawing exclusively on general revenues.98 Noting that “Congress can’t stand the pressure of the Townsend Plan unless we have a real old age insurance system,” FDR resolved to take the initiative in the development of social programs, and to maintain careful control of the entire process of legislating social security.99 Looking toward the future, Roosevelt was especially concerned to insulate his legislation from popular and political pressure for expansion by including contributions from beneficiaries. Roosevelt’s timing in introducing permanent as opposed to “emergency” measures also reflects the cautiousness of his approach. Planning for social insurance initiatives was undertaken immediately, but no legislation was proposed by the administration. In contrast, the president introduced federal emergency relief legislation almost immediately upon taking office, thus responding to the plight of the unemployed—and to the not inconsiderable protests of state and local welfare officials, whose agencies were overwhelmed financially by the proportions of need produced by the crisis.100 By 1934, the year Roosevelt appointed the Committee on Economic Security (CES)—the group responsible for drafting the administration’s proposed social security legislation—the most urgent demands for relief had been satisfied by public works and relief programs, so that the permanent program to be designed by these advisors could more easily be kept separate from the “dole” so abhorred by FDR, but so politically necessary in the first months of his administration.101 What, then, can be said of the role of mass pressures in stimulating the development of new state social-spending measures in the United States? Given the character of the U.S. state structure, popular pressures were critical in providing an opening for elite initiatives. Indeed, popular political activity encouraged policymakers in the Roosevelt administration to take initiatives, but, as we shall see below, they were largely able to insulate the process of formulating policy from popular demands. There is no doubt that some kind of social-welfare initiative was politically necessary, but the specific form it took was shaped by the institutional arrangements of the American state, the power of certain congressional groups, and the reaction of New Deal policymakers to the U.S. policy inheritance.

Presidential Initiatives for Social Security


446

Roosevelt had long harbored political ambitions—for himself as well as for the Democratic party and the cause of progressive reform—and he eagerly worked to build and lead a cross-class coalition for social policy reform at the national level. There was substantial legislative activity on proposed pension and unemployment benefits legislation in Congress during 1933 and 1934, yet Roosevelt was interested in developing a comprehensive approach to social security, as well as in taking political credit for any legislation that did succeed. Roosevelt seized the initiative in formulating policy and was able to maintain control over the policymaking process by setting up the CES.102 This group of cabinet officers, established in early June of 1934, was chaired by Secretary of Labor and longtime social reform advocate Frances Perkins, who had come to Washington with Roosevelt from the position of Industrial Commissioner in his gubernatorial administration in New York.103 The CES would study social insurance and assistance, aided by an expert staff, and have a legislative proposal ready for the start of the 1935 congressional session. The key position of executive director was filled by Edwin Witte, a former Commons student trained at the University of Wisconsin and a veteran of years of administrative service in Wisconsin with the State Industrial Commission and the Legislative Reference Library (another state agency with progressive origins).104 Roosevelt’s choice of personnel for the important task of drafting the social security bill reflected his generally cautious approach to social policy. Roosevelt and the people he chose to be the architects of the Social Security Act came of age politically in the Progressive Era; as products of the progressive movement, they shared its concerns with “good government” and fiscal “responsibility.” Moreover, they had received their political training and developed their distinctive outlook in the “expert,” state-level administrative agencies dealing with labor and social legislation that were the enduring result of progressive state-building and social reform efforts. In particular, Roosevelt’s social policymakers came from backgrounds that ensured their concern with fiscal “soundness.” They would attempt to build programs able to withstand what they perceived as the ever-present danger of democratic pressure for expansion of government benefits into fiscally irresponsible and politically motivated “handouts.”105 Social policy experts who advocated more liberal policies were excluded from the inner circles of policy formulation, relegated to relatively powerless positions within advisory bodies, or left out of the official policy formation process altogether.106 The Deliberations of the Committee on Economic Security From the very outset of the deliberations of the CES, the tension between national standards and coordination and state-level autonomy and diversity was of overriding importance in debates over the character of the programs it would propose.107 Although politics had acquired a more national focus in the 1930s, there were substantial obstacles to national uniformity and centralized administration. Policymakers faced the threat of any national social insurance programs they recommended being declared unconstitutional by the Supreme Court. In addition, they had to cope with the political exigencies represented by the distribution of power within the Democratic party, which


447

favored the representatives of conservative and states’-rights viewpoints as much as or more than those of urban liberalism. In particular, the extraordinary political leverage of Southern representatives in Congress, out of proportion to the population and economic might of their states, was a fact that could never be ignored as FDR and the CES formulated the bill.108 Moreover, the fact that states had been the principal actors in social policy until the mid-1930s meant that diverse programmatic and administrative interests in the various state programs were politically well entrenched and difficult to bypass by the time the CES began its work. Indeed, Roosevelt, Witte, and Perkins, all of whom had been involved in state-level labor and welfare administration, actually preferred to maintain the states as important administrative actors, even as they hoped to increase the uniformity and adequacy of social programs. Other politicians representing the interests of progressive Northern states, such as Senator Robert Wagner of New York, the Roosevelt administration’s chief ally in the Senate, also took some stances which undercut nationalizing trends for programmatic uniformity in order to protect favored state legislation.109 One can see the working out of the tensions around national uniformity and state-level autonomy especially clearly in the struggles over the character of unemployment protection. Despite the many obstacles to national uniformity and administrative coordination, it is clear from these debates that important nationalizing tendencies coexisted with tendencies favoring decentralization of administration and lack of standardization. The policy options for unemployment insurance being debated by the CES and the U.S. community of social insurance experts included a purely national plan and a federal subsidy plan that would allow the federal government substantial control over standards, in addition to the favorite plan of Roosevelt, Perkins, and Witte—a federal-state tax-offset scheme, in which states would retain complete decision-making power over many aspects of their unemployment programs.110 Some sort of nationally standardized plan (either purely federal or a federal subsidy program that allowed substantial national control while addressing fears about constitutionality) was the overwhelming choice of policy experts, including those serving in technical positions within the CES.111 Further, the more national approaches were favored by trade union leaders and the few businessmen who had not already rejected the New Deal, and social security along with it.112 Despite this support for national standards, Roosevelt, Perkins, and Witte managed—after considerable resistance from their own technical staff, members of the official advisory council, and outside experts—to get the CES to recommend the tax-offset approach that preserved the most state-level autonomy.113 The tax-offset plan left to the states the politically difficult decisions about merit rating, employee contributions, and pooled versus individual reserves. This meant the CES could avoid taking stands on these tense issues, and Witte in particular would not have to undermine the Wisconsin experiment in “preventive” unemployment legislation.114 In the area of old-age protection, the tension between nationalizing and decentralizing tendencies played itself out slightly differently, and the debate was complicated by the politically sensitive issue of whether to include contributions from future beneficiaries. Policymakers in the CES had to contend with the growing strength of Townsendism and other movements calling for noncontributory federal pensions, and with the fact that members of Congress were far more supportive of pensions than of contributory


448

insurance for the aged. Roosevelt and Witte, reacting to a policy inheritance which dramatized the dangers of politically motivated expansion of benefits, were determined to keep the federal government out of the business of giving direct grants to citizens, and they never wavered from a commitment to contributory features for whatever permanent old-age program was settled on. The arguments of liberal New Dealers for a different approach found no favor. A vivid example of this intransigence came when Harry Hopkins, federal emergency relief administrator and member of the CES, eloquently argued for noncontributory old-age and unemployment benefits as a matter of right for citizens. Roosevelt saw this as being “the very thing he had been saying he was against for years—the dole,” and he vetoed any such proposals.115 Roosevelt, Perkins, and Witte preferred to develop what they saw as the only reasonable and “fiscally sound” long-term solution to the problem of old-age dependency: a contributory program of old-age benefits, to be firmly distinguished from noncontributory social assistance. Even so, Witte and the members of the CES were willing to include a governmental contribution to the old-age insurance fund once payroll taxes and the contributory principle were well established; the draft bill used by the CES showed a subsidy from federal revenues beginning in 1965.116Yet Roosevelt, upon discovering at the last moment that the CES plan included a future subsidy from general revenues, demanded a redrafting of the proposed legislation so that it would always be self-sustaining from employer and beneficiary contributions alone. Even a governmental contribution to the social security fund was problematic for FDR, with his “prejudice about the ‘dole.’”117 In the end, Roosevelt’s view prevailed, and the U.S. social insurance program for the aged received no financial input from federal coffers, a logical outcome of Roosevelt’s concern to preserve a sharp distinction between social assistance and social insurance programs. Given the preferences of the CES and FDR himself for state-level administration, it is worth explaining the fact that the contributory old-age insurance system was the only one of the several programs included in the Social Security Act that was administered completely at the federal level of government. To some extent this reflected the fact that Witte and Perkins were preoccupied with resolving the disputes about unemployment insurance, which by all accounts occasioned the most attention from the members of the CES and the policy experts who advised them.118 The staff members in charge of developing the old-age security portions of the program were strongly in favor of establishing a national old-age insurance system; they and the CES actuaries were unanimous in advising that, given extensive labor mobility, a federal-state scheme involving contributions over workers’ lifetimes would be unworkable from a technical point of view.119 Moreover, old-age insurance was the only area of policy in which the states had not enacted legislation and, hence, the only one in which there were no statelevel vested administrative and political interests to defend, as there were in the old-age assistance and unemployment programs. Despite their strong commitment to establishing contributory old-age insurance, CES members believed that some sort of old-age assistance was needed to cope with existing poverty among the aged, at least until the contributory schemes matured.120 The fact that four times as many elderly people were collecting benefits under federal relief programs as under state old-age assistance systems worried Roosevelt administration policymakers,


449

who did not want to encourage Americans to continue looking to the federal government to take care of the needs of aged citizens directly through noncontributory programs.121 State old-age assistance programs, with their inadequate financing and benefits, would have to be strengthened if the federal government was to be kept out of direct relief to the aged. The CES therefore recommended the enactment of the Old Age Assistance (OAA) program, a federal subsidy (of 50 percent) for state old-age assistance plans that met certain minimum requirements. Most importantly, the draft legislation required that plans be state-wide in operation, and that pensions be paid at a level that would ensure the “health and decency” of the elderly recipients.122 Though not seen by CES members as a long-term solution to poverty among the aged, OAA would please congressional Democrats and allow the popular enthusiasm for public old-age protection to be harnessed to the Social Security Act as a whole. The bill submitted to Congress by the CES also recommended that the federal government initiate a program of subsidies to state-level assistance schemes for fatherless families with dependent children—Aid to Dependent Children (ADC). These state programs would be modified versions of the mothers’ pension systems that had been enacted in most states in the Progressive Era and into the 1920s. As had been the case with OAA, the CES did not want to bypass state-level administration of existing programs and was opposed to establishing permanent federal assistance. Likewise, the CES saw a need to provide stronger financial backing of these programs, which were completely inadequate in the face of the economic crisis. The CES recommended the broadening of eligibility requirements, so that all children of single mothers could be aided; many mothers’ pension schemes had extremely restrictive provisions.123 Policymakers deliberately opted not to mandate assistance to all needy children regardless of family situation, a decision which has had far-reaching implications for the operations of and debates over public assistance to this day. In addition, the proportion of state expenses to be met by federal subsidy was substantially lower than that for OAA— one-third as opposed to one-half. Policymakers, noting the far greater interest of Congress in aid to the aged and the resistance of congressional conservatives to the bill as a whole, worried that more generous provisions would threaten passage of any program for poor children.124 The politically motivated decision to omit health insurance from the legislation the CES proposed obviously had important consequences for the future of American social provision. Roosevelt and his advisors on the CES looked on health insurance quite favorably, although they—along with most social insurance experts—saw unemployment and old-age provision as more pressing.125 Roosevelt had directed the CES to study health insurance, and to develop a legislative proposal. But even the decision merely to consider publicly run health insurance drew massive opposition from the medical profession. Although there was a brief moment in 1934 when Witte was “unsure that health insurance is out of the question”—because the American Medical Association (AMA) had been silenced temporarily by some concessions from the CES and by a desire to overcome internal dissension—FDR and the members of the CES ultimately decided that including health insurance in the proposed Economic Security Act was politically impossible.126 Witte wrote that he believed the inclusion of health insurance “would spell defeat for the entire bill.” Roosevelt was of the same opinion, and he would not authorize


450

a health insurance title for the administration bill, or even the publication of a CES report favoring health insurance.127 The hope of members of the CES and of FDR himself that health insurance would be introduced at a later date went unfulfilled as the conservative coalition in Congress gained ground in the years following the passage of the Social Security Act.128 The adoption of universal national health insurance during the New Deal was ultimately precluded by the failure of social insurance in the Progressive Era. After the health insurance campaign of 1915–20 ended in failure, the field was left open for the foes of health insurance, principally the medical profession, to consolidate their position.129 The AMA became virtually unopposable, given their capacity to pressure congressmen, especially in the absence of any significant and politically well-situated popular movement for health insurance (such as existed in the case of old-age protection). Thus, when the possibilities for social reform generally opened again in the 1930s, and despite public opinion favorable to new public health protections,130even so popular and powerful a politician as Franklin Roosevelt was unwilling to risk the passage of his other proposed programs by including health insurance in the package. The fact that patronage practices continued unabated in many states, and had not been completely eradicated even from the federal government, meant that the CES faced a challenge in constructing realms of administration that could manage the new programs while avoiding entanglements in patronage politics. That it was willing to face the challenge, rather than await the development of a public administration more suited to its preferences, is testament to the crisis atmosphere produced by the Great Depression and to its conviction that the continuing popular pressure for some kind of federal action on the social welfare front would produce “unwise” legislation if it did not act. Yet the CES carefully included as many guarantees against corruption, patronage, and mismanagement as possible in its proposed permanent social security legislation, and it tried to design programs capable of withstanding periodic democratic onslaughts demanding “unwarranted” expansion.

The Social Security Act: Charter Legislation of America’s Welfare State The legislative recommendations of the CES included a federal-state contributory unemployment system; federal subsidies to state programs for the needy aged, dependent children, and the blind; and a purely federal contributory old-age benefits program.131 The pragmatic formulators of the CES social security proposals had opted for substantial state-level autonomy in most of the programs they recommended. Yet the Report to the President of the Committee on Economic Security also recommended the establishment of minimum national standards, such as the “health and decency” requirement for old-age assistance benefit levels, and practices that would lead to administrative regularity, such as the recommendation for merit personnel systems in state-level administration of the programs. Moreover, the CES hoped to see all social insurance administration united within a single social insurance board located in the Department of Labor, with the social assistance programs grouped together under the aegis of the Federal Emergency Relief Administration. Finally, the CES recommended continuing research and planning for


451

national coordination of the new social welfare activities, endorsing the establishment of a national planning board.132 The CES proposal had been designed to propitiate important congressional interests, of course, especially those representing the South. These had largely determined that an approach favoring state-level autonomy would be taken. But even more concessions were wrested from the congressional sponsors of the administration bill by Southern lawmakers and other conservatives.133 Federal standards in general came under fire, and the CES proposal that state programs be administered along merit system lines had to be dropped. In addition, the administration of the insurance programs was taken away from the Department of Labor and grouped with the assistance programs in an independent administrative agency, the Social Security Board.134The CES recommendation that all employed persons be included in the unemployment and old-age insurance systems was not accepted by Congress, with the result that agricultural and domestic workers were excluded from coverage.135 This meant that the insurance programs of the Social Security Act would not reach most American blacks, who were still overwhelmingly concentrated in rural, agricultural regions of the South.136 Southern senators and congressmen led the fight for “states’ rights” and against national standards and universal coverage. They especially attacked the provision of OAA that mandated that states pay pensions sufficient to provide, “when added to the income of the aged recipient, a reasonable subsistence compatible with decency and health,” fearing it would represent an entering wedge for federal interference in their states’ system of segregation and discrimination. In the end, the Southern-led bloc won on that fight and the requirement was dropped.137 After the compromises, the Social Security Act was finally signed into law in August of 1935. States soon acted to pass legislation establishing programs under the act, and by 1937, the programs were nationwide in operation. Also in 1937, the act was declared constitutional by the Supreme Court, and the first payroll taxes were taken out of covered employees’ paychecks, though benefits under the old-age insurance program were not scheduled to begin until 1942. Yet before the first benefit check was ever mailed, important changes were made in the contributory old-age benefits part of social security. A new advisory council began discussions in 1937 on certain controversial aspects of the social security program.138 In addition to dissatisfactions with some of the more technical aspects of the law, there was widespread support for adding survivors’ benefits. As a result of the council’s recommendations and new negotiations in Congress, the 1939 amendments to the Social Security Act were passed. The most notable changes included a shift from full reserves to a modified pay-as-you-go system, and, in an early demonstration of the potential for expansion that even contributory social insurance systems can have in the United States, benefit payments became payable two years earlier (in 1940), payroll tax increases were delayed, and benefits for wage-earners’ dependents and survivors were added.139 Whatever its shortcomings as a system of social protection, it was already clear that social security represented an expandable set of social programs, despite the intentions of its formulators. While the 1939 amendments to the Social Security Act showed the channels along which the nascent U.S. welfare state might be expanded, other events in the late 1930s undermined the capacity of social security programs to ensure Americans’ economic well-being. The CES had believed that the success of their package of social programs in


452

guaranteeing social security for the American people depended upon “employment assurance”—the government’s provision of suitable jobs for all who wanted to work.140 As the Report of the CES pointed out: Any program for economic security that is devised must be more comprehensive than unemployment compensation, which of necessity can be given only for a limited period. In proposing unemployment compensation we recognize that it is but a complementary part of an adequate program for protection against the hazards of unemployment, in which stimulation of private employment and provision of public employment…are the other major elements.141 The CES expected that employment assurance would come through the institutionalization of the New Deal public employment programs; as the papers by Margaret Weir and Theda Skocpol in this volume detail, however, this did not happen, and social insurance and assistance programs—however inadequate to the task—were left to deal with problems of economic and social insecurity.142 1935 turned out to have been the highpoint of congressional willingness to go along with Roosevelt’s social reform initiatives, as a coalition of congressional conservatives gained strength in the late 1930s.143 During this period, FDR was thwarted in his attempts at administrative and judicial reform—which likely would have enhanced the capacity of the American state to undertake economic and social interventions—and new social reforms were turned back as well.144 It would be many years before Americans again enjoyed an era hospitable to social reform. The Social Security Act represented one of the most important of all the New Deal reforms of American social and political life. Along with such legislation as the National Labor Relations (Wagner) Act and the Fair Labor Standards Act, it established critical social rights for American citizens. Yet these rights were compromised by uneven national standards, the sharp distinction between assistance and insurance, and the omissions of health insurance, employment assurance, and allowances for all needy children. This largely reflected the inability of the New Deal social reform coalition, led by the Roosevelt administration, to overcome the deep resistance of congressional conservatives and some congressional constituencies to the changes they wanted to effect in American social policy, but the policymakers themselves were responsible for some of the choices that left the U.S. welfare system incomplete. Ultimately, it was the larger obstacles produced by the legacies of U.S. state-building, state structure, and past policy that prevented the achievement of a more complete welfare state in New Deal America. From Civil War Pensions to a Belated Welfare State: The Fate of Social Policy in America We have seen that throughout the history of American social policy, the possibilities for reform have been constrained by the legacies of state-building as well as by the characteristics of the American state structure. The fragmentation of authority encouraged by federalism and the division of powers has consistently undermined


453

coordinated national welfare initiatives, even in the periods most hospitable to reform. The patronage practices initially encouraged by early mass democracy and the lack of bureaucratic state-building in America deprived reformers of readily available institutional capacities for carrying out new social-spending activities. Successful reforms have had to work around the obstacles represented by America’s institutional and policy legacies, and they have built the bureaucratic capacity for social intervention slowly when it has been possible to build it at all. We have seen as well that American policy innovators and reformers have developed policy preferences in response to earlier state social interventions and policies, from the Progressives’ reactions against the excesses of the Civil War pension system and patronage democracy to the New Dealers’ preference for contributory social insurance. Moreover, the historically varying possibilities for cross-class coalitions—joining elite, middle-class, and working-class interests—have been critical to explanations of the fates of failed and successful social-spending programs. In the Progressive Era, it proved impossible to unite these interests around new social-spending programs, but during the New Deal, a cross-class coalition was able to form around the Social Security Act. Again today, the feedback from the programs established by the Social Security Act has shaped the possibilities for joining interests around alternative futures for social policy. If I have made some sense here of the origins of America’s belated welfare state, perhaps the insights derived from such an institutional-political process approach will be of use to both academics and those active in today’s policy debates. Policy options are always constrained by the legacies of existing policy, politics, and administration, and our choices today are burdened by our past:—not least by the legacies of the initial legislation of America’s incomplete welfare state, the Social Security Act—and we will do well to understand the limits to our alternatives. Yet the past development of American social policy shows as well that alternatives did exist. In the 1980s, we again find ourselves faced with alternative futures for American social provision. I conclude with the hope that these lessons about our past will serve to enhance the possibilities for more generous public social provision in America.

Notes I would like to thank John Myles, Hugh Heclo, George Steinmetz, and my co-editors, Theda Skocpol and Margaret Weir, for their helpful comments on this paper. Thanks also to Eric Parker for providing research assistance. Partial support for this research was provided by a grant from the University of Wisconsin Graduate School Research Committee. 1. 2. 3. 4.

Edward Devine, quoted in John Gillen, Poverty and Dependency (New York: Century, 1926), p. 284. Charles Henderson, Industrial Insurance in the United States (Chicago: University of Chicago Press, 1909), p. 277. Isaac M.Rubinow, Social Insurance (New York: Henry Holt, 1913), p. 404. Elizabeth Brandeis, “Labor Legislation,” in Don Lescohier and Elizabeth Brandeis, History of Labor in the United States, 1896–1932, vol. 3 (New York: Macmillan, 1935), pp. 399–


5. 6. 7. 8. 9.

10. 11.

12. 13.

14. 15.

16.

17. 18. 19. 20.

21. 22.

454

700; Mark Leff, “Consensus for Reform: The Mother’s-Pension Movement in the Progressive Era,” Social Service Review 47 (1973): 397–417. Stuart Brandes, American Welfare Capitalism, 1880–1940 (Chicago: University of Chicago Press, 1976). Theda Skocpol and John Ikenberry, “The Political Formation of the American Welfare State,” Comparative Social Research 6 (1983): 90. Hugh Heclo, Modern Social Politics in Britain and Sweden (New Haven: Yale University Press, 1974). Martin Shefter, “Party and Patronage,” Politics and Society 7 (1977): 404–51. Charles Tilly, “War Making and State Making as Organized Crime,” in Peter Evans, Dietrich Rueschemeyer, and Theda Skocpol, eds., Bringing the State Back In (New York: Cambridge University Press, 1985), pp. 169–91. Richard McCormick, “The Party Period and Public Policy: An Exploratory Hypothesis,” Journal of American History 66 (1979): 279–98. Morton Keller, Affairs of State: Public Life in Nineteenth Century America (Cambridge, Mass.: Harvard University Press, 1977), chaps. 7, 8, 14; Heywood Sanders, “Paying for the ‘Bloody Shirt’: The Politics of Civil War Pensions,” in Barry Rundquist, ed., Political Benefits (Lexington, Mass.: Lexington Books, 1980), pp. 137–59. Richard Bensel, Sectionalism and American Political Development (Madison: University of Wisconsin Press, 1984), pp. 60–73. Leonard D.White, The Republican Era: A Study in Administrative History, 1869–1901 (New York: The Free Press, 1958), p. 218. GAR membership increased from 60,000 in 1880 to 428,000 in 1890, when the organization included about one-third of all union Army veterans who survived the Civil War. Yet the initial legislative liberalization of pensions— the Arrears Act of 1879—was not a product of GAR lobbying; rather, the growth of the GAR was stimulated by the Arrears Act. On the GAR, see Mary Dealing, Veterans in Politics: The Story of the G.A.R. (Baton Rouge: Louisiana State University Press, 1952), and Donald McMurry, “The Political Significance of the Pension Question, 1885–1897,” Mississippi Valley Historical Review 9 (1922): 19–36. William Glasson, Federal Military Pensions in the United States (New York Oxford university Press, 1918); McMurry, “Political Significance of the Pension Question.” U.S. Bureau of Pensions, Laws of the United States Governing the Granting of Army and Navy Pensions (Washington, D.C.: Government Printing Office, 1925), p. 43, Carole Haber, Beyond Sixty-Five (New York: Cambridge University Press, 1983), p. 112. Ann Shola Orloff, “The Politics of Pensions: A Comparative Analysis of the Origins of Pensions and Old Age Insurance in Canada, Great Britain, and the United States, 1880s1930s” (Ph.D.diss., Princeton University, 1985), p. 71. McMurry, “Political Significance of the Pension Question,” p. 28. Keller, Affairs of State, p. 311. Sanders, “Paying for the ‘Bloody Shirt.’” William Glasson, “The South’s Care for Her Confederate Veterans,” American Monthly Review of Reviews 36 (1907): 40–47; Rubinow, Social Insurance, pp. 408–9; Ann Shola Orloff and Theda Skocpol, “‘Why Not Equal Protection?’: Explaining the Politics of Public Social Spending in Britain, 1900–1911, and the United States, 1880s-1920,” American Sociological Review 49 (1984): 728. Henderson, Industrial Insurance, p. 277. Leah Hannah Feder, Unemployment Relief in Periods of Depression (New York: Russell Sage, 1936), pp. 168–88.


455

23. Raymond Mohl, “Three Centuries of American Public Welfare,” Current History 65 (1973): 6–10, 38–39. 24. By the latter half of the century, some central administrative supervision of poor relief functions had emerged; in European countries, this was often at the level of the national government, while in the U.S., it was at the state level. On early U.S. poor relief administrative practices, see Josephine Brown, Public Relief, 1929–1939 (New York: Henry Holt 1940), chap 1, and Michael Katz, In the Shadow of the Poorhouse: A Social History of Welfare in America (New York: Basic Books, 1986), pt. 1. 25. Keller, Affairs of State, p. 377. 26. Ari Hoogenboom, Outlawing the Spoils: A History of the Civil Service Reform Movement (Urbana: University of Illinois Press, 1968), p. 1. 27. Frank Bruno, Trends in Social Work, 1874–1956 (New York: Columbia University Press, 1957), p. 79; Sophonisba Breckinridge, Public Welfare Administration in the United States: Select Documents (Chicago: University of Chicago Press, 1927), sec. 4; Harry C. Evans, The American Poorfarm and its Inmates (Des Moines, Iowa: Loyal Order of Moose, 1926); Keller, Affairs of State, p. 501. 28. Brown, Public Relief, p. 16. 29. Feder, Unemployment Relief, p. 188; see also pp. 22, 158. 30. John Pratt, “Boss Tweed’s Welfare Program,” New York Historical Quarterly 45 (1961): 396–411; William L.Riordan, Plunkett of Tammany Hall (New York: McClure, Phillips, 1905). 31. Philip Garret, “The Merit System in Public Institutions,” Proceedings of the National Conference of Charities and Correction (Boston: George Ellis, 1896), p. 369. 32. Eugene Smalley, “The United States Pension Office,” Century Magazine 28 (1884): 427. 33. Martin Schiesl, The Politics of Efficiency: Municipal Administration and Reform in America, 1880–1920 (Berkeley and Los Angeles: University of California Press, 1977), chap. 2; Martin Shefter, “Party, Bureaucracy and Political Change in the United States,” in Louis Maisel and Joseph Cooper, eds., Political Parties: Development and Decay (Beverly Hills: Sage, 1978), pp. 211–65. 34. Stephen Skowronek, Building a New American State: the Expansion of National Administrative Capacities, 1877–1920 (New York: Cambridge University Press, 1982), p. 68 and chap. 3 generally. 35. Shefter, “Party, Bureaucracy and Political Change,” p. 228. 36. Skowronek, Building a New American State, pp. 68–69. 37. Brandeis, “Labor Legislation,” pp. 575–77. 38. Orloff, The Politics of Pensions, chap. 6. 39. Irwin Yellowitz, Labor and the Progressive Movement in New York State, 1897–1916 (Ithaca: Cornell University Press, 1965), p. 10. 40. See, for one example among many, Gaston Rimlinger’s Welfare Policy and Industrialization in Europe, America and Russia (New York: John Wiley, 1971). 41. Among the better-known social-scientific surveys were in the United States, Jacob Riis’s How the Other Half Lives (New York: Charles Scribner’s sons, 1890), and, in Britain, Charles Booth’s Life and Labour of the People of London (London: Williams and Norgate, 1889 and 1891) and The Aged Poor in England and Wales (New York and London: Macmillan, 1894). On the “social question”—that is, how to incorporate industrial workers into the democratic polity, see, for example, John Graham Brooks, The Social Unrest (New York: Macmillan, 1903), and J.A.Hobson, The Crisis of Liberalism: New Issues of Democracy (London: P.S.King and Son, 1909).


456

42. On the British “New Liberals,” see Michael Freeden, The New Liberalism: An Ideology of Social Reform (Oxford: Clarendon Press, 1978); P.F.Clarke, “The Progressive Movement in England,” Transactions of the Royal Historical Society 24 (1974): 159–82; and Stefan Collini, Liberalism and Sociology (New York: Cambridge University Press, 1979). On their American counterparts, see Sidney Fine, Laissez-Faire and the General Welfare State (Ann Arbor: University of Michigan Press, 1965), and Robert Bremner, From the Depths (New York: New York University Press, 1956) chap. 8. 43. See, e.g., Hobson The Crisis of Liberalism, pp. 92–113. 44. Orloff, The Politics of Pensions, chap. 3.; Collini, Liberalism and Sociology, p. 107; Henderson, Industrial Insurance; Henry Seager, Social Insurance (New York: Macmillan, 1910), pp. 4–5, 148–50. 45. Kirk Porter and Donald B.Johnson, eds., National Party Platforms (Urbana: University of Illinois Press, 1970), p. 177; U.S. Bureau of the Census, Historical Statistics of the United States (Washington, D.C.: Government Printing Office, 1960), p. 682. 46. Massachusetts Special Commission on Social Insurance, Report (Massachusetts House No. 1850) (Boston: Wright and Potter, 1917), p. 57. 47. See, for example, Rimlinger, Welfare Policy and Industrialization, pp. 80–84. 48. Phillip Taft, Organized Labor in American History (New York: Harper and Row, 1964), p. 233; Gary Fink, Labor’s Search for Political Order: The Political Behavior of the Missouri Labor Movement, 1890–1940 (Columbia: University of Missouri Press, 1973), pp. 161–82. 49. “Unemployment Survey,” American Labor Legislation Review 5(1915): 591–92; Daniel Nelson, Unemployment Insurance: The American Experience, 1915–1935 (Madison: University of Wisconsin Press, 1969), p. 18. 50. Nelson, Unemployment Insurance, pp. 70–71. 51. Ronald Numbers, Almost Persuaded: American Physicians and Compulsory Health Insurance, 1912–1920 (Baltimore: Johns Hopkins University Press, 1978), p. 79; Paul Starr, The Social Transformation of American Medicine (New York: Basic Books, 1982), p. 250; Hace C.Tishler, Self-Reliance and Social Security, 1870–1917 (Port Washington, N.Y.: Kennikat Press, 1971), p. 173: “Labor Getting Behind Health Insurance,” The Survey 39 (1918):708–9. 52. On the British case, see C.L.Mowat, “Social Legislation in Britain and the United States in the Early Twentieth Century: A Problem in the History of Ideas,” in J.C.Beckett, ed., Historical Studies: Papers Read Before the Irish Conference of Historians (New York: Barnes and Noble, 1969), pp. 81–96; Bentley Gilbert, The Evolution of National Insurance in Great Britain (London: Michael Joseph, 1966), chaps. 5–7; and Heclo, Modern Social Politics, pp. 78–90. 53. Louis Reed, The Labor Philosophy of Samuel Gompers (New York: Columbia University Press, 1930), p. 117. 54. American Federation of Labor, History, Encyclopedia, Reference Book (Washington, D.C.: American Federation of Labor, 1919), pp. 303–4. 55. David Fischer, Growing old in America (New York: Oxford University Press, 1978), p. 171. 56. The text of the bill is reprinted in the Report of the Massachusetts Commission on Old Age Pensions, Annuities, and Insurance (Massachusetts House Document No. 1400) (Boston: Wright and Potter, 1910), pp. 339–40. 57. On events in Massachusetts, see Alton Linford, Old Age Assistance in Massachusetts (Chicago University of Chicago Press, 1949), chap. 1; Massachusetts Commission on Old Age Pensions, Annuities, and Insurance, Report; and Fischer, Growing Old in America, p.

The Political Origins of America’s Belated Welfare State 58. 59. 60. 61. 62.

63. 64. 65.

66.

67.

68.

69. 70. 71.

72.

73.

74.

75. 76.

77.

457

161. Massachusetts Commission on Old Age Pensions, Annuities, and Insurance, Report, p. 335. “Social Insurance,” American Labor Legislation Review 4 (1914): 578–79. “Unemployment Survey,” American Labor Legislation Review 5(1915): 573–75. Schiesl, The Politics of Efficiency. Robert Wiebe, The Search for Order. 1877–1920 (New York: Hill and Wang, 1967), chap. 5; John D.Buenker, Urban Liberalism and Progressive Reform (New York: W.W.Norton, 1978), chap. 6. Skowronek, Building a New American State, chap. 6. Buenker, Urban Liberalism. See Charles Francis Adams, “Pensions—Worse and More of Them,” World’s Work 23 (1911–12): 188–92, 327–33, 385–98. World’s Work carried many articles on pension abuse even as it championed various social, political, and labor reforms during the years between the turn of the century and the outbreak of World War I. Leff, “Consensus for Reform”; the entire third volume of The History of Labor in the United States by Lescohier and Brandeis chronicles the progress of workers’ compensation and legislation regulating work conditions, wages, hours, and related topics in this era. John Commons and John Andrews, Principles of Labor Legislation (New York: Harper and Brothers, 1927), p. 471; Lee Welling Squier, Old Age Dependency in the United States (New York: Macmillan, 1912), pp. 286–91. Roy Lubove, The Struggle for Social Security, 1900–1935 (Cambridge: Harvard University Press, 1968), pp. 149–57; on British unemployment insurance, see José Harris, Unemployment and Politics (London: Oxford University Press, 1972). Glasson, Federal Military Pensions, p. 283. Samuel McCune Lindsay, “Next Steps in Social Insurance in the United States,” American Labor Legislation Review 9 (1919): 111. Carroll Woody, The Growth of Federal Government, 1915–1932 (New York: McGraw Hill, 1934), pp. 371–72; I.W Litchfield, “United States Employment Service and Demobilization,” Annals of the American Academy of Political and Social Science 81 (1919): 19–27. David Brody, “The Rise and Decline of Welfare Capitalism,” in Brody, Workers in Industrial America (New York: Oxford University Press, 1980), p. 61; see also Brandes, American Welfare Capitalism. Joan Hoff Wilson, Herbert Hoover: Forgotten Progressive (Boston: Little, Brown, 1975), p. 89 and chap. 4 generally; Ellis Hawley, “Herbert Hoover, the Commerce Secretariat, and the Vision of an ‘Associative State,’ 1921–28,” Journal of American History 61 (1974): 116–40. William Chenery, “Unemployment at Washington,” Survey 37 (1921): 47; ibid., “The President’s Conference and Unemployment in the United States,” International Labour Review 5 (1922): 359–76; Carolyn Grin, “The Unemployment Conference of 1921: An Experiment in National Cooperative Planning,” Mid-America 55 (1973): 83–107; Wilson, Herbert Hoover, pp. 90–93. Clarke Chambers, Paul U.Kellogg and the Survey (Minneapolis: University of Minnesota Press, 1971), p. 77. Lafayette Harter, John R.Commons: His Assault on Laissez-Faire (Corvallis: Oregon State University Press, 1962); Nelson, Unemployment Insurance, chap. 6; Lubove, The Struggle for Social Security, chap. 7. John R.Commons, Labor and Administration (New York: August Kelley, [1913] 1964);


458

Commons and Andrews, Principles of Labor Legislation. 78. Lubove, The Struggle for Social Security, p. 33. 79. Nelson, Unemployment Insurance, chap. 6. 80. Louis Leotta, “Abraham Epstein and the Movement for Old Age Security,” Labor History 16 (1975): 359–78; Jackson Putnam, Old Age Politics in California (Stanford, Calif.: Stanford University Press, 1970), pp. 18–24; Fraternal Order of Eagles, The Fraternal Order of Eagles: What it Is, What it Does, What it Stands For (Kansas City, Mo.: Organization Department, Fraternal Order of Eagles, 1929); Fischer, Growing old in America, pp. 17374. 81. U.S. Bureau of the Census, Historical Statistics of the U.S., p. 738; Fischer, Growing old in America, pp. 173–74; Massachusetts Commission on Pensions, Report on Old Age Pensions (Boston: Wright and Potter, 1925), p. 37; U.S. Social Security Board, Social Security in America: The Factual Background of the Social Security Act as Summarized from the Staff Reports to the Committee on Economic Security (Washington, D.C.: Government Printing Office, 1937), p. 161. 82. Wilson, Herbert Hoover, chap. 5. 83. Porter and Johnson, eds., National Party Platforms, p. 331. 84. James Holt, “The New Deal and the American Anti-Statist Tradition,” in John Braeman, Robert Bremner, and David Brody, eds., The New Deal: The National Level (Columbus: Ohio State University Press, 1975), p. 29; Daniel Fusfeld, The Economic Thought of Franklin D.Roosevelt and the origins of the New Deal (New York: Columbia University Press, 1956), pp. 158–59. See also Frances Perkins, The Roosevelt I Knew (New York: Viking, [1946] 1964), pp. 166–67,182. 85. Perkins, The Roosevelt I Knew, p. 182. 86. Two classics in a huge literature on the New Deal and the popular political upsurge include William Leuchtenberg, Franklin D.Roosevelt and the New Deal (New York: Harper and Row, 1963), and Arthur Schlesinger, Jr., The Politics of Upheaval, vol. 3 of The Age of Roosevelt (Boston: Houghton Mifflin, 1960). 87. Abraham Holtzman, The Townsend Movement (New York: Bookman Associates 1963); Alan Brinkley, Voices of Protest: Huey Long, Father Coughlin and the Great Depression (New York: Alfred A.Knopf, 1982). 88. See, especially, the work of Frances Fox Piven and Richard Cloward, Regulating the Poor (New York: Random House, 1971), pt. 1, and ibid., Poor People’s Movements (New York: Random House, 1977), pp. 30–31. 89. George Martin, Madame Secretary: Frances Perkins (Boston: Houghton Mifflin, 1976), p. 341. 90. Perkins, The Roosevelt I Knew, p. vi. 91. Christopher Leman, “Patterns of Policy-Development: Social Security in Canada and the United States,” Public Policy 25 (1977): 264; ibid., The Collapse of Welfare Reform: Political Institutions, Policy, and the Poor in Canada and the United States (Cambridge: MIT Press, 1980), pp. 135, 165. 92. The fact that U.S. parties had not completely given up patronage practices, or moved toward thoroughly programmatic electoral appeals, reflected the uneven and incomplete success of earlier attempts to reform the civil administration. Patronage politicians remained strong in many areas of the United States and coexisted with politicians making more issue-oriented appeals. On the character of American political parties, see Theodore Lowi, “Party, Policy, and Constitution in America,” in W.Chambers and W.Burnham, eds., The American Party Systems (New York: Oxford University Press, 1967), pp. 238–76.


459

93. This effect was reinforced by the fact that Roosevelt and his closest advisors had been influenced quite negatively by European experiences with unemployment insurance in the 1920s; there—at least partially in response to popular political pressure—benefits had lost any connection to contributions and had become a “dole.” Thus, Roosevelt noted in a 1934 address, “Let us profit by the mistakes of foreign countries and keep out of unemployment insurance every element which is actuarially unsound.” See Franklin D. Roosevelt, “Addresses to the Advisory Council on Social Security” (November 14, 1934), in Samuel Rosenman, ed., The Public Papers and Addresses of Franklin D.Roosevelt, vol. 3 (New York: Random House, 1938), pp. 453–54. On the issue of the effects of popular pressures, see also Skocpol and Ikenberry, “The Political Formation of the American Welfare State,” pp. 123–26. 94. Irving Bernstein, A Caring Society: The New Deal Confronts the Great Depression (Boston: Houghton Mifflin, 1985), p. 66; Holtzman, The Townsend Movement, chap. 16; W.Andrew Achenbaum, Old Age in a New Land (Baltimore: Johns Hopkins University Press, 1978), pp. 129, 132. 95. While only six states had passed old-age pension legislation prior to 1929, twenty-eight states had initiated programs by the beginning of 1935, when the Roosevelt administration’s draft social security legislation was introduced in Congress. See U.S. Social Security Board, Social Security in America, pp. 161, 166. 96. Achenbaum, Old Age in a New Land, p. 129. 97. Theron Schlabach, Edwin E.Wine: Cautious Reformer (Madison: State Historical Society of Wisconsin, 1969), pp. 109–10. 98. Mark Leff, “Taxing the ‘Forgotten Man’: The Politics of Social Security Finance in the New Deal,” Journal of American History 70 (1983): 359–81. 99. Perkins, The Roosevelt I Knew, p. 294. 100. Bernstein, A Caring Society, chap. 1; Katz, In the Shadow of the Poorhouse, pp. 216–22; Piven and Cloward, Regulating the Poor, chap. 2. 101. Perkins, The Roosevelt I Knew, p. 284 102. Bernstein, A Caring Society, pp. 41–42, 51–53; Edwin Witte, The Development of the Social Security Act (Madison: University of Wisconsin Press, 1962), pp. 4–7, 18. 103. Martin, Madame Secretary, pp. 205–6. 104. Schlabach, Cautious Reformer, chaps. 2, 3. 105. Perkins, The Roosevelt I Knew, chap. 1, 2; Bernstein, A Caring Society, pp. 43–45; Schlabach, Cautious Reformer. 106. Lubove, The Struggle for Social Security, p. 176; Skocpol and Ikenberry, “The Political Formation of the American Welfare State,” pp. 127–28. 107. Schlabach, Cautious Reformer, pp. 114–15. 108. See Perkins, The Roosevelt I Knew, p. 291; Arthur Altmeyer, The Formative Years of Social Security (Madison: University of Wisconsin Press, 1968), pp. 14–15; and the paper by Jill Quadagno in M.Weir, A.Orloff and T.Skocpol, eds., The Politics of Social Policy in the United States (Princeton: Princeton University Press, 1988), chap. 6. 109. J.Joseph Huthmacher, Senator Robert F.Wagner and the Rise of Urban Liberalism (New York: Atheneum, 1968). 110. Altmeyer, Formative Years, pp. 17–25; Nelson, Unemployment Insurance, chap. 9; Witte, The Development of the Social Security Act. 111. Schlabach, Cautious Reformer, pp. 117–26; Bernstein, A Caring Society, p. 55. 112. Skocpol and Ikenberry, “The Political Formation of the American Welfare State,” pp 126– 31; Nelson, Unemployment Insurance, pp. 192–97, 202–3.


460

113. Schlabach, Cautious Reformer, p. 129. 114. Although Witte himself did not believe the preventive approach was satisfactory, neither did he wish to interfere with the operation of Wisconsin’s law. See Schlabach, Cautious Reformer, pp. 119–31. 115. Perkins, The Roosevelt I Knew, p. 284. 116. Schlabach, Cautious Reformer, pp. 127–28; Perkins, The Roosevelt I Knew, pp. 294–96. 117. Perkins, The Roosevelt I Knew, p. 296; Leff, “Taxing the ‘Forgotten Man.’ ” 118. See, for example, Altmeyer, Formative Years, chap. 1; Witte, The Development of the Social Security Act; Perkins, The Roosevelt I Knew, chap. 23; and Schlabach, Cautious Reformer, chap. 6. 119. Schlabach, Cautious Reformer, pp. 111–12; Altmeyer, Formative Years, pp. 25–26. 120. Achenbaum, Old Age in a New Land, p. 134; Committee on Economic Security, Report to the President (Washington, D.C.: Government Printing Office, 1935), pp. 4–5. 121. Committee on Economic Security, Report, p. 27; Schlabach, Cautious Reformer, p. 110. 122. Achenbaum, Old Age in a New Land, pp. 134–35. 123. It is perhaps worth noting, in light of later developments, that policymakers deliberately designed this program to allow poor mothers to stay at home to care for their children, just as better-off mothers were expected to do. See U.S. Social Security Board, Social Security in America, p. 233, and Irwin Garfinkel and Sara S.McLanahan, Since Mothers and Their Children: A New American Dilemma (Washington, D.C.: The Urban Institute, 1986), pp. 101–5. 124. Witte, The Development of the Social Security Act, pp. 163–65; Winifred Bell, Aid to Dependent Children (New York: Columbia University Press, 1965), pp. 27–28. 125. Witte, The Development of the Social Security Act, p. 187; Schlabach, Cautious Reformer, pp. 108, 112; Perkins, The Roosevelt I Knew, chap. 13. 126. Schlabach, Cautious Reformer, pp. 112–14; Stair, The Social Transformation of American Medicine, pp. 266–69. 127. Witte, The Development of the Social Security Act, p. 188. 128. James T.Patterson, Congressional Conservatism and the New Deal (Lexington: University of Kentucky Press, 1967). 129. Lubove, The Struggle for Social Security, p. 89. 130. U.S. Department of Health, Education, and Welfare, Social Security Administration, Office of Research and Statistics, Public Attitudes Toward Social Security, 1935–1965, by Michael E.Schlitz, Research Report No. 33 (Washington, D.C.: Government Printing Office, 1970), pp. 128–29. 131. To deal with concerns about the constitutionality of the fully national old-age insurance program, separate titles were used to establish taxes and benefits; the CES was depending on the taxing power of the federal government to ensure the plan’s constitutionality. See Achenbaum, Old Age in a New Land, pp. 136–37. 132. Committee on Economic Security, Report, p. 9. 133. On getting the Social Security Act through Congress, and the changes this necessitated, see Witte, The Development of the Social Security Act; Perkins, The Roosevelt I Knew, pp. 296–301; Achenbaum, Old Age in a New Land, pp. 134–35; Altmeyer, Formative Years, p. 35; and the papers by Kenneth Finegold and Jill Quadagno in M.Weir, et al., eds., The Politics of Social Policy in the United States, chaps. 5 and 6, respectively. 134. Witte, The Development of the Social Security Act, pp. 101, 144–45. 135. Witte, The Development of the Social Security Act, pp. 131,152–53. On Roosevelt’s preference for universal insurance coverage, see Perkins, The Roosevelt I Knew, pp. 282–


461

83. 136. Nancy J.Weiss, Farewell to the Party of Lincoln: Black Politics in the Age of FDR (Princeton, N.J.: Princeton University Press, 1983), 166–68. 137. Witte, The Development of the Social Security Act, pp. 144–45. 138. Schlabach, Cautious Reformer, chap. 8. 139. Achenbaum, Old Age in a New Land, pp. 136–37; Altmeyer, Formative Years, chap. 3. 140. Committee on Economic Security, Report, pp. 3–4, 7–10. 141. Committee on Economic Security, Report, pp. 7–8. 142. See also Katz, In the Shadow of the Poorhouse, pp. 227–33, on the problems of institutionalizing work relief in the New Deal United States. 143. Patterson, Congressional Conservatism and the New Deal. 144. Richard Polenberg, “The Decline of the New Deal, 1937–1940,” in Braeman, Bremner and Brody, eds., The New Deal, pp. 246–66.


462


SECTION FIVE: American Exceptionalism


Commentary The United States was born from a revolt against taxes, and the constitution with which it is endowed deliberately seeks to create a situation in which the emergence of despotism would be impossible. Remarkably, the United States still has the constitution with which it was born. The papers in this section add to comments made by others in this collection in order to examine the peculiarities of the United States, the greatest power in the world but one which resolutely sets its face against the very concept of the state. R.Salisbury’s ‘Why No Corporatism in America?’ begins with empirical analysis that does very firmly give him a subject matter. Interest group organisation, of capital as much as of labour, is relatively decentralised and so effectively underdeveloped. Salisbury examines a series of reasons for this. The fact the United States is a continental country makes it extremely hard for a unitary set of interests to emerge: in the Reagan years, for example, economic decline of the old industrial heartland went hand-in-hand with extraordinary growth in the south and in California. Further, the political culture of the United States is resolutely hostile to monopolies, and there have continuously been striking attempts to destroy overmighty corporate subjects. The end result of all this is that it is impossible to run a corporatist style of government since the corporate interest blocks are either not present or so weak as to be incapable of policing their own members. The United States thus has neither corporatism, nor central planning, nor much indicative planning: this is a source of weakness at times, but of strength at others—but the institutional pattern is deeply set and very unlikely to be changed. Curiously enough, the very weakness of domestic interest groups, in tandem with the fact that their struggles all take place within this huge society, allows foreign policy makers a very large degree of freedom. I.Katznelson and K. Prewitt’s ‘Constitutionalism, Class and the Limits of Choice in US Foreign Policy’ demonstrates that a relatively weakly organised society exerts no real control over foreign policy—unless it beings to impinge upon them in a major way, most obviously in terms of battle deaths. The President of the United States has a free hand for much of the time, provided that any wars in which his country is involved are small and costless in terms of American lives. This may or may not be an advantage, but it too is ‘the American way’—which outsiders must understand in order to avoid confusion just as much as social scientists must in order to understand the historical record.



467

57 Why No Corporatism in America?* Robert H.Salisbury

* Source: P.Schmitter and P.Lehmbruch, eds., Contemporary Political Sociology, vol. 1, Trends Toward Corporatist Intermediation, 1979, ch. 8, pp. 213–230.

In the early 1970s, after President Nixon had created tripartite machinery to bring labor, business, and government together to try to halt inflation, several scholars proclaimed that the corporate state had arrived in America [Fusfeld, 1972:1–20; Peterson, 1974:483– 506]. For the most part it was the strength of corporate business in the circles of decision that most impressed these observers, but the new institutional arrangements were taken to be confirming evidence of corporatist tendencies stretching back through World War II, the National Recovery Administration, all the way to Theodore Roosevelt. But the evidence quickly disintegrated. Official labor-management collaboration in an incomes policy was ended and the machinery abandoned. Part of the reason could be found in the curious weakness of those peak associations that would be expected, in a corporatist system, to play crucial roles. Despite the temptations to seek unity, major sectoral groups remained divided. In January 1977, a small newspaper item announced that the National Association of Manufacturers and the US Chamber of Commerce had suspended talks about the possible merger of their two organizations. Later in the spring, accompanying the accession of Douglas Fraser to the presidency of the United Auto Workers were speculations about whether the union would now reenter the AFLCIO. These two stories were among the recent examples of a phenomenon that has long been known but not much examined; peak interest group associations in the United States have great difficulty achieving enough comprehensiveness of membership to be able effectively to represent their respective sectors in the political process. The analysis of this condition provides the basis for this paper. To begin, let us say what we mean by the notion of peak association. Comparativists may not think there is much difficulty here. In Britain such groups as the National Farmers Union, the Confederation of British Industries or the Trades Union Congress include some 80–90 percent of their respective potential members. In West Germany the Spitzenverbände must be consulted in drafting legislative proposals. And elsewhere in Western Europe the phenomenon is not only familiar, it is a critical organizational ingredient in the emergence of what Schmitter calls societal corporatism. But in the United States it is not quite so clear which organizations are the “peaks”, and so we must seek definition for our term. It turns out that the interest group literature does not provide much help. The term “peak association” is used, often in passing, by such scholars as Key [1964] and

Why No Corporatism in America?

468

Eldersveld [1958]. It is employed as an important part of his analysis by Wooten [1970]. But none of these writers defines the words very clearly, and Eldersveld seems to imply that he thinks the notion is unclear. My own previous comments concerning “peak associations” are brief but nearly all of what there is. I suggested that the term referred to “sector-wide organizations which embrace a comprehensive array of constituent sector organizations” [Salisbury, 1975:187]. This is a reasonable beginning but uncertainties remain. First, it is not clear what is meant by saying that a peak association is composed of other organizations. Indeed, the entire notion of group membership itself is far more ambiguous than is commonly realized. For instance, in the National League of Cities the unit of membership can be either the individual city or the state leagues of municipalities. The American Trucking Associations include both individual firms and specialized associations. And the National Association of Manufacturers has several different kinds of membership. Moreover, there is a considerable difference between a group like the American Farm Bureau Federation, a federation of state farm bureaus, and the Consumer Federation of America, an amalgam of over two hundred very diverse kinds of organizations. Yet both are “organizations of organizations”. A second uncertainty arises over the definition of sector. The term is most commonly employed to refer to the major sections of economic self-interest in modern industrial society; labor, business and agriculture. What is unclear is how many such sectional or producer groups should be designated as sectors without destroying the meaning of the term. What of the professions? A good case can be made for including medicine, law and education as sectors. Each is reasonably well-bounded. Each has an “interest”, a stake in society and in public policy. In recent years the “PIGs”, the Public Interest Groups which include public official organizations of mayors, governors, counties, and the like, have certainly emerged as a self-interested set with clear stakes in public policy [Stanfield, 1976]. The Nader groups and Common Cause are probably outside the definition. So perhaps are the environmentalists. Consumers are a borderline case. We need not come to a definitive position on the matter; only indicate the fuzziness of the boundary. Let us attempt a definition. A peak association is an organization which purports, and is taken, to speak for a particular sector of society. The term leaves out those groups who defend the public interest for they deny a “selfish” sectoral concern. The term sector is intended to apply to larger rather than smaller slices of society. Neither the petroleum industry nor Texas constitutes a sector in this sense. The definition involves a reciprocal relationship. A group cannot simply declare itself to be the spokesman of a sector. It must be acknowledged to be so by those to whom it speaks: decision makers, elites more generally, or the broad public. Once the notion of audience response is taken into account, it makes the concept probabilistic rather than definitional. An organization may be acknowledged as the legitimate representative of a sector by some elites but not others. It may be supposed that there are threshold points beyond which a group achieves the status of peak association, and these points are located at the intersection of the organi-zation’s actual hegemony in its sector—density of membership, absence of intrasector rivals, forceful assumption of sector leadership—and the recognition of that hegemony by the relevant “others”. Presumably, the greater the hegemony, the fuller the recognition. But the opposite is also true. By conferring recognition on a group as the rightful spokesman of a sector, policy makers may greatly enhance its actual dominance.


469

Indeed, it appears that many of the peak associations in Western Europe reached their hegemonic status with major contributions from the more or less official recognition by key government agencies, especially in the bureau. In the United States, however, it has been unusual for official recognition to be granted any organized interest group, and it has been even more rare for such recognition to be given to groups purporting to speak for an entire sector. There are some examples, however, and let us note them. One of the best known is the support given through the Extension Service to the formation and continuing strength of the American Farm Bureau Federation [Block, I960]. It took decades for this connection to be severed, and during at least part of the time from the Bureau’s establishment in 1919 until the late 1940s there were many, inside of government and out, who asserted or acknowledged the Bureau’s suzerainty [McConnell, 1953]. Yet all the while the Grange and the Farmer’s Union existed as general farm organizations open to all kinds of farmers, and when there was a major consultation with USDA or Congress about farm policy they too were included [Campbell, 1962]. From the time of the Brannan Plan in 1949 until 1977 the dominant motif of American farm politics was partisan division with each party joined together in close working relationship with a general farm organization, Democrats with the Farmers Union and Republicans with the AFBF [Key, 1964:159; Heinz, 1962:952–78]. Even though the latter was much the largest, it could hardly be regarded as a peak organization when its access depended so heavily on having Republicans in power. And periodic efforts to transcend the partisan division by establishing ad hoc conditions of farm organizations have all quickly failed as the coalitions find themselves unable to contain the centrifugal forces generated by diverse and conflicting farm interests. Another substantive area has witnessed several efforts of federal government officials to encourage the formation of interest groups [Zeigler, 1964:94–109; Fainsod et al., 1959:46–71]. First during World War I, again under Secretary of Commerce Hoover in the 20s, during the NRA period of 1933–35, and in World War II trade associations were organized with the active support of government, primarily in order to assist in the administration of federal regulatory programs. At other times federal policy has sought to restrict trade association activities, too, of course, but even during the periods of encouragement there was no sustained move to support or consult with groups that purported to speak for all of industry. Organizational recognition was confined to the level of the specific industry. When Washington officials wanted to acknowledge the importance of business (or labor) as a whole, to confer symbolic recognition, and to consult with private interests about public policy, they worked with individuals, like William Knudsen and Sidney Hillman, whom they, the officials, not the sector organizations, chose [Blum, 1976]. A case for full peak status might be made for the AFL-CIO. Certainly since the 1955 merger there has been no rival organization in a position to speak on behalf of all of organized labor. Nevertheless, when compared, say, to the TUC, several points of difference appear. For one, the AFL-CIO includes only about seventy-five percent of unionized workers who consist, in turn, of less than one-fourth of the total work force. Moreover, any labor group that does not include the auto workers, the teamsters, or the mine workers must stand in stark contrast to the TUC in which the three equivalent unions are among the most significant. AFLCIO hegemony is further reduced by its


470

declining rate of success in winning representation elections. This is not meant to dismiss them as unimportant. Mr George Meany has certainly had a significant voice in national policy discussions, and Andrew Biemiller leads what is widely regarded as among the most skilled lobbying crews in Washington [Singer, 1976]. But neither lobbying strength nor electioneering clout has been sufficient to assure the AFL-CIO of a decisive, officially acknowledged, voice on labor questions. It is the President who selects the Secretary of Labor and other labor representatives on official bodies, not the AFL-CIO, even when a Democrat is in the White House. A final example of the difficulties of sustaining peak association status in the United States can be drawn from medicine. When Oliver Garceau’s splendid study was published in 1941 there was little doubt of the AMA’s dominance over the “medical profession in politics” [Garceau, 1941]. This remained largely true through the 1950s. But it is not true any longer. Rival organizations of doctors have been formed, and other groups with different interests, such as the hospital administrators, have emerged to contest with the AMA for position and power in health policy making. Thus peak associations in the United States are either weak and incomplete or ineffectual. None can claim quasi-monopolistic hegemony over a significant sector of socio-economic self-interest. Consequently, the United States lacks an essential ingredient of a corporatist polity, “a limited number of singular, compulsory, noncompetitive, hierarchically ordered and functionally differentiated categories….” [Schmitter, 1974:93–4]. Now there is a long tradition among American intellectuals that asks, “Why no Socialism in America?” [Laslett and Lipset (eds.), 1974], To that query, or perhaps instead of it, we now would ask, “Why no corporatism in America?” I propose to approach this question from three quite different perspectives: one, macro-social; two, in terms of the patterns of public policy; and three, as a problem in organizational analysis.

Social Diversity and Institutional Fragmentation Perhaps the most immediate and common response to the question of “Why no corporatism?” would be the one that harks back to the Federalist Papers and identifies two interrelated factors, social diversity and institutional fragmentation as primarily responsible. The relative extent of social diversity and its effects in American life can be argued. Beer and Sapolsky, for instance, both point out that British agriculture is also diverse in commodity interest, yet successfully brought under a single organizational tent [Beer, 1958:130–140; Sapolsky, 1968:355–76]. How much diversity is too much to contain? To the element of heterogeneity one may add that of sheer size. Compared to the nations of Western Europe the physical scope and diversity of the United States is immense. This is undoubtedly true, yet corporate concentrations of power have emerged, despite the anti-trust laws, in sufficient proportion to suggest that American heterogeneity can be overcome by skillful entrepreneurs. A variant on the diversity factor, and one of considerable interest, is presented by those who contend that the pace and timing of socio-economic growth significantly affects the prospects for corporatist organization. At one end of the spectrum we find Sweden with


471

its rapid and nationwide advance into mature industrialism. The United States is surely close to the other end [Sharkansky, 1975]. Industrialization struck New England a century and a half before it reached many parts of the South or West; today, while Utah, Wyoming and Montana emerge as fuel-rich industrial sites, the Northeast attempts to shore up its sagging employment prospects. The unevenness of growth is not confined to a single period of history. The late-starting sections do not eventually catch up and smooth out the differences. The Old South may become part of the Sun Belt, but Phoenix remains different from Savannah in the value and interests of those who live there. Within any given sector, therefore, the tensions among competing groups will be greater and more difficult to reconcile to the extent that historical socioeconomic growth and development have been distributed geographically in an uneven pattern. If the historical pattern has been uneven, it will continue to be. Moreover, this tendency will be greatly accentuated if political institutions are designed to accommodate this spatially distributed diversity. And this is what American institutions are preeminently designed to do. The structural elements are familiar: federalism, separation of powers, legislators nominated and elected from single-member districts. The elements interact to perpetuate the pattern where groups have multiple access points and governmental officials find it extremely difficult to assemble enough authority to act on a comprehensive scale, whether it involves enacting policy or negotiating with a socio-economic sector. At the same time, the continued existence of dispersed centers of authority provides opportunities for influence to interest groups which are similarly organized in a fragmented, geographically dispersed manner [Truman, 1951; McConnell, 1966]. Simply to illustrate the familiar point, if lawyers are able to affect the selection of judges and the development of procedural rules at the state level for state courts, the need to act on a national basis is much reduced and the ability of the ABA to mobilize the legal profession is likewise reduced. If federalism makes it difficult, or unnecessary, for a sector to be mobilized on a nationwide basis, so all the devices for fragmenting governmental authority make it difficult, or impossible, to assemble the capacity to act in a focused and forceful fashion. The other side of the corporatism equation is “the state”, an entity capable of recognizing, licensing or creating a peak association and then bargaining with it in a substantively meaningful way. But in the United States no such monistic state exists in any real behavioral sense. The Department of Agriculture cannot meaningfully engage in an annual price review, even if there were an agreed spokesman for agriculture [Self and Storing, 1962], because neither that bureaucracy nor the administration of which it is a part can commit “the state” to a specific course of action except within a very narrow range. Congress has now allowed it even in its most extreme moments of delegation of authority. HEW cannot exercise even its delegated authority over universities or the medical profession without multiple end runs immediately going to Congress to mitigate the effects of the order. Corporatism is a form of collective bargaining. In the United States there is no party, on any side, with enough authority to bargain effectively and commit a sector-wide following to accept the result. It is instructive to note that the partial exceptions to this statement have mainly occurred during emergencies. The extensive (and, as the Supreme Court ruled in the Schechter case, excessive) corporatist delegation contained in the NIRA was a response


472

to the Depression. Trade association formation and cooperative bargaining flourished also during the wars, especially in conjunction with administering wage and price control. But even national emergencies have not suppressed for long the centrifugal tendencies built into America’s governing institutions. And yet, governmental arrangements are, after all, subject to change. Public policy might have overcome the tendency toward fission. The mere existence of fragmented institutional structures does not guarantee their persistence, though it helps. Corporatist institutional possibilities might have been established by law, as they were in a good many countries after World War II. In the United States the functional need for a “stable, bourgeois-dominant regime” was surely as compelling as in other industrialized democracies [Schmitter, 1974]. But policy has not moved in that direction. Why not?

Interest Group Legitimacy The factors adduced already surely help to account for the predominant thrust of public policy. The central point to be added here is this. An important reason the American public policy has not enhanced corporatist tendencies is that monopolistic interest groups are regarded with deep suspicion in the American political culture.1 It may seem absurd, at first blush, to suggest that in the United States, where interest group analysis was, in a sense, invented and given its most enthusiastic hearing, where associational membership has been the very hallmark of national character, the legitimacy of interest group activity is less than it is in other democratic systems. Nevertheless, it is the case. First, we offer as evidence the point made earlier; namely, there is virtually no official incorporation of formal associations as participants in policy discussions. They are not invited to designate sector representatives on governmental advisory bodies, or to name key policy makers who are to deal with their sector. There have, of course, been occasional advisory relationships, such as the ABA’s role in “clearing” judicial appointments [Grossman, 1965]. And we are in no way suggesting that groups have less informal influence in the United States than elsewhere. We shall comment on this aspect later. Here our point is only that their formal position is negligible. A corollary to this is that bureaucrats in America deal directly with constituent units, not with associations. HEW negotiates with individual hospitals or universities, not with the organizations of hospitals or universities. The Defense Department does not contract with the Aerospace Industries Association but with individual firms. It is specific unions which must conform to pension fund regulations, and the AFL-CIO does not formally intervene. A second piece of evidence bearing on the legitimacy issue is the amount of official regulation of group activity. We do not have a comprehensive survey on the matter, but it seems clear that no democratic nation even remotely approximates the seriousness of the American effort to regulate the details of group-government interactions. For a long time scholars seemed to believe that this was because there was so little group activity in other systems, but once interest groups were discovered abroad some two decades ago that illusion was swept away. Lobbying regulation in America plainly rests on a deep suspicion of organized groups


473

and of the probable consequences for the polity of their activities. This same suspicion permeates significant fragments of the academic community. In political science the heirs to the progressive tradition characteristically regarded interest groups as the enemy of the public interest.2 Schattschneider, Lowi, and a rather disparate array of latter-day antipluralists have kept this perspective very much alive [Schattschneider, 1960; Lowi, 1969]. The point is that an anti-group orientation is widely diffused through the American polity, and it permeates both textbook literature and newspaper editorials. The organizational success of Common Cause, itself an interest group to be sure, cannot be understood apart from this profound conviction that organized interest groups are not fully legitimate participants in the processes of government. Why, then, is this so? The matter can be approached from at least two directions. One is that fount of so much interpretation of American politics; John Locke, Louis Hartz and the Liberal Tradition [Hartz, 1955]. The argument is that American political culture is so rooted in individualist assumptions that groups have no integral place. Our theorists and publicists have sustained a mythic structure that was created in order to undermine an older corporatism of post-feudal Europe. [I] is clear enough why the democratic image these men gave us should be hostile to half of the machinery that was later invented to make democracy work…Seeking to emancipate men from the rigid pluralism of church, guild, and province, those thinkers were bound to be “individualistis.” How could they say, even if they understood the fact, that democracy itself would function through a new pluralism of association, parties, and groups [Hartz, 1960:13– 14]. Presthus is one of the very few who has recognized how hostile American political culture is toward organized groups as policy-making participants [1974]. He regards Canada as possessing substantially more of a corporatist value orientation, though the empirical data he develops are not entirely compatible with this interpretation.3 But we need not rest the argument entirely on the elusive variable of political culture.4 Tentatively, I would suggest that group legitimacy and group hegemony are mutually interdependent. What makes an organization the legitimate spokesman for a socioeconomic sector? Confidence that the organization is representative of the values, opinions and interests (which terms may all mean the same thing) of the members of that sector. How can we be confident of this? From evidence that the organization encompasses a substantial portion of the sector in either actual or virtual membership; from evidence that the organization’s leaders are not embroiled in internal conflicts over policy [Truman, 1951:84; Masters et al., 1964]; from the existence of procedures, such as elections or referenda, that give some reason to suppose the leadership to be reasonably representative of the membership. There are examples of this kind of unity/legitimacy which, in turn, results in a very impressive accumulation of group influence, albeit informal, over policy [Masters et al., 1964; Bailey et al., 1963]. But in the US these cases are rare and generally rather fragile. The overwhelming tendency is for membership to be fractional, cohesion to be threatened, and representativeness to be very doubtful. Hence, no legitimacy. Hence, no corporatist role.


474

The importance of the legitimacy dimension is that it makes it highly unlikely that American public policy will move in a corporatist direction, granting privileged access to particular organizations so that they gain enough organizing advantages to achieve quasimonopolistic status vis-à-vis their respective sectors.5 This being so, the only way for a corporatist system to develop would be for would-be peak associations to achieve hegemonic status through their own organizational processes.

Peak Association and Organizational Exchange If it is up to organized groups to make themselves into legitimate peak associations, what would they need to do? To consider this question we employ the analytical framework that Mancur Olson and others have used to think about interest groups generally [Olson, 1965; Salisbury, 1969]. The would-be peak must offer prospective members selective benefits, unavailable outside the peak organizations, which are sufficiently appealing to induce a large proportion of the sector involved to join. (We pass over Olson’s other possibilities, coercion or one member bearing the whole cost.) What kinds of benefits could hold such appeal? Since we are dealing with peak associations we can assume that, despite the ambiguities we noted earlier regarding the nature of membership units, the prospective members of the peak are themselves organizations, such as labor unions, business firms or universities. It would be rare that material benefits in their usual forms—cheap insurance, charter flights, or strike funds—could be utilized to build a peak organization. The main reason for this is that the constituent units themselves are organized around the exchange of material benefits. Moreover, each of the constituent units typically has a substantial staff. This staff is unlikely cheerfully to surrender this autonomy or the membership to subordinate a prosper-ing benefit exchange to a peak association, and, conversely the peak group cannot prevail merely by offering staff support or selective material benefits. American constituent groups tend to be far stronger in staff and material resources than their counterparts elsewhere, and hence less susceptible to this kind of appeal from peak organizations. Let us consider four types of benefits that are sometimes utilized, or at least proposed, as having the potential to attract and hold the requisite membership. These are information, the regulation of jurisdictional disputes, recognition of the sector organization as legitimate and/or expert, and increased sector weight in policy making. All are of material concern, but their value is not as direct as such things as insurance, and indeed their worth to prospective members is highly problematic. Sapolsky suggests that a major benefit accruing to peak association members in Britain is access to information about ministry intentions regarding sector policy [1968:368]. There is no doubt that information is a very valuable commodity. But in the United States two factors conspire to make it extremely difficult for peak associations to monopolize such information so as to make it a selective benefit available only to its members. One is that given the institutional fragmentation of American government there are multiple sources of information about what is happening in Washington. It is very difficult to imagine who could sign an exclusive dealing agreement on behalf of the government,


475

effectively limiting the flow of information to a single peak group, and make it stick. The second reason is that there is a highly developed and enormously competitive information business already developed, and any organization already has many options available to find out what they need to know.6 Differential access to information gives advantages to specific organizations vis-à-vis competitors in their own sector. At some point, of course, such access may become too costly to maintain and a common information pool, as would be provided by a peak association, would then be attractive. Essentially, this is the calculus followed by a newspaper that uses a wire service rather than maintaining its own Washington correspondent. But it has not yet become operative in labor, business or the other major interest sections. Aspiring peak associations in most sectors seem not to have been as efficient in providing information to members as more specialized groups, and the prospects for peak group hegemony over this function are not at all encouraging. The regulation of jurisdictional disputes is a different kind of matter. The possibilities can best be seen in the case of labor unions. One of the chief advantages of union participation in the AFL-CIO is protection against raiding by rival unions and support from rival unions in organizing activities. A good recent example can be seen in the AFLCIO support for the United Farm Workers against the Teamsters in their efforts to organize in California. Surely this kind of help can be of significant benefit to all but the very strongest organizational units in a sector. The Teamsters, of course, are very strongly situated and hence quite content to remain outside the AFL-CIO fold. For other unions and in other sectors, however, there is an additional factor that limits the ability of organizations in the United States to establish sector hegemony by regulating jurisidictional disputes. There is an extensive array of extant public policy that already deals precisely with this problem in both positive and negative ways. On the negative side the antitrust laws preclude business organizations from allocating markets among competitors, and there is a considerable number of cases in which trade associations have been convicted on exactly those grounds. The NRA period was an exception, of course, and enforcement of competition has always been uneven, to say the least. But in large sectors of the economy peak associations would be legally forbidden to provide the “benefit” of settling jurisdictional arguments for the members.7 Consider also such policies as commodity marketing orders which do indeed have the effect of regulating conflicts among potential competitors and which have been written into positive law as a result of the specialized pressures of particular commodity groups. In this situation there is no regulatory function left for a peak organization to perform. When an organization is put together, as the National Conference of Commodity Organization was in the late 1950s, it is left with the much weaker function of coordinating a log-rolling alliance based on the mutual support of each commodity group for every other group’s needs. Hence the peak group does not long survive. There is a broad array of legislation regulating group operations, including representation elections and pension fund disclosures in the labor field, securities issues in business, and degree program certification in higher education, to maintain only a few which might have been subject to private control of strong peak associations. Conversely, it seems generally to be the case that in the more fully corporatist systems of Western Europe, with vigorous associations of comprehensive sector-wide scope, there is considerably less legislation


476

regulating the internal affairs of that sector. We have already discussed the question of peak association legitimacy at some length. In the present context the point is that legitimacy requires substantial sector unity and this can perhaps best be achieved by means of an effective peak organization. Let us consider further, however, what else is associated with sector unity, what kinds of groups especially need it, and what trade-offs there are against it. Professional groups provide good illustrations of the arguments. When doctors or lawyers or educators are cohesive and speak with one organizational voice, their associations are often regarded as legitimate spokesmen for their sectors. More broadly, the whole profession is conceded a degree of expertise that entitles its members to deference on sector-related public policy, and to substantial self-regulation. One aspect of this is the right to charge a fee for services which, if they are incompetent, will not be discovered until it is too late to do much about it, except, of course, to sue for malpractice. Such a reputation for expertise is thus highly valued and placed at risk whenever significant policy conflicts develop within the sector organization. Once there is substantial internal disagreement about what is best for the sector, there is no choice but to have non-experts make the decision, in coalition with some portion of the sector and against some other part, and to do so according to standard political considerations. From the perspective of the sector, the thing to do is first to suppress the conflict or mask its expression. If the decisive forces cannot be contained, however, the indicated strategy is to cultivate political clout. Mobilized electoral strength can affect political decisions when deference is no longer afforded to expertise. In both education and medicine the decline in professional unity has been met by a rise in political militancy.8 The point is that within virtually every professional sector policy conflicts have erupted with increasing frequency and severity, reducing thereby the chances of sector hegemony. The fourth possible basis for peak association formation is to increase the weight or influence of the sector in bargaining over policy. Surely there is appeal to potential members in enhanced influence, but this really is a problem in organizing for collective benefits, and Olson’s arguments apply. He does provide for the possibility of small groups organizing for collective purposes and some sectors might already be organized into oligopolistic form with a sufficiently modest number of discrete units to qualify as a small group with the accompanying face-to-face pressures that help overcome free rider tendencies. Elsewhere I have argued that within some limits, that might be fairly broad, it would be in the interest of each potential member of large groups to join so long as it is uncertain whether that membership increment may be decisive to the establishment of requisite sector influence [Salisbury]. But above those limits free riderism must reappear in the absence of selective benefits, and, as we have seen, these are difficult to manage effectively in peak organizations. We cannot say, on these grounds, that strong peak associations are impossible in the United States but at a comprehensive sectoral level the odds are against them. There is another factor in the American context that makes sector influence itself of problematic value. The assumption that influence is desirable for a sector is tied to another assumption: namely, that policy decisions will affect the entire sector in substantially the same way. For that to be so the policies themselves must be framed in


477

comprehensive terms, and the members of the sector must be in sufficiently similar circumstances as to have a common interest in the outcome. In fact, however, heterogeneity of interest or situation characterizes most sectors of American life. Pension fund regulation will not affect the Teamsters the same way it does the Steelworkers, for example. Nor are Harvard and Slippery Rock similarly situated regarding NSF policy. More than that, much American public policy tends to be highly disaggregated so that the differential effects within a sector are further heightened. The general point is that whenever a given policy has differential impacts, it adds to the centrifugal tendencies of the system, and in the United States the continued reinforcement of those tendencies constitutes a formidable obstacle for would-be peak associations to overcome. Again, medicine serves as an example of a sector where the expansion of explicit public policy has resulted in the steady proliferation of interests, each somewhat differently affected by existing programs and taking a distinct and separate posture toward new proposals.9 Another Possibility In one sense the point of this essay is a small one, and perhaps some might consider it even trivial. We are not suggesting that organized interests are without influence over American public policy. That influence is informal and often suspect, but it is not necessarily less consequential because it is unofficial. Some observers indeed have interpreted American politics as very largely a public, legitimate, process that masks the exercise of private and unaccountable power of the business elite.10 From this perspective the absence of corporatist mechanism might be regarded as irrelevant or even a matter of deliberate choice to hide the realities of economic power over political decision and to exclude potential rivals from the bargaining table. If there were genuine corporatism in the United States, labor would have to be a full participant in the negotiations over economic policy. They would have a full voice in determining which social sectors were to get what share of the rational product. Government, in turn, would hold not only the swing vote between business and labor but an important potential for affecting the demands and interactions of both sides. Moreover, corporatist mechanisms presumably operate regardless of election outcomes, whereas informal access may be much affected by them. Thus corporatist arrangements might well make a difference and a perceptive business elite might therefore wish to prevent a political evolution in that direction. Such an hypothesis cannot be refuted, of course, and the possibility, even as only a part of the total explanation, should not be dismissed. My own judgement is that insofar as a selfconscious business elite exists in the United States it has not devoted much attention to what economic planning mechanisms would or would not best suit its needs. Most of its energies have gone into the more general defense of the free market, on one hand, and the quest for specific subsidies and other direct benefits for particular industries, on the other. Still, the political strength of business in America and the concomitant weakness of labor may well help us understand why it has only been in times of severe crisis that both sides have been willing to sit together.


478

Conclusion It is possible that there is a somewhat simpler answer, or partial answer, to our original query. The argument put forward by Schmitter is that corporatism is a response, perhaps the model response, to the need for stability in an advanced capitalist system. Shonfield [1965:231] stresses the need for stability but sees corporatism as one of three possible styles of national economic planning by which such stability may be sought. In addition to corporatist planning there is state intervention, utilizing the levers of public power, and indicative planning whereby the planners persuade rational economic men to accept dispassionate analyses and act accordingly. In the United States there is some of each of these kinds of planning but not much use of corporatist devices. On that policy area most representative perhaps of corporatist tendencies, wage-price or incomes policy, the American approach has been mainly to utilize jawboning, occasionally to exercise the levers of federal power, and to play off one sector against the other. But there has been almost no serious effort to cultivate consensus or to facilitate the development of labormanagement agreements which would hold down inflation [Goodwin (ed.), 1975:368–9]. Panitch has argued that in Europe corporatism has not really worked as well as Schmitter and others believe, that the stability it brings is brief and the problems soon return [Panitch, 1976c, also Wilensky, 1976 and Wheeler, 1975]. In the United States there has generally been enough slack within each sector that a relatively free market process could operate at least some of the time, to bring downward pressure on prices. But even in the face of stubborn stagflation there is not much corporatism in America.11 Corporatism, in turn, is an aspect of comprehensive national economic planning and we may conclude this review by observing that there is not much planning in America either.12 Notes An earlier version of this paper was presented at the annual meeting of the American Political Science Association, Washington, DC, September, 1977. 1. For a parallel argument, viz., that it is values that differentiate American from European practice, see Anthony King, “Ideas, Institutions and the Policies of Governments Part III”, British Journal of Political Science, Vol. 3 (Oct., 1973), pp. 409–423. 2. In a wonderfully revealing statement Professor Jewel Cass Phillips once distinguished between “good” pressure groups and “bad” pressure groups. He then asked, “Are good pressure groups desirable?” After careful deliberation he concluded that they were not! State and Local Government in America, New York: American Book Co., 1954, p. 122. 3. Another example is provided by Joseph La Palombara’s characterization of the British style of operation: “When in doubt, clear it some more before taking action”, Politics Within Nations, Englewood Cliffs: Prentice-Hall, 1974, p. 354. 4. Roy C.Macridis is the author of another well-known argument embedding patterns of interest group behavior in the context of political culture. “Interest Groups in Comparative Analysis”, Journal of Politics, vol. 23 (Feb., 1961), pp. 24–45. 5. One might add that insofar as there is increasing insulation of elected officials from electoral


6. 7.

8.

9. 10.

11.

12.

479

insecurity, there will be still less reason to negotiate with organized associations. Rather, officials may feel more and more able to dominate the discussion. It would seem that such a development characterizes the contemporary politics of medical care more than would have been thought possible in the 1950’s. On the increasing safety of electoral margins, see Morris P.Fiorina, Congress, Keystone of the Washington Establishment, New Haven: Yale University Press, 1977. As a single example of a large and mostly uncharted area, see the large volume published by Congressional Quarterly entitled Washington Information Directory, 1977–78. A notable exception to the general ban on peak organizations resolving disputes among their members is provided in the exemption from antitrust prosecution granted to organized baseball. On the rising militancy in the National Education Association, see “NEA’s Growing Political Power”, National Journal, August 30, 1975. An instructive recent evaluation of the politics of medicine is John K.Iglehart’s “No More Doctor Nice Guy,” National Journal, March 6, 1976. A part of this proliferation process is treated by John K.Iglehart, “Health Report/ Economy Takings Steam from National Insurance Drive”, National Journal, January 18, 1975. I am grateful to Professors G.William Domhof and Robert Alford for their perceptive and probing comments on an earlier version of this paper. Both of them, in somewhat different ways, made suggestions that persuaded me I could not ignore the relevance of de facto corporate influence in a discussion of corporatism in America. N.H.Keehn has written a rather interesting essay, in part predicting and in part urging the coming of corporatism to the United States. I think he is wrong, but the analysis deserves attention. See “A World of Becoming: From Pluralism to Corporatism”, Polity, vol. 9 (Fall, 1976), pp. 19–39. I have tried to address this side of the problem in another paper, “On Centrifugal Tendencies in Interest Systems: The Case of the United States”, prepared for the World Congress of Sociology, Uppsala, Sweden, August, 1978.

References Bailey, S.K. et al. (1963) Schoolman in Politics. Syracuse: Syracuse University Press. Beer, S. (1958) “Group Representation in Britain and the United States”. The Annals 319 (September): 130–140. Block, W.J. (1960) The Separation of the Farm Bureau and the Extension Service. Urbana: University of Illinois Press. Blum, I.M. (1976) V was for Victory. New York: Harcourt, Brace, Jovanovich. Campbell, C.M. (1962) The Farm Bureau and the New Deal. Urbana: University of Illinois Press. Eldersveld, S.J. (1958) “American Interest Groups” in H.Ehrmann (ed.) Interest Groups on Four Continents. Pittsburgh: University of Pittsburgh Press. Fainsod, M., L.Gordon and J.Palamountain (1959) Government and the American


480

Economy. 3rd ed. New York: Norton. Fusfeld, D.R. (1972) “The Rise of the Corporate State in America”. Journal of Economic Issues 6 (March): 1–20. Garceau, O. (1941) The Political Life of the American Medical Association. Cambridge: Harvard University Press. Good win, C.D. (1975) Exhortation and Controls. The Search for a Wage-Price Policy, 1945–1971. Washington, DC: Brookings. Grossman, J.B. (1965) Lawyers and Judges, The ABA and the Politics of Judicial Selection. New York: Wiley. Hartz, L. (1960) “Democracy: Image and Reality” in W.N.Chambers and R.H.Salisbury (eds.) Democracy in the Mid-Twentieth Century. St.Louis: Washington University Press. Hartz, L. (1955) The Liberal Tradition in America. New York: Harcourt, Brace. Heinz, J. (1962) “The Political Impasse in Farm Support Legislation”. Yale Law Journal 71 (April): 952–78. Key, V.O. Jr. (1964) Politics, Parties and Pressure Groups. 5th ed. New York Crowell. Laslett, J.M. and S.M.Lipset (eds.) (1974) Failure of a Dream? Essays in the History of American Socialism. Garden City, New York: Anchor/Doubleday. Lowi, T. (1969) The End of Liberalism. New York: Norton. Masters, N.A. et al. (1964) State Politics and the Public Schools. New York: Knopf. McConnell, G. (1966) Private Power and American Democracy. New York: Alfred A. Knopf. McConnell, G. (1953) The Decline of Agrarian Democracy. Berkeley: University of California Press. Olson, M. (1965) The Logic of Collective Action. Cambridge, Mass.: Harvard University Press. Panitch, L. (1976c) “The Development of Corporatism in Liberal Democracies.” Paper presented to the American Political Science Association, September. Peterson, W.C. (1974) “The Corporate State, Economic Performance and Social Policy.” Journal of Economic Issues 8 (June): 483–506. Presthus, R. (1974) Elites in the Policy Process. London: Cambridge University Press. Salisbury, R.H. (1975) “Interest Groups” in F.Greenstein and N.Polsby (eds.) Handbook of Political Science. Vol. 4. Reading, Mass. Salisbury, R.H. (1969) “An Exchange Theory of Interest Groups,” in Midwest Journal of Political Science 13 (February): 1–32. Sapolsky, H. (1968) “Organizational Competition and Monopoly.” Public Policy 17:55–


481

76. Schattschneider, E.E. (1960) The Semi-Sovereign People. New York: Holt Rinehart & Winston. Self, P. and H.J.Storing (1962) The State and the Farmer. London: Allen & Unwin. Sharkansky, I. (1975) The United States: A Study of a Developing Country. New York: David McKay. Shonfield, A. (1965) Modern Capitalism, The Changing Balance of Public and Private Power. London: Oxford University Press. Singer, J.W. (1976) “Buttonholing and Buttering up for Labor—and ‘the People’.” National Journal (April 24). Truman, D.B. (1951) The Governmental Process. New York: Alfred A.Knopf. Wheeler, C. (1975) White Collar Power. Urbana: University of Illinois Press. Wilensky, H. (1976) The New Corporatism, Centralization and the Welfare State. Beverly Hills, Calif.: Sage. Wooten, G. (1970) Interest Groups. Englewood Cliffs: Prentice-Hall. Zeigler, H. (1964) Interest Groups in American Society. Englewood Cliffs: Prentice-Hall.


482

58 Constitutionalism, Class, and the Limits of Choice in U.S. Foreign Policy* Ira Katznelson and Kenneth Prewitt

* Source: T.Fagen, ed., Capitalism and the State in US—Latin American Relations, 1979, pp. 25–40.

As an object of explanation, U.S. policy toward Latin America has been more diverse and more changeable than many analysts have allowed. Latin America is an unwieldly unit for analysis. The immense differences between nation-states—in size, wealth, geopolitical capacity, character of regime, and independence from U.S. direction—and the vast range of alternative policies practiced by the United States at any one time— including disaster aid, counterinsurgency training, credit, economic “destabilization,” immigration control, drug traffic management, and, on occasion, direct military intervention—make it difficult to speak in unitary terms about U.S. policies toward the hemisphere as a whole. This difficulty is compounded by the obvious oscillations in emphasis in both the rhetoric and the performance of U.S. policy makers, who have alternated over time between “soft” policies associated with being a Good Neighbor and with the Alliance for Progress and “hard” policies such as the Bay of Pigs, the Dominican invasion of 1965, and the unremitting hostility to the Allende experiment in constitutional socialism. From Kennedy to Carter the overall trajectory of U.S. policies has been one of less overt intervention, some economic disengagement, and more rhetorical concern with human rights and the quality of life. No marines have landed since 1965. With the obvious exception of Chile U.S. administrations have worked hard to transmit a more benign image of their activities in the hemisphere. Neither the complexity of U.S.-Latin American relations nor the changes in administration emphases, however, should be allowed to obscure a very high degree of fundamental continuity in U.S. policy. Two underlying principles have been most important: first, the maintenance of a skewed pattern of north-south geopolitical capacity; and second, the maintenance of a skewed pattern of distribution of economic opportunities and rewards. The whole gamut of U.S. policies concerned with aid, trade, activities of international organizations, investment, and military security has been directed, at least since the end of the Second World War, at “the preservation and extension of North American political, economic, and cultural influence and domination in the hemisphere, at the lowest possible cost.”1 These practices have maintained dependent industrialization characterized by growing foreign control over industrial development, competitive advantages to foreign firms, a deformation of local industry, planned expansion in harmony with the needs of external purchasers of exports rather

Constitutionalism, Class, and the Limits of Choice in U.S. Foreign Policy

483

than with those of the home market, and the outflow of profits. In addition, they have supported the formation and dominance of clientele local classes with a vested interest in such patterns. The varied policies of the United States have been vital to the maintenance of this status quo in a number of ways. The United States has set clear and widely understood limits to the acceptable behavior of Latin American governments. In so doing, it has provided regimes willing to operate within these bounds with the means to survive in spite of their shaky legitimacy. The United States also provides the security necessary to insure the capacity of its firms to participate in the bounty of dependency. Policy analyses of U.S. activities in Latin America have generally proceeded at two very different levels: on the one hand, political economy explanations have made sense of the overriding goals of dominance and the maintenance of a structure of dependent inequality; on the other hand, ad hoc treatments have made sense of alterations in policy over time, stressing changing situational factors. Neither approach is wrong so much as insufficient. Each fails to describe “middle-range” characteristics of the U.S. regime that are special in the universe of capitalist democracies yet more enduring than the monthly or yearly flow of events. Concretely, we argue that two closely related features of the U.S. regime are basic to an understanding of the pattern of oscillations in foreign policy within a consistent set of hegemonic goals. The first of these characteristics is the “low stateness” of the polity. Though the size and functions of the state have grown enormously in this century, and especially since the Second World War, the United States is a peculiar democracy with an ideology and practice of divided and circumscribed government. One keen observer, J.P.Nettl,2 went so far as to say that at the domestic level the United States lacks a state in the German or French sense. There is, in short, a paradox of big but diffuse government. The second characteristic is the “low classness” of social and political life in the United States. The country is the most developed of all the capitalist societies and thus is divided objectively between collective capital and collective labor more fully than any other. Yet with the partial exception of workplace relations, U.S. politics is organized on a plurality of affiliations of a nonclass kind, and the country lacks even the modest kind of social democracy associated with Great Britain or West Germany. The impact of these distinctive features of the regime on domestic politics and policymaking is overwhelming; in various ways, to be sure, virtually all modes of social and political analysis deal with them when internal U.S. affairs are being considered. Their relevance for a consideration of foreign policy obviously is more opaque. Accordingly, we wish to complement work in the political economy tradition as well as ad hoc case studies by exploring connections between the state as an actor in the international system and those of its durable features that may condition and limit foreign policies. First, we will explore the middle-range characteristics of the U.S. regime and specify the dynamic relationship between the character of state and class relations over time. Second, we will suggest ways in which these middle-range features in fact act importantly, if mostly at the margin, to inform the content of U.S. foreign policy.

State, Class, and the U.S. Regime


484

The term “capitalist state” has been the subject of much theoretical and historical discussion within the Marxist tradition, revolving around the question, How can democracy and capitalism be maintained and recreated simultaneously? Virtually all the answers—whether they emphasize differential recruitment patterns, the special capacity of capital as an interest group, the “selective” character of the mechanisms that link society and politics, or the dependence by the state on a revenue surplus generated by successful capitalist accumulation—share with Nicos Poulantzas the formulation that it is the state that manages the tensions between capitalism and democracy by functioning as the “factor of political organization of the dominant classes” and as the “factor of political disorganization” of the working classes.3 This approach rightly understands that the state is not simply a passive recipient of class and group pressures but an entity whose very structure conditions whether and how class will be made politically relevant. This process and its effects cannot be understood, however, at the generic level of the “capitalist state” except to indicate or project broad tendencies. Rather, within the structural logic imposed by industrial capitalism, individual states with their own form of government are connected to social structures conditioned by different historical traditions. If the ideal-typical capitalist state tends to organize the dominant and disorganize the subordinate classes, we still must ask, How do distinctive states actually perform these functions? With what mechanisms and with what degree of success? How does the state shape the unequal and competing capacities of classes as historical social groups? Our discussion of the regime in the United States revolves around these issues. The pattern of class formation, one of “low classness,” developed before the Civil War and was caused largely by the constitutional organization of the polity. In turn the pattern of class relations has contributed to the reproduction of the “low stateness” of U.S. democracy. Over time these mutually reinforcing elements have fashioned a regime distinctive in the West, one whose key features inform and constrain the making of foreign policy. The United States is characterized by an extraordinary diversity of collective affiliations. Class does not provide the central theme organizing political conflict. Class relations do not seem as simple or transparent as elsewhere in the industrial capitalist world. The links between categories of individuals shaped by their places in production and social and political behavior appear more remote, or at least less obvious. Given that the United States has the most developed of the capitalist economies this anomaly has been very puzzling. As in all the other capitalist states, the “economic motion” of capital underpins struggles at work, in residence communities, and between citizens and the state. But more than in any other capitalist society, each of these arenas of conflict has been unusually enclosed and encapsulated in the United States. The links, especially between work and community, have been tenuous. Each locus of conflict has had its own vocabulary and form of institutional expression—residence and groups, local parties, churches, and voluntary associations; workplace and unions. Class has been lived and politically expressed as a series of partial relationships and is experienced, therefore, as one of a number of competing bases of social life. The differentiation of social life into work, community, and state relations is, of course,


485

an objective feature characteristic of all capitalist (and perhaps all industrial) development. Because life outside the workplace is free from the authority of capital or management and because workers, unlike slaves, sell their labor time and not themselves, patterns of life and culture in residence communities constitute in large measure an autonomous sphere of private life. Here the worker is a consumer, a member of a family, a resident of a territory, a person interacting with others, sharing ethnicity, race, and interests. Such interactions may be affective or instrumental or both. It is hardly surprising to find that work lives and community lives may be perceived as unconnected. What is distinctive about the experience of the United States, however, is the extent to which the linguistic, cultural, and institutional meanings given to this differentiation have taken such an extremely serial character in the polity for so long. What has been exceptional about the experience of class in the United States, in short, is that the split between work and community has been reproduced ideologically and institutionally in ways that have fragmented patterns of class in qualitatively distinctive fashion. Elsewhere in the West the tendency to parcelization has been partly countered by competing “global” institutions, which provide understandings of class. Such interpretations are to be found in the history, rhetoric, programs, and organizational strategies (joining work and home) of social democratic, socialist, laborist, and communist parties, and of voluntary associations. They often, but do not necessarily, entail revolutionary interpretations of capital and the class structure. It is the virtual absence of even moderate global approaches to class—and the massive resistance to such attempts in the past—that is so striking about the U.S. experience. As early as the Civil War, the American working class, at the level of class formation, had become labor; it has remained so to the present. The nonwork aspects of the lives of working people have been organized into politics on other bases, including those of race, ethnicity, and territoriality. Because workers are not structured into political life in terms that produce a coherent working class at the level of dispositions, workers lack the capacity to counterpose the interests of capital as a whole to those of the working class as a whole. Rather, workers affect political life as members of interest groups that sometimes utilize the language and grammar of class and sometimes the language and grammar of other group identifications. The sources of this unusual pattern of class relations may be located in the intersection of the dynamics of capitalist development and citizenship in the three decades before the Civil War. In this period the creation of a modern working class, principally in the older, predominantly mercantile cities of the East, which entailed the physical and social separation of work and community, was accompanied and given meaning by the culmination of a number of reinforcing political trends that collectively defined the terms by which workers would be linked to the polity: federalism, franchise extensions, a modern national party system, and its neighborhood (not yet centralized) machine affiliates. Citizenship intersected community, not work. In this way citizenship and its bases were given communal meanings separate from work relations. The segmented pattern of class understandings in the United States thus was caused principally by features of the polity created by the operation of a federal constitutional system. The constitutionalism of the regime has continued to matter, not only for the character of class relations but also for the ways constitutionalism has shaped a polity of “low stateness.”


486

From the perspective of the international system, “the state is the basic irreducible unit, equivalent to the individual person in a society.”4 The salience of a state—“stateness,” in Nettl’s term—is highest where foreign affairs are concerned. However, not all nations exhibit the same degree of stateness. The salience of a state is highly variable, depending in large measure on the character of its organization, the autonomy of other sectors of society, the degree of plurality in modes of representation, and the language and idiom of political life. “Low stateness,” paradoxically, is compatible with big government and with an assertive foreign policy; and such is the case in the United States. The “stateness” of the United States in the international arena is at least as great as that of any other nation-state. The growth in the size and capacity of government is also not in doubt. Total government expenditures accounted for only about 7 percent of the GNP at the turn of the century, compared to 37 percent today. Yet, understood comparatively, the state—the autonomous organized sector of society that authoritatively taxes, preserves order, potentially conscripts, makes foreign policy—is less tangible, more diffuse, and more interpenetrated by nonstate actors than virtually anywhere in Western Europe. “Low stateness” in the United States is foreshadowed in the design of the constitutional system and can best be described by turning to the originating document, the Constitution itself. Now in existence for almost 200 years, the Constitution proclaims a theory of government, popularizes a language of politics, establishes decision rules and institutional arrangements, and creates a political regime accepted largely without question. To be sure, there have been serious political conflicts about the interpretation of the Constitution, the most violent of which became a full-fledged civil war; but with the exception of the Civil War era, major political movements have not been organized around a demand for a different constitution. The Constitution is more than a set of political rules; it was designed to be compatible with an economic framework under which citizens expect to find happiness by maximizing private gain. The Constitution does not establish a state that in turn manages the affairs of society toward some clear conception of the public welfare; rather, it establishes a political economy in which the public welfare is the aggregate of private preferences. The constitutional order has benefited quite directly from the impressive standard of living provided many citizens by U.S. industrial capitalism. Consumer satisfactions in a materialistic society shade over into commitment to the “American way of life,” which in turn translates in the public mind as loyalty to the constitutional order. There are additional and more “political” reasons for the stability enjoyed by the constitutional regime in U.S. history. For example, it is hard to rebel against “low stateness,” for there is not much of a target for rebellion. This does not mean that every segment of the population is necessarily satisfied with public policy. But dissatisfaction most frequently takes the form of trying to force a new and more favorable interpretation of the Constitution. Black anger, for example, has been directed not at the dissolution of public order but at racist public policies and laws, at making the Constitution work. Social movements, especially when fueled by egalitarian or redistributive demands, tend to express anger at those public policies that get in the way of some group’s ability to maximize private interests (upward mobility for blacks, equal treatment for women). Public policies neutralize this anger by improving the access of the deprived group to the


487

rewards of a consumer-oriented capitalism. That this may deprive some other group (white male workers, for instance) only calls forth another round of policies to ameliorate their discontent. Never in this endless shuffling does the Constitution itself become the target. Rather, constitutional principles legitimate claims for a fair share of “the American way of life,” and constitutional interpretations and reinterpretations are the means for forcing reallocations. The design of the Constitution was not merely an attempt to create a framework for the pursuit of gain. It was an experiment with unprecedented political forms aimed at avoiding political tyranny and establishing real, if limited, popular sovereignty. The fear of centralized authority and the search for democratic safeguards led the designers of the new government to the principle of limited powers, which was to be realized primarily through dividing and diffusing sovereignty. The United States was the first nation to selfconsciously experiment with the heretical doctrine that sovereignty need not be established with a single pyramid of power. There was to be no unified state at the apex of public life in the United States. The writers of the Constitution identified their primary challenge as the creation of a government that would have sufficient powers to control the governed while also being obliged to control itself against the tendency toward tyranny. This was to be accomplished, according to The Federalist Papers, “by so contriving the interior structure of the government so that its several constituent parts may, by their mutual relations, be the means of keeping each other in their proper place.”5 A government of divided powers was established in two steps. First, formal government powers should be divided between two layers of government—national and state, or central and regional. The second dimension of divided powers is the doctrine that the different branches of government—legislative, judicial, and executive—were autonomous in their own spheres, each being given the constitutional means to resist domination by the other two. Together the two dimensions of divided powers significantly lowered the salience of the state. Nettl understands this point precisely in his review of the separation of political authority among different layers and branches of government. “Hence in the United States,” he writes, “the real boundaries of autonomy fall not between the state and other institutions but within and between the complex of institutions that elsewhere would be encapsulated within the collectivity of the state…. This internalization of distinct areas of institutional autonomy is the fundamental implication of the separation of powers.”6 Limited government was to be achieved not only through separation of powers but also through the application of such doctrines as “a government of laws, not men.” The independent judiciary is particularly critical here. In addition to adjudicating civil and criminal transgressions against society, the judiciary measures the actions of government itself against the principles of the Constitution. Stated more generally, the United States is a government of legislation and litigation. Where the political order is legitimate, as U.S. constitutionalism is, politics becomes the struggle to translate social and economic interests into law. Once group interests become law, they cease to be group interests and appear as those of society in general. The government then assumes the task of getting others in society to do things in a manner that promotes the interests of the group. If legislation is at the core of those processes that translate group interests into general law,


488

the political culture defines political power as getting a law passed. Though starting from a different point, Nettl reaches a similar conclusion. In his description of the United States, he notes that “in practice only law is sovereign, and probably the ‘function’ of sovereignty can indeed best be taken as being fulfilled and institutionalized by the law.”7 Constitutional democracy and the sovereignty of law, together with the segmented character of workplace and community bases of affiliation, constitute the idiom of U.S. politics. Political business could not be conducted without such terms as “citizen participation,” “majority rule,” “due process of law,” “accountable leaders,” “regular elections,” “civil rights,” “civilian control of the military,” and “free press.” Except in rare instances and in small numbers, U.S. citizens do not question whether they are governed by democratic procedures, let alone whether democracy is the best form of government. Group and class conflicts, tensions, and struggles are refined in this crucible.

The Limits of Choice in U.S. Foreign Policy The regime elements of “low classness” and “low stateness” affect U.S. foreign policy not by providing the basic impetus for the international activities of the state but by creating a context for policymaking that is distinctive among the advanced capitalist democracies of the West. These special features of the U.S. regime modify, shape, and also obscure basic imperatives of the political economy in a number of important ways. They form the group and class actors who pay attention to foreign policy. They fashion a powerful ideology that organizes public discourse about foreign policy. The language of policy discussions between groups and classes in constitutionally defined arenas is ideological not only in the sense of masking exploitative relationships but also in the sense of creating a policy-relevant political culture that shapes apparently contradictory initiatives like support both for authoritarian regimes and for the promotion of human rights. The character of the arena of U.S. politics and of the relevant contending groups defines “legitimate” choices that policymakers can consider in public. It also compels them to operate covertly where this political culture does not permit open activity. Finally, the “low classness” and “low stateness” of the regime make it very difficult, if not impossible, to fashion the political agency required to make even modest progressive changes in current U.S. practices in Latin America. Let us take up these matters in turn. All the capitalist democracies of the West act to reproduce a capitalist order because they cannot do otherwise. These states depend in the first instance—whatever their aims apart from the recreation of capitalism—on the productive apparatus to generate a surplus that can be taxed to provide revenue for state activities. Making the economy work is both the primary goal and the political test of incumbent governments. Dictates of political prudence thus reinforce the imperatives of the political economy to make the needs of business the chief business of government. Virtually everywhere in the West, however, these powerful imperatives are modified by working-class organizational and ideological capacities. Yet the distinctive “low classness” of political and social life in the United States largely precludes these working-class pressures from playing an important role in the making of domestic and foreign policy. As a result, the U.S. state possesses an unusually strong capacity to insulate class-interested policies from the countervailing


489

pressures of an organized working class, and to make many public policies appear technically and politically neutral with solutions beneficial to all members of the society. Intense public attention to issues is thus limited largely to those matters that directly affect highly organized groups and associations. The very pluralism of American life outside the workplace thus fashions a degree of freedom for the relations between the state and capital virtually unknown elsewhere. Given the absence of a socialist or social democratic ethic, ideology, and definition of the working class as a whole, workers appear in the political process almost exclusively as organized labor. The capacities of organized labor to affect domestic and foreign policy are considerable. In the area of social welfare expenditures, for example, precisely those programs in the relatively emaciated American welfare state that directly affect the material well-being and security of the unionized 20 percent of the workforce— especially programs of social insurance—have been the ones to grow the fastest and to show the least discrepancy with expenditure patterns in Western Europe. With respect to foreign policy, organized labor is a member—albeit the least important member—of a specialized coalition whose other members are transnational capital and the national security managers of the state. Once policies are considered in terms not of class conflict but of a cross-class, hegemonic interest-group framework, labor by and large joins with capital and the state in the role of interest group. Organized labor, then, fully shares with its coalition partners an interest in current relationships of domination, for it harvests with its partners the fruits of the structure of dependency. Most of labor’s activities in Latin America in the past half century, including its close cooperation with U.S. intelligence agencies, are comprehensible only in those terms. The plurality of nonwork identifications and affiliations in the United States produces much intense attention to domestic policymaking. Clients of specific state programs, for example, together with those who administer such programs develop a stake in their continuation and expansion.8 Such collectivities—along with organized labor, traditional interest groups, and racial and ethnic groups—all press with varying degrees of success to shape government activity. By contrast, the freedom of the state to act in foreign affairs is much greater, because those outside the foreign policy coalition rarely pay much attention to international matters. Indeed, unless a foreign relations issue cuts close to questions of economic well-being (more often than not defined in hegemonic terms) it is unlikely to draw widespread public notice. Selective inattention leaves the foreign policy coalition relatively unconstrained, except in cases where particular subpublics closely monitor particular issue arenas, as is the case with pro-Israel groups and Middle East policies. On the whole, such subpublics do not exist with respect to Latin America, the main exceptions being associations like Amnesty International, concerned with civil and political rights. “Low classness” and “low stateness” also fashion an idiom of politics that makes an active role in foreign affairs very difficult for the public. The ordinary democratic, constitutional grammar of U.S. political discourse has no place for terms like “imperialism,” “military aggression,” “covert operations,” “puppet regime,” and “subversion.” The absence of an appropriate vocabulary and language does not mean that U.S. citizens support military aggression, such as the Bay of Pigs, or covert operations, such as the undermining of the Allende regime; but these acts, even when recognized for


490

what they are, are dismissed as temporary deviations. Consider Vietnam. There was a brief period during which the anti-war movement called into question the benign nature of U.S. foreign and military policy. But the general public, even when it began to question the wisdom of the war, never adopted the language of the anti-war movement. Moreover, the end of the war quickly removed the rhetoric of the anti-war movement from public discourse, dismissing it to the left-wing journals and scattered academics where and among whom it had originated. American foreign policy was scarred by Vietnam, but the democratic idiom so central to the political culture remained intact. Even in more routine periods the democratic idiom disguises the imperialism of much of U.S. foreign policy. As we noted earlier, it also plants in the political culture the lesson that politics is primarily law-passing—not a lesson easily applied in the international setting. The language of U.S. politics, however, does not merely contribute to a masking of state activities; it also helps shape them. This “positive” effect takes two forms. First, the routine grammar of political life makes it important for policymakers to emphasize those features of their activities that mesh with the constitutionalism of the regime: the United States acts to promote liberty and human rights abroad, to alleviate suffering, to help the poor help themselves. Though hardly the only cause of the “soft” side to U.S. policies abroad, the need to explain and legitimate state activities in terms of the ordinary language of a regime of “low stateness” and “low classness” should not be underestimated. Those features of foreign policy most easily apprehended in the language of democracy—as in the case of individualistic human rights—are the most clearly recognized by the public at large, thus placing limits on what government can do. Second, the language and understandings of constitutionalism help define the boundary between overt and covert activity. Obviously, a “constitutionally dominated” foreign policy is hardly the whole picture. Many features of U.S. policy are clearly incompatible with the rules and spirit of the constitutional game. The image of a “limited government” cannot easily mesh with the realities of a national security state that include a far-flung intelligence network, support for military regimes around the world, a propaganda program aimed at audiences within the United States and abroad, active promotion of and participation in regional military-security alliances, and sustained research and development in weaponry. Such ideas as legislation-dominated political processes, and such doctrines as separation of powers, are in apparent contradiction with an executivecentered foreign policy process and descriptions of the “imperial presidency” that have become fashionable in the United States. More strikingly, how can it be said that constitutional principles guide a foreign policy that makes extensive use of covert operations, repression of dissident opinions, bribery, secrecy, and the like? Alan Wolfe has suggested that the United States has been governed by a “Dual State”: “In domestic politics, there existed a state that was popular, democratic, constitutional…. Centered in the legislative branches of government, it was based on rules that were reassuring: executives executed, legislators legislated, and judges judged. Certain standards of conduct—like due process of law, democratic representations, and appeals to history and tradition—were expected to receive homage.”9 Wolfe argues that these constitutional principles were impotent in the 1960’s and early 1970’s, at least in foreign policy, as they increasingly were replaced by


491

authoritarian and secret government. “At some point during the 1960’s, then, a fully developed Dual State had come into existence. Because it was secret, very few were aware of its creation…. Because it had no public accountability, it was arrogant and ruthless. …Because it considered itself omniscient, it developed its own language, its own code words, its own rules of proper reality .”10 Wolfe is arguing that in an effort to protect its commercial empire and to extend its means of accumulating capital, the United States suspended the rules of the Constitution in order to pursue a repressive foreign policy. Borrowing from Wolfe, we can conclude that a more inclusive picture of the foreign policy process would have to understand the conditions under which there is constitutionally dominated policy and the conditions under which the Constitution is for all practical purposes suspended.11 The suspension of the Constitution no doubt occurs when there is a crisis. We suggest that foreign policy oscillates between being constitutionally dominated and being crisis-dominated. Of course, this only begs the question. The analysis would have to deal with what a crisis is and who gets to do the labeling. Such an analysis would too greatly extend our discussion, but in preliminary and tentative fashion we can suggest a distinction between “real crises” and “manipulated crises.” A “real crisis” might be a complete oil embargo, invasion of Taiwan by Communist China, Soviet-supported coups across Africa, and so forth. Attentive publics in the United States would probably be willing to tolerate suspension of certain constitutional constraints, as they did with the Tonkin Gulf resolution or the internment of Japanese-American citizens at the beginning of the Second World War. A “manipulated crisis” might occur when some special interests abroad needed to be protected, such as American business interests when Allende was elected. There might be a suspension of constitutional constraints, but covert action would be much more likely, because it would be difficult to sustain a suspension. One way to think about foreign policy under different presidents is to ask how much room there is for “manipulated” crises. Certainly there is less room under Carter than there was under Nixon. Finally, the middle-range features of the U.S. regime make progressive reforms in Latin American policy more difficult to achieve than in other capitalist democracies. There is no dearth of concrete policy proposals that would assist more progressive forces in the hemisphere. One such list includes changes in the present pattern of trade preferences and commodity arrangements; debt rescheduling; the normalization of relations with Cuba; the cessation of police and counterinsurgency aid; the liquidation of the Overseas Private Investment Corporation, which protects U.S. investors from market risks; and a shift from loans to grants in development assistance.12 None of these changes would require revolutionary transformations in U.S. capitalism or democracy. Nevertheless, they are highly unlikely to be implemented (even by an administration anxious to project benign intentions to the Third World)—first because of the features of U.S. politics and society discussed above, and second, because of the absence of state capacity to direct the internal distribution of costs entailed by a less imperialist foreign policy. The constitutional doctrines of limited government, divided powers, and individual liberty, coupled with the practices of a market economy and the absence of a working class at the level of class formation, put a limit on the type of central


492

authoritative agency that can emerge in the United States. The “low stateness” of the regime is especially demonstrated by the absence of a centralized planning function. Central government planning with respect to land use, regional distributions, labor market trajectories, and industrial policy are common in virtually all of the capitalist democracies. In the United States, by contrast, although individual state policies and clusters of policies have important ramifications in each of these areas, national economic planning on the European model does not exist. The absence of a planning tradition or function in the United States makes it difficult (some say impossible) for the United States to absorb the costs of any substantial reform in foreign policy toward Latin America. These policies are constructed upon and maintain asymmetric power relations and skewed distributions of economic resources between the north and the south. The consequent U.S. advantages work to the benefit of nearly all segments and classes of U.S. society. The high consumption standards of advanced industrial capitalism, it should be remembered, produce a standard of living even for its poorer classes that easily outdistances the subsistence-level conditions of the majority of peasants and workers in the Third World. A progressive foreign policy toward Latin America (or, more generally, toward the nonindustrialized, low-consumption nations) would have equity as one of its goals and would include some attempt to reduce the enormous disparity between the hemispheres. At a minimum this would involve tolerance of Latin American regimes committed to equity within their own societies—exactly the regimes that present U.S. policy consistently subverts. What impedes the development of a more progressive foreign policy is suggested by the larger point we have made about “low stateness” and the absence of a planning function in the United States: any policy that would close the relative gap between the north and the south would involve some reduction in economic growth and consumption in the United States. The costs of this reduction would have to be allocated across the different regions and classes of U.S. society. In the absence both of any larger views about the appropriate levels of production and consumption and of some planning function to implement these views, the allocation of such costs becomes exceedingly difficult. Given the diffuseness of power, the democratic controls, and the hold of marketplace values in the United States, any attempt to allocate the costs of reduced consumption would impose heavy political costs. The fate of affirmative-action programs is instructive. Affirmative action was introduced and promoted as a way to overcome several centuries of racism, but once understood by the public as a policy to redistribute rewards (jobs and schooling for jobs) from the classes that have controlled those rewards to the classes that have not, it aroused political opposition with a forcefulness that has attenuated it, if not dismantled it. An analogy to foreign policy is not far-fetched. It took a century of internal strife for the majority of white Americans to acknowledge the racism that they lived with every day. To recognize the more distant imperialism and exploitation in foreign affairs is obviously a much more difficult task. Without this recognition, however, “affirmative action” in foreign policy is not likely. Assume for the sake of argument that the U.S. public could be made to understand that it shares responsibility for the exploitation of Latin American societies. Should we then expect “affirmative action” in which some deliberate attempt to atone for that


493

exploitation would be made? The question is obviously rhetorical. Given the sharpened competition for the rewards of the international marketplace, the United States will not voluntarily forgo its “rightful” share, measured by its present standard of living. The regime of “low classness” and “low stateness” is not compatible with the emergence of the kind of “agency” in the United States that could internalize and politically allocate the costs of an international order based on global equity. We thus reach the conclusion that political processes internal to the United States will not generate a major transformation of foreign policy toward Latin America, though there will continue to be oscillations in policies depending on the character of different presidential administrations and the access enjoyed by different elements within the foreign policy elite. Oscillations from “hard” to “soft” policies, however, will occur within limits set by the policy goals of maintaining geopolitical hegemony and hemispheric inequities. If there is to be a serious challenge to current foreign policies, it will come from outside the United States, most likely from Latin American nations themselves.

Notes Ira Katznelson chairs the Department of Political Science at the University of Chicago. He is the author of Black Men, White Cities: Race, Politics, and Migration in the United States, 1900–1930, and Britain, 1948–1968, and City Trenches: The Patterning of Class in the United States (forthcoming). Kenneth Prewitt is President of the Social Science Research Council and has taught political science at the University of Chicago and Stanford University. He is the author of The Recruitment of Political Leaders: A Study of Citizen-Politicians, and co-author of Labyrinths of Democracy: Adaptations, Linkages, Representation, and Policies in Urban Politics. 1. Richard R.Fagen, “Commentary on Einaudi,” in Julio Cotler and Richard R. Fagen, eds., Latin America and the United States. The Changing Political Realities (Stanford, Calif., 1974), p. 262. 2. J.P.Nettl, “The State as a Conceptual Variable,” World Politics 20 (July 68), pp. 559–92. 3. Nicos Poulantzas, Pouvoir politique et classes sociale (Paris, 1971), vol. 2, p.l 16. 4. J.P.Nettl, “The State as a Conceptual Variable,” p. 563. 5. Federalist Paper No. 51. 6. J.P.Nettl, “The State as a Conceptual Variable,” p. 569. 7. Ibid., p. 514. 8. Morris Janowitz, Social Control of the Welfare State (Chicago, 1977). 9. Alan Wolfe, The Limits of Legitimacy (New York, 1977), p. 179. 10. Ibid., p. 203. 11. Wolfe raised this point in discussion at the conference where an earlier draft of our essay was presented. 12. Fagen, “Commentary on Einaudi,” in Cotler and Fagen, eds., p. 260.

CONCLUSION: The End of the State?


Commentary It is hard enough to explain the past, let alone to predict the future. Nonetheless, the spectacular recent growth in the interdependence of the world economy suggests the need for at least considering the possibility that the role of the state may be diminished in the future. The articles in this conclusion differ in tone, some being more prescriptive than descriptive, but all can usefully be seen as warning against any complete embrace of the view that the future will be stateless. R.Gilpin’s Three Models of the Future’ is by now nearly twenty years old, but it retains great power—not least because it structured much debate that follows. Two of the futures that Gilpin envisages are ones in which economics dominates politics: in the first transnational corporations will increasingly bring prosperity to the world economy, whilst in the second the rich core of capitalism will exploit ever more systematically its peripheral regions. The third model that he had at the front of his mind, that of state-led mercantilist reactions against the market has, at first sight, suffered a considerable blow from the failure of protectionism, whether in Latin America or in state socialism. On reflection, however, states still seek to advance their economies: if the particular tools of protectionism have failed, this merely means that greater attention will now be given to export-led industrialisation. The tenor of Susan Strange’s ‘Supranational and the State’ can be gleaned from her observation that transnational corporations are seen as hollowing out the American economy and as ensuring the subservience of the Third World. Both views are plausible, but they cannot be absolutely right at the same time. Strange’s highly sophisticated view depends upon a refusal to think in either/or terms. Differently put, she considers that the state is losing some of its powers precisely at the moment in which it is strengthening other of its competencies. The workings of the international political economy are seen by Strange as being so fast and so novel that we are hard pressed even to understand, let alone to control, what is actually happening. Particular interest attaches to Robert Reich’s ‘Who is “Us”?’ given his move from academia to mainstream politics. Increasing interdependence has resulted, according to Reich, in a bifurcation of social interests within the United States: those capable of dealing with high technology have done extremely well, and accordingly have an interest in the openness of the world economy, but those without such skills have seen their real position decay. Reich presents a prescriptive argument for national renewal, by means of educational provision for all citizens, so as to ensure America’s future success in the world economy. This is an argument in favour of state intervention, albeit Reich insists that such intervention, even though it is based on nationalism, need not be directed against any particular enemy. Past nationalism has, however, often been precisely a zero-sum affair, and it is this which lends especial interest to R.Lepsius’s ‘Beyond the Nation-State: The Multinational State as the Model for the European Community’. The attempt to create a European super-state, that is, a genuinely transnational political organisation akin in character to the United States, still looks both unlikely and, given its tendency to reinforce nationalist

sentiments, undesirable. Lepsius offers a model of a multinational community seeking to combine the uniquenesses of states within a larger federal culture. This proposal is at once realistic, and historically sensitive.


498

59 Three Models of the Future* Robert Gilpin

* Source: International Organization, 1975, vol. 29, pp.37–60.

Edward Hallet Carr observed that “the science of economics presupposes a given political order, and cannot be profitably studied in isolation from politics.”1 Throughout history, the larger configurations of world politics and state interests have in large measure determined the framework of the international economy. Succeeding imperial and hegemonic powers have sought to organize and maintain the international economy in terms of their economic and security interests. From this perspective, the contemporary international economy was the creation of the world’s dominant economic and military power, the United States. At the end of the Second World War, there were efforts to create a universal and liberal system of trade and monetary relations. After 1947, however, the world economy began to revive on the foundations of the triangular relationship of the three major centers of noncommunist industrial power the United States, Western Europe, and Japan. Under the umbrella of American nuclear protection and connected with the United States through military alliances, Japan and Western Europe were encouraged to grow and prosper. In order to rebuild these industrial economies adjacent to the Sino-Soviet bloc, the United States encouraged Japanese growth, led by exports, into the American market and, through the European Economic Community’s (EEC) common external tariff and agricultural policy, also encouraged discrimination against American exports.2 Today, the triangular relationship of the noncommunist industrial powers upon which the world economy has rested is in disarray. The signs of decay were visible as early as the middle 1960s, when President John F.Kennedy’s grand design failed to stem the coalescence of an inward-looking European economic bloc and to achieve its objective of an economic and political community extending from Scandinavia to Japan and pivoted on the United States.3 Believing that the world trading and monetary system was operating to America’s disadvantage, the administration of Richard Nixon took up the challenge with a completely different approach. On 15 August 1971, former President Nixon announced a new foreign economic policy for the United States. In response to the first trade deficit since 1893 and to accelerating attacks on the dollar, the president imposed a surcharge on American imports, suspended the convertibility of the dollar, and took other remedial actions. Subsequently the dollar was devalued twice (December 1971 and February 1973); the world moved toward a system of flexible exchange rates; and intense negotiations were initiated to create a new international monetary and trading system. A new economic policy was necessary for several reasons. The United States believed

Three Models of the Future

499

an overvalued dollar was adding significantly to its unemployment rate.4 American expenditures abroad for military commitments, foreign direct investment, and goods and services required, in the 1970s, greater outlays of foreign exchange than the United States could earn or wished to borrow. The US rapprochement with China, its moves toward détente with the Soviet Union, and President Nixon’s announcement of the New Economic Policy appeared to signal the end of the political order that American economic and military supremacy had guaranteed; this political order had been the foundation for the post-World War II world economy. All these policy initiatives were efforts to adjust to the growing economic power of America’s partners, Europe and Japan, and to the growing military power of its primary antagonist, the Soviet Union. In terms of the present article, these economic and political changes raised the question of whether the interdependent world economy could survive in the changing political environment of the 1970s and beyond. In this brief article I make no attempt to give a definitive answer to this question. Rather, my purpose is to present and evaluate three models of the future drawn from current writings on international relations. These models are really representative of the three prevailing schools of thought on political economy: liberalism, Marxism, and economic nationalism. Each model is an amalgam of the ideas of several writers who, in my judgment (or by their own statements), fall into one or another of these three perspectives on the relationship of economic and political affairs. Each model constitutes an ideal type. Perhaps no one individual would subscribe to each argument made by any one position. Yet the tendencies and assumptions associated with each perception of the future are real enough; they have a profound influence on popular, academic, and official thinking on trade, monetary, and investment problems. One, in fact, cannot really escape being influenced by one position or another. Following the presentation of the three models, I present a critique that sets forth the strengths and weaknesses of each. On the basis of this critique, I draw some general conclusions with respect to the future of international economic organization and the nature of future international relations in general.

The Sovereignty-at-Bay Model I label the first model sovereignty at bay, after the title of Raymond Vernon’s influential book on the multinational corporation.5 According to this view, increasing economic interdependence and technological advances in communication and transportation are making the nation state an anachronism. These economic and technological developments are said to have undermined the traditional economic rationale of the nation state. In the interest of world efficiency and domestic economic welfare, the nation state’s control over economic affairs will continually give way to the multinational corporation, to the Eurodollar market, and to other international institutions better suited to the economic needs of mankind. Perhaps the most forceful statement of the sovereignty-at-bay thesis is that of Harry Johnson—the paragon of economic liberalism. Analyzing the international economic problems of the 1970s, Johnson makes the following prediction:


500

In an important sense, the fundamental problem of the future is the conflict between the political forces of nationalism and the economic forces pressing for world integration. This conflict currently appears as one between the national government and the international corporation, in which the balance of power at least superficially appears to lie on the side of the national government. But in the longer run economic forces are likely to predominate over political, and may indeed come to do so before the end of this decade. Ultimately, a world federal government will appear as the only rational method for coping with the world’s economic problems.6 Though not all adherents of the sovereignty-at-bay thesis would go as far as Johnson, and an interdependent world economy is quite conceivable without unbridled scope for the activities of multinational corporations, most do regard the multinational corporation as the embodiment par excellence of the liberal ideal of an interdependent world economy. It has taken the integration of national economies beyond trade and money to the internationalization of production. For the first time in history, production, marketing, and investment are being organized on a global scale rather than in terms of isolated national economies. The multinational corporations are increasingly indifferent to national boundaries in making decisions with respect to markets, production, and sources of supply. The sovereignty-at-bay thesis argues that national economies have become enmeshed in a web of economic interdependence from which they cannot easily escape, and from which they derive great economic benefits. Through trade, monetary relations, and foreign investment, the destinies and well-being of societies have become too inexorably interwoven for these bonds to be severed. The costs of the ensuing inefficiencies in order to assert national autonomy or some other nationalistic goal would be too high. The citizenry, so this thesis contends, would not tolerate the sacrifices of domestic economic well-being that would be entailed if individual nation states sought to hamper unduly the successful operation of the international economy. Underlying this development, the liberal position argues, is a revolution in economic needs and expectations. Domestic economic goals have been elevated to a predominant position in the hierarchy of national goals. Full employment, regional development, and other economic welfare goals have become the primary concerns of political leadership. More importantly, these goals can only be achieved, this position argues, through participation in the world economy. No government, for example, would dare shut out the multinational corporations and thereby forgo employment, regional development, or other benefits these corporations bring into countries. In short, the rise of the welfare state and the increasing sensitivity of national governments to the rising economic expectations of their societies have made them dependent upon the benefits provided by a liberal world economic system. In essence, this argument runs, one must distinguish between the creation of the interdependent world economy and the consequences of its subsequent dynamics.7 Though the postwar world economy was primarily a creation of the United States, the system has since become essentially irreversible. The intermeshing of interests across


501

national boundaries and the recognized benefits of interdependence now cement the system together for the future. Therefore, even though the power of the United States and security concerns may be in relative decline, this does not portend a major transformation of the international economy and political system. The multinational corporation, for example, is now believed to be sufficiently strong to stand and survive on its own. The flexibility, mobility, and vast resources of the corporations give them an advantage in confrontations with nation states. A corporation always has the option of moving its production facilities elsewhere. If it does, the nation state is the loser in terms of employment, corporate resources, and access to world markets. Thus the multinationals are escaping the control of nation states, including that of their home (source) governments. They are emerging as sufficient powers in their own right to survive the changing context of international political relations. On the other hand, it is argued that the nation state has been placed in a dilemma it cannot hope to resolved.8 It is losing control over economic affairs to transnational actors like the multinational corporation.9 It cannot retain its traditional independence and sovereignty and simultaneously meet the expanding economic needs and desires of its populace. The efforts of nation states to enhance their security and power relative to others are held to be incompatible with an interdependent world economy that generates absolute gains for everyone. In response to the growing economic demands of its citizens, the nation state must adjust to the forces of economic rationality and efficiency. In the contemporary world, the costs of disrupting economic interdependence, of territorial conquest, and of risking nuclear warfare are believed to be far greater than any conceivable benefits. The calculus of benefits and risks has changed, and “the rational relationship between violence as a means of foreign policy and the ends of foreign policy has been destroyed by the possibility of all-out nuclear war.”10 In contrast to the nineteenth century, the cost of acquiring territory is viewed as having simply become too great. In the contemporary world, there is more to be gained through economic cooperation and an international division of labor than through strife and conflict. Thus, in the opinion of Saburo Okita, formerly president of the Japan Economic Research Center, the exercise of force for economic gain or to defend economic interests is an anachronism: We are living in a century when such military action is no longer viable. To build up military power just to protect overseas private property is rather absurd in terms of cost-benefit calculations. The best course for the Government in case of nationalization or seizure of overseas private Japanese assets is to compensate Japanese investors directly in Japan rather than to spend very large amounts of money to build up military strength.11 Just as the nuclear revolution in warfare now inhibits the exercise of military power, the revolution in economic relations now inhibits the national exercise of economic power by increasing the cost. Advances in transportation and communications have integrated national economies to the point where many believe it is too costly to threaten the severance of economic relations in order to achieve particular political and economic goals. Economically as well as militarily in the contemporary world, nations are said to


502

be mutually deterred from actions that would disrupt the interdependent economy. This mutual vulnerability of necessity limits and moderates the economic and political struggle among nation states. It provides the necessary minimum political order where the multinational corporations of all the major industrial powers can flourish and bring benefits to the whole of mankind. The sovereignty-at-bay view also envisages a major transformation of the relationships among developed and underdeveloped countries. The multinational corporations of the developed, industrial economies must not only produce in each other’s markets, but the locus of manufacturing industry will increasingly shift to underdeveloped countries.12 As the economies of developed countries become more service oriented, as their terms of trade for raw materials continue to deteriorate, and as their labor costs continue to rise, manufacturing will migrate to lesser-developed countries. United States firms already engage in extensive offshore production in Asia and Latin America. Western Europe has reached the limits of importing Mediterranean labor, which is the functional equivalent of foreign direct investment. Japan’s favorable wage structure and undervalued currency have eroded. With the end of the era of cheap energy and of favorable terms of trade for raw materials, the logic of industrial location favors the underdeveloped periphery. Increasingly, the multinational corporations of all industrial powers will follow the logic of this manufacturing revolution. Manufacturing, particularly of components and semiprocessed goods, will migrate to lesser-developed countries. This vision of the future has been portrayed most dramatically by Norman Macrae, in an issue of The Economist, who foresees a world of spreading affluence energized perhaps by “small transnational companies run in West Africa by London telecommuters who live in Honolulu.”13 New computer-based training methods and information systems will facilitate the rapid diffusion of skills, technologies, and industries to lesserdeveloped countries. The whole system will be connected by modern telecommunications and computers; the rich will concentrate on the knowledge-creating and knowledgeprocessing industries. More and more of the old manufacturing industries will move to the underdeveloped world. The entire West and Japan will be a service-oriented island in a labor-intensive global archipelago. Thus, whereas the telephone and jet aircraft facilitated the internationalization of production in the Northern Hemisphere, the contemporary revolution in communications and transportation will encompass the whole globe. “The logical and eventual development of this possibility,” according to management consultant John Diebold, “would be the end of nationality and national governments as we know them.”14 This sovereignty-at-bay world, then, is one of voluntary and cooperative relations among interdependent economies, the goal of which is to accelerate the economic growth and welfare of everyone. In this model, development of the poor is achieved through the transfer of capital, technology, and managerial know-how from the continually advancing developed lands to the lesser-developed nations; it is a world in which the tide of economic growth lifts all boats. In this liberal vision of the future, the multinational corporation, freed from the nation state, is the critical transmission belt of capital, ideas, and growth.


503

The Dependencia Model In contrast to the sovereignty-at-bay vision of the future is what may be characterized as the dependencia model.15 Although the analysis underlying the two approaches has much in common, the dependencia model challenges the partners-in-development motif of the sovereignty-at-bay model. Its Marxist conception is one of a hierarchical and exploitative world order. The sovereignty-at-bay model envisages a relatively benevolent system in which growth and wealth spread from the developed core to the lesser-developed periphery. In the dependencia model, on the other hand, the flow of wealth and benefits is seen as moving—via the same mechanisms—from the global, underdeveloped periphery to the centers of industrial financial power and decision. It is an exploitative system that produces affluent development for some and dependent underdevelopment for the majority of mankind. In effect, what is termed transnationalism by the sovereignty-at-bay advocates is considered imperialism by the Marxist proponents of the dependencia model. In the interdependent world economy of the dependencia model, the multinational corporation also reigns supreme. But the world created by these corporations is held to be far different from that envisaged by the sovereignty-at-bay school of thought. In the dependencia model the political and economic consequences of the multinational corporation are due to what Stephen Hymer has called the two laws of development. The law of increasing firm size, and the law of uneven development. The law of increasing firm size, Hymer argues, is the tendency since the Industrial Revolution for firms to increase in size “from the workshop to the factory to the national corporation to the multidivisional corporation and now to the multinational corporation.”16 The law of uneven development, he continues, is the tendency of the international economy to produce poverty as well as wealth, underdevelopment as well as development. Together, these two economic laws are producing the following consequence: …a regime of North Atlantic Multinational Corporations would tend to produce a hierarchical division of labor within the firm. It would tend to centralize highlevel decision-making occupations in a few key cities in the advanced countries, surrounded by a number of regional sub-capitals, and confine the rest of the world to lower levels of activity and income, i.e., to the status of towns and villages in a new Imperial system. Income, status, authority, and consumption patterns would radiate out from these centers along a declining curve and the existing pattern of inequality and dependency would be perpetrated. The pattern would be complex, just as the structure of the corporation is complex, but the basic relationship between different countries would be one of superior and subordinate, head office and branch office.17 In this hierarchical and exploitative world system, power and decision would be lodged in the urban financial and industrial cores of New York, London, Tokyo, etc. Here would be located the computers and data banks of the closely integrated global systems of


504

production and distribution; the main computer in the core would control subsidiary computers in the periphery. The higher functions of management, research and development, entrepreneurship, and finance would be located in these Northern metropolitan centers. “Lower” functions and labor-intensive manufacturing would be continuously diffused to the lesser-developed countries where are found cheap pliable labor, abundant raw materials, and an indifference to industrial pollution. This global division of labor between higher and lower economic functions would perpetuate the chasm between the affluent northern one-fifth of the globe and the destitute southern four-fifths of the globe. The argument of the dependencia thesis is that the economic dependence of the underdeveloped periphery upon the developed core is responsible for the impoverishment of the former. Development and underdevelopment are simultaneous processes; the developed countries have progressed and have grown rich through exploiting the poor and making them poorer. Lacking true autonomy and being economically dependent upon the developed countries, the underdeveloped countries have suffered because the developed have a veto over their development: By dependence we mean a situation in which the economy of certain countries is conditioned by the development and expansion of another economy to which the former is subjected. The relation of interdependence between two or more economies, and between these and world trade, assumes the form of dependence when some countries (the dominant ones) can expand and be self-sustaining, while other countries (the dependent ones) can do this only as a reflection of that expansion, which can have either a positive or negative effect on their immediate development.18 Though this particular quotation refers to trade relations, much of the dependence literature is addressed to the issue of foreign direct investment. In content, most of this literature is of a piece with traditional Marxist and radical theories of imperialism. Whether because of the falling rate of profit in capitalist economies or the attraction of superprofits abroad, multinational corporations are believed to exploit the underdeveloped countries. Thus, Paul Baran and Paul Sweezy see the multinationals necessarily impelled to invest in lesser-developed countries.19 Constantine Vaitsos has sought to document the superprofits available to American corporations in Latin America.20 The message conveyed by this literature is that the imperialism of free investment has replaced the imperialism of free trade in the contemporary world.

The Mercantilist Model A key element missing in both the sovereignty-at-bay and the dependencia models is the nation state. Both envisage a world organized and managed by powerful North American, European, and Japanese corporations. In the beneficial corporate order of the first model and the imperialist corporate order of the second, there is little room for nation states, save as servants of corporate power and ambition. In opposition to both these models,


505

therefore, the third model of the future—the mercantilist model—views the nation state and the interplay of national interests (as distinct from corporate interests) as the primary determinants of the future role of the world economy.21 According to this mercantilist view, the interdependent world economy, which has provided such a favorable environment for the multinational corporation, is coming to an end. In the wake of the relative decline of American power and of growing conflicts among the capitalist economies, a new international political order less favorable to the multinational corporation is coming into existence. Whether it is former President Nixon’s five-power world (US, USSR, China, the EEC, and Japan), a triangular world (US, USSR, and China), or some form of American-Soviet condominium, the emergent world order will be characterized by intense international economic competition for markets, investment outlets, and sources of raw materials. By mercantilism I mean the attempt of governments to manipulate economic arrangements in order to maximize their own interests, whether or not this is at the expense of others. These interests may be related to domestic concerns (full employment, price stability, etc.) or to foreign policy (security, independence, etc.). This use of the term mercantilism is far broader than its eighteenth-century association with a trade and balance-of-payments surplus. The essence of mercantilism, as the concept is used in this article, is the priority of national economic and political objectives over considerations of global economic efficiency. The mercantilist impulse can take many forms in the contemporary world: the desire for a balance-of-payments surplus; the export of unemployment, inflation, or both; the imposition of import and/or export controls; the expansion of world market shares; and the stimulation of advanced technology. In short, each nation will pursue economic policies that reflect domestic economic needs and external political ambitions without much concern for the effects of these policies on other countries or on the international economic system as a whole. The mercantilist position in effect reverses the argument of the liberals with respect to the nature and success of the interdependent world economy. In contrast to the liberal view that trade liberalization has fostered economic growth, the mercantilist thesis is that several decades of uninterrupted economic growth permitted interdependence. Growth, based in part on relatively cheap energy and other resources as well as on the diffusion of American technology abroad, facilitated the reintroduction of Japan into the world economy and the development of a closely linked Atlantic economy. Now both cheap energy and a technological gap, which were sources of rapid economic growth and global interdependence, have ceased to exist. International competition has intensified and has become disruptive precisely because the United States has lost much of its technological lead in products and industrial processes. As happened in Britain in the latter part of the nineteenth century, the United States no longer holds the monopoly position in advanced technologies. Its exports must now compete increasingly on the basis of price and a devalued dollar. As was also the case with Great Britain, the United States has lost the technological rents associated with its previous industrial superiority. This loss of industrial supremacy on the part of the dominant industrial power threatens to give rise to economic conflict between the rising and declining centers of industrial power.22 From the mercantilist perspective, the fundamental problem of modern international


506

society has been how to organize an industrial world economy. This issue arose with the spread of industrialism from Great Britain and the emergence of several competing capitalist economies in the latter part of the nineteenth century.23 In the decades prior to the First World War, the issue of how to organize a world economy composed of several competing industrial economies was at the heart of international politics. The resulting commercial and imperial struggle was a major factor in the subsequent outbreak of the First World War. The issue was never resolved during the interwar period. During the Second World War, the organization of the world economy was regarded, at least in the United States, as a central question for the postwar era. Would it be a universal liberal system or a fragmented system of regional blocs and preference arrangements? With the outbreak of the cold war and the undisputed hegemony of the United States over other capitalist economies, however, the issue faded into the background. Former President Nixon’s 15 August 1971 speech signaled to mercantilist writers that with the easing of the cold war the issue has once again moved to the fore. These mercantilist writers tend to fall into the two camps of malevolent and benign mercantilism. Both tend to believe the world economy is fragmenting into regional blocs. In the wake of the relative decline of American power, nation states will form regional economic alliances or blocs in order to advance their interests in opposition to other nation states. International trade, monetary arrangements, and investment will be increasingly interregional. This regionalization of economic relations will replace the present American emphasis on multilateral free trade, the international role of the dollar, and the reign of the American multinational corporation. Malevolent mercantilism believes regionalization will intensify international economic conflict.24 Each bloc centered on the large industrial powers—the United States, Western Europe, Japan, and the Soviet Union—will clash over markets, currency, and investment outlets. This would be a return to the lawlessness and beggar-thy-neighbor policies of the 1930s. Benign mercantilism, on the other hand, believes regional blocs would stabilize world economic relations.25 It believes that throughout modern history universalism and regionalism have been at odds. The rationale of regional blocs is that one can have simultaneously the benefits of greater scale and interdependence and minimal accompanying costs of economic and political interdependence. Though the material gains from a global division of labor and free trade could be greater, regionalism is held to provide security and protection against external economic and political forces over which the nation state, acting alone, has little influence or control. In short, the organization of the world economy into regional blocs could provide the basis for a secure and peaceful economic order.26 Benign mercantilism derives from the view of John Maynard Keynes and other Englishmen who were highly critical of an increasingly interdependent world economy. The loss of national self-sufficiency, this more benign view of mercantilism holds, is a source of economic-political insecurity and conflict.27 Liberalism, moreover, is detrimental to national cultural and political development. Therefore, this benign mercantilist position advocates a regionalization of the world economy as the appropriate middle road between a declining American-centered world economy and a global conflict


507

between the capitalist economies. An inevitable clash between industrial economies can be prevented through the carving out of regional spheres of influence and the exercise of mutual self-restraint among them. In the opinion of benign mercantilism, the thrust of much domestic and international economic policy, especially since the end of the First World War, has in fact been away from interdependence. Nations have placed a higher priority on domestic stability and policies of full employment than on the maintenance of international links. They have sought to exert national control over their monetary and other economic policies. This is what the Keynesian revolution and its emphasis on management of the domestic economy is said to be all about. The same desire for greater latitude in domestic policy underlies the increasing popularity today of flexible over fixed exchange rates and the movement toward regional blocs. Mercantilists point out that in many industrialized economies there is, in fact, a renewed questioning of whether the further benefits of trade liberalization and interdependence are worth the costs. Interdependence accentuates domestic economic adjustment problems as economic instabilities in one economy spill over into others. It causes labor dislocations, may accentuate inequalities of income distribution, and makes national planning more difficult. In short, according to these mercantilists, the world has reached the limits of interdependence and loss of national self-sufficiency.

A Critique of the Three Models In this section of the article, I evaluate the three models and draw from each what I consider to be important insights into the nature of contemporary international economic relations. This critique is not meant to cover all the points of each model but only those most directly relevant to this essay. Sovereignty at Bay Fundamentally, the sovereignty-at-bay thesis reduces to a question of interests and power: Who has the power to make the world economy serve its interests? This point may be best illustrated by considering the relationship of the multinational corporation and the nation state. In the writings I identified with the sovereignty-at-bay thesis, this contest is held to be most critical. On one side of this contest is the host nation state. Its primary source of power is its control over access to its territory, that is, access to its internal market, investment opportunities, and sources of raw material. On the other side is the corporation with its capital, technology, and access to world markets.28 Each has something the other wants. Each seeks to maximize its benefits and minimize its costs. The bargain they strike is dependent upon how much one wants what the other has to offer and how skillfully one or the other can exploit its respective advantages. In most cases, the issue is how the benefits and costs of foreign investment are to be divided between the foreign corporation and the host economy. The sovereignty-at-bay thesis assumes that the bargaining advantages are and always


508

will be on the side of the corporation. In contrast to the corporation’s vast resources and flexibility, the nation state has little with which to bargain. Most nation states lack the economies of scale, indigenous technological capabilities, or native entrepreneurship to free themselves from dependence upon American (or other) multinational corporations. According to this argument, the extent to which nation states reassert their sovereignty is dependent upon the economic price they are willing to pay, and it assumes that when confronted with this cost, they will retreat from nationalistic policies. In an age of rising economic expectations, the sovereignty-at-bay thesis rests on an important truth: A government is reluctant to assert its sovereignty and drive out the multinational corporations if this means a dramatic lowering of the standard of living, increasing unemployment, and the like. But in an age when the petroleum-producing states, through cooperation, have successfully turned the tables on the multinational corporations, it becomes obvious that the sovereignty-at-bay thesis also neglects the fact that the success of the multinational corporation has been dependent upon a favorable political order. As this order changes, so will the fortunes of the multinationals. This political order has been characterized by an absence of unity on the part of the economies that have been host to American and other corporations. The divisions between and within the host countries themselves, and the influence of the American government, left the host countries with little power to bargain effectively or to increase their relative benefits from foreign investments in their countries. Thus, in the case of Canada, the competition between the provinces and particularly between English Canada and Quebec greatly weakened Canada’s position vis-à-vis American investors. Similarly, nationalistic competition for investment has weakened attempts, such as the Andean Pact, that have tried to develop a common policy toward foreign corporations. But the importance of political factors in the overseas expansion of American corporations may be best illustrated by the case of Western Europe and Japan. American corporations coveted both the Japanese and Western European markets: they have been able to establish hundreds of subsidiaries in the latter but only a few in the former. The reason for this difference is largely political. Whereas the former has one central government controlling access to Japan’s internal market of 100 million population, six (now nine) political centers have controlled access to the European Common Market. By interposing itself between powerful American corporations and intensely competitive Japanese firms that desired American capital and technology, the Japanese government has been able to prevent the latter from making agreements not desired by the government. As a consequence, the Japanese home market has been protected as the almost exclusive domain of Japanese industry. American firms have had, therefore, a strong incentive to license their technology to the Japanese or to form corporate arrangements in which the American firms were no more than a minor partner. What the Japanese succeeded in doing was to break up the package of capital, technology, and entrepreneurship that foreign direct investment entails. The Japanese did not need the capital; they got the technology without managerial control by American corporations; entrepreneurship remained in the hands of Japanese. This Japanese example of untying the package and obtaining the technology, and in many cases the capital, required for development without loss of control has become an inspiration for economic nationalists in Latin America, Canada, and elsewhere.


509

In Western Europe, on the other hand, an American firm denied the right to establish a subsidiary in one Common Market country has had the option of trying another country and thereby still gaining access to the whole Market. Moreover, the strong desire of individual European countries for American investment has enabled American corporations to invest on very favorable terms. In certain cases, the firms have followed a divide-and-conquer strategy. Denied permission by President de Gaulle to invest in France, General Motors established in Belgium one of the largest automobile assembly plants in the Common Market. Through this route, the corporation gained access to the French market as well as to other European markets. In response to this situation, de Gaulle sought to obtain West German cooperation against American investment in EEC countries. Together these two most powerful of the Six could dictate a policy the others would be forced to accept. Through the instrumentality of the Franco-German Friendship Treaty of 1963, therefore, de Gaulle sought to form a Bonn-Paris axis directed against American hegemony in Western Europe. Although there was sentiment in West Germany favorable to taking measures to limit the rapidly growing role of American subsidiaries in EEC countries, the West German government refused to take any action that might weaken the American commitment to defend Western Europe. The United States government not only reminded the West Germans that a continued American military presence was dependent upon West German support of measures to lessen the American balance-of-payments deficit, but it also pressured West Germany to increase its military purchases from the United States and to avoid competitive arrangements with France. Largely as a result of these American pressures, the Friendship Treaty was, in effect, aborted. The first serious counteroffensive of the nation state against the multinational corporation collapsed. It is clear, however, that the outcome of this tale would have been altogether different if West Germany had desired greater military and economic independence from the United States. In short, the American corporate penetration of the European Common Market has been dependent upon the special security relationship of the United States and West Germany. One could extend this type of analysis for the whole of American overseas investment. American investment in the Middle East, Africa, Latin America, Canada, and elsewhere has benefited from America’s dominant position in the world. This position is now seriously challenged not only by the Soviet Union but by Japan, Western Europe, China, the Arabs, and Brazil in Latin America. Throughout these areas, economic nationalism is on the rise, threatening American investments and the income they bring to the United States. The thrust of this attack has been to break up the package of capital, technology, and management in order to acquire the first two without the third; the goal is greater local control through joint ventures, nationalization, and other policies. While the host countries are unlikely to “kill off” the American multinational corporations, they will increasingly make them serve local interests. This in turn will undoubtedly make direct investment abroad less attractive to American corporations. A reversal of fortunes has already been seen in the case of the oil multinationals. The significance of the offensive by the oil-producing states against the large international oil companies is not merely that the price of oil to the United States and to the rest of the world has risen but also that the United States may lose one of its most lucrative sources


510

of investment income. The oil crisis and Arab oil boycott which followed the 1973 ArabIsraeli war was a profound learning experience for Europe, Japan, and even the United States. The oil boycott and the behavior of the oil multinationals set into motion a series of events that cannot help but transform national attitudes and policies toward the oil multinationals. The sudden appreciation of how vulnerable governments were to the policies of the oil multinationals and how far their “sovereignty” had been compromised awakened them to the inherent dangers of overdependence on the corporations and their policies. The French and, to a lesser extent, the Japanese responses to this experience have received the most attention. But perhaps more noteworthy was the reaction of the West German government—after the United States the nation most committed to a liberal world economy. It was the West German representative at the February 1973 Washington conference of oil-consuming nations who demanded that the United States and Western Europe undertake “a joint analysis of the price policies, profits, and taxes of oilmultinationals.” While the proposal, which became part of the Washington Declaration, does not mean the demise of the oil multinationals, it does suggest that the policies of nation states will increasingly impinge on the freedom of action of these particular multinational corporations. This change in attitude toward the oil multinationals can be witnessed in the United States itself. The role of the companies as instruments of the Arab boycott has had a significant impact on American perceptions. Prior to that time, few probing questions about the oil multinationals had been raised in the press or in Congress. Other than a few “radicals,” few had challenged the fact that Exxon, Gulf, and other oil multinationals paid virtually no taxes to the United States government and that they acted as sovereign entities in their dealings with the oil-producing countries. When the tables were turned, however, and the oil companies became the instruments of the Arab boycott against the United States, then even their staunchest defenders began to raise questions about tax avoidance. More importantly, the United States government took into its own hands some of the task of negotiating with the oil-producing states. Thus, when the multinationals were perceived as no longer supportive of the national interests of the United States, there was a reassertion of national sovereignty. The case of oil and the oil multinationals is perhaps unique. Yet it does suggest that nation states have not lost their power or their will to act when they believe the multinational corporations are threatening their perceived national interests and sovereignty. The experience of the oil boycott and the role of the multinationals in carrying it out reveal the extent to which the operators and the success of these corporations have been dependent upon American power. With the relative decline of American power and the rise of governments hostile to American interests and policies, this case history at least raises the question of how the weakening of the Pax Americana will effect the status of other American multinational corporations throughout the world. Dependencia The weakness of the dependencia, or ultraimperialism, model is that it makes at least three unwarranted assumptions. In the first place, it assumes much greater common


511

interest among the noncommunist industrial powers—the United States, Western Europe, and Japan—than is actually the case. Secondly, it treats the peripheral states of Asia, Africa, Latin America, Canada, and the Middle East solely as objects of international economic and political relations. Neither assumption is true. As the first assumption is considered in more detail in the next section, let us consider the second for a moment. After nearly two centuries, the passivity of the periphery is now past. The Soviet challenge to the West and the divisions among the capitalist powers themselves have given the emerging elites in the periphery room for maneuver. These nationalist elites are no longer ignorant and pliable colonials. Within the periphery, there are coalescing centers of power that will weigh increasingly in the future world balance of power: China, Indonesia, India, Iran, Nigeria, Brazil, and some form of Arab oil power. Moreover, if properly organized and led, such centers of power in control over a vital resource, as the experience of the Organization of Petroleum Exporting Countries (OPEC) demonstrates, may reverse the tables and make the core dependent upon the periphery. For the moment at least, a perceptible shift appears to be taking place in the global balance of economic power from the owners of capital to the owners of natural resources.29 The third unwarranted assumption is that a quasi-Marxist theory of capitalist imperialism is applicable to the relationship of developed and lesser-developed economies today. Again, I illustrate my argument by considering the role of the multinational corporation in the lesser-developed countries, since its allegedly exploitative function is stressed by almost all dependencia theorists. The dependencia theory undoubtedly has a good case with respect to foreign direct investment in petroleum and other extractive industries. The oil, copper, and other multinationals have provided the noncommunist industrial world with a plentiful and relatively cheap supply of minerals and energy. The dramatic reversal of this situation by the oil-producing countries in 1973–74 and the steady rise of prices of other commodities support the contention that the producing countries were not getting the highest possible price and possibly not a just price for their nonrenewable resources. But what constitutes the just price for a natural endowment that was worthless until the multinationals found it is not an easy issue to resolve. With respect to foreign direct investment in manufacturing, the case is far more ambiguous. Even if technological rents are collected, does the foreign corporation bring more into the economy in terms of technology, capital, and access to world markets than it takes out in the form of earnings? The research of Canadian, Australian, and other economists, for example, suggests that it does. They find no differences in the corporate behavior of domestic and foreign firms; on the contrary, foreign firms are given higher marks in terms of export performance, industrial research and development, and other economic indicators.30 Nonetheless, it would be naive to suggest that no exploitation or severe distortions of host economies have taken place. On the other hand, it may not be unwarranted to suggest that a strong presumption exists for arguing that in terms of economic growth and industrial development, foreign direct investment in manufacturing is to the advantage of the host economy. A major cause of foreign direct investment is the sector-specific nature of knowledge and capital in the home economy.31 In order to prevent a fall in their rate of profits through


512

overinvesting at home or diversifying into unknown areas, American corporations frequently go abroad to guard against a lower rate of profit at home rather than because the superprofits abroad are attractive. Insofar as this is true, and there is sufficient evidence to warrant its plausibility, foreign direct investment benefits both the corporation and the host economy at a cost to other factors of production in the home economy. Thus, though the Marxists may be right in saying that there is an imperative for capitalism to go abroad, the effect is not to exploit but to benefit the recipient economy— a conclusion, by the way that Marx himself would have accepted.32 While it is true that, in general, lesser-developed countries are economically dependent upon developed countries, the conclusions to be drawn from this fact are not self-evident. Are the countries underdeveloped because they are dependent as dependencia theorists assume, or are they dependent because they are underdeveloped? China is underdeveloped, but it is not dependent upon any external power (though one could argue a historical case). As Benjamin Cohen has pointed out, the critical question is whether the poor are worse off economically because of this dependence.33 Does dependence upon the developed countries entail a new loss, or foreclose opportunities of greater benefit to the economy of the undeveloped country? While the opportunity to exploit may be there, is it exercised? These are empirical questions to which no general answers can be given. Whether foreign direct investment is exploitative or beneficial depends on the type of investment, its terms, and the policies of the recipient economy itself. The dependencia argument that foreign direct investment by multinational corporations preempts the emergence of an indigenous entrepreneurial middle class and creates a situation of technological dependence provides a clue to what is the central concern of dependence theory. Though most frequently couched solely in economic terms, the concepts of underdevelopment and dependence are more political than economic in nature. They involve an assessment of the political costs of foreign investment. They refer both to the internal political development of the recipient country and its external relations. As one of the better dependence theorists has put it, the problem “is not so much growth, i.e., expansion of a given socio-economic system, as it is ‘development,’ i.e., rapid and fundamental politico-socio-economic transformation.”34 In other words, foreign direct investment fosters an international division of labor that perpetuates underdevelopment and politico-economic dependencia. This distinction between growth and development is crucial.35 Economic growth is defined by most development economists simply as an increase in output or income per capita; it is essentially a positive and quantitative concept. The concepts of development and underdevelopment as used by dependence theorists are primarily normative and qualitative; they refer to structural changes internal to the lesser-developed economy and in external relations with the developed world. Dependencia theory really calls for a change in the current international division of labor between the core and the periphery of the international economy, in which the periphery is a supplier of raw materials and whose industries are branch plants of the core’s multinational corporations. Whatever its economic merits, the dependencia model will continue to generate opposition against the structure of the contemporary world economy and the multinational corporation throughout the underdeveloped periphery of the world economy. As these peripheral societies grow in power, one can anticipate that they will


513

undertake initiatives that attempt to lessen their dependence upon developed countries. Mercantilism It seems to me that mercantilists either ignore or ascribe too little significance to certain primary facts. Although the relative power of the United States has declined, the United States remains the dominant world economy. The scale, diversity, and dynamics of the American economy will continue to place the United States at the center of the international economic system. The universal desire for access to the huge American market, the inherent technological dynamism of the American economy, and America’s additional strength in both agriculture and resources—which Europe and Japan do not have—provide a cement sufficient to hold the world economy together and to keep the United States at its center.36 Furthermore, the United States can compensate for its loss of strength in one issue area by its continued strength in another. For example, the American economic position has indeed declined relative to Europe and Japan. Yet the continued dependence of Europe and Japan on the United States for their security provides the United States with a strong lever over the economic policies of each. Thus, the fundamental weakness of the mercantilist model is the absence of a convincing alternative to an American-centered world economy. Western Europe, the primary economic challenger to the United States, remains internally divided; it is as yet unable to develop common policies in such areas as industry and energy or with respect to economic and monetary union. It is merely a customs union with a common agricultural policy. Moreover, like Japan, it continues to be totally dependent upon the United States for its security. As long as both Europe and Japan lack an alternative to their military and economic dependence on the United States, the mercantilist world of regional blocs lacks credibility. The so-called energy crisis has affirmed this assessment. In the first place, the Arab oil boycott revealed the fragility of European unity. Threatened with the loss of vital supplies of Middle Eastern oil, every nation fended for itself. But subsequently, despite their reluctance, both Europe and Japan participated in the American-sponsored Washington energy conference. The American purpose in calling the conference was in part to reinforce its Middle Eastern diplomacy. But the purpose was also to reassert America’s influence over its allies and to forestall policies such as competitive currency depreciation, creation of new trade barriers, and bilateral deals that would tend to fragment the world economy. No doubt, too, as the French and others charge, the United States hoped to find a solution to the energy crisis that did not threaten the position of the American oil multinationals. Calling for cooperation from its European and Japanese allies, the United States reminded them that their security still rested on American goodwill. Moreover, in the event of a conflict over oil, America’s economic weapons were far superior. Thus chastened and reminded where power continued to rest, all but the French fell into line. For the time being at least, the United States demonstrated that it retained sufficient power to maintain intact an American-centered world economy. Yet sufficient tensions and conflicts of interests remain within this world economy to


514

prevent one from dismissing so quickly the mercantilist thesis. Undoubtedly, the interstate conflict that will be the most vexing is the growing demand and competition for raw materials, particularly petroleum.37 The loss of energy self-sufficiency by the United States and the growth in demand for petroleum and other raw materials have already shifted the terms of trade against developed economies, and commodity prices have become major factors in world inflation. In the longer term, these changes have put the industrial powers in competition for these limited resources. They are also competing for export markets in order to finance these vital imports and for the capital the oil-producing states now have to invest. Thus, whereas in the past America’s virtual control over the noncommunist world’s supply of petroleum was a source of unity, today the United States is struggling with other industrial powers to insure its own position in a highly competitive environment. In fact, one witnesses in the contemporary world the reemergence of the neoMalthusian and Social Darwinist fears that swept industrial society and were so disruptive in the latter part of the nineteenth century. A common factor in the several imperialisms that burst forth after 1880 and fragmented the world economy was the growing fear of the potential consequences of exclusion from resources and markets. With expanding populations and productive industries believed to be dependent on foreign sources of food and raw materials, the insecurity of European states was magnified by the loss of their former relative self-sufficiency. The paradox of an interdependent world economy is that it creates sources of insecurity and competition. The very dependence of one state on another and the necessity for access to external markets and sources of raw materials cause anxieties and suspicions that exacerbate international relations. The other reason for believing that there may be some validity in the mercantilist vision of the future is the weakening of political bonds between the United States, Western Europe, and Japan. During the height of the cold war, the foreign economic policies of these three countries were complementary. Potential conflicts over economic matters were subordinated to the necessity for political unity, against the Soviet Union and China. The United States encouraged export-led growth and accepted anti-American trade discrimination in order to enable Japan and Europe to rebuild their shattered economies. Reciprocally, Japan and Europe supported the international position of the dollar. Through foreign direct investment, American corporations were able to maintain their relative share of world markets. Neither the Europeans nor the Japanese challenged America’s dominant position with respect to the industrial world’s access to vital raw materials, particularly Middle Eastern petroleum. Until the early 1970s, the political benefits of this arrangement were regarded as outweighing the economic costs to each partner. With the movement toward détente and with the revival of the European and Japanese economies, however, the political benefits have receded in importance and the concern over costs has increased. As a consequence, the United States and its industrial partners now desire reforms of the world’s trading and monetary systems that would enable each to pursue its own particular set of interests and to limit that of the others. For example, the United States has proposed reforms of the trade and monetary systems that would limit the ability of the Europeans and the Japanese to run up huge trade surpluses. Europe and Japan, for their part, desire to


515

preserve this scope and to limit the privileges of the United States as world banker. Regardless of the outcome of the negotiations over the future of the international monetary system, one thing is certain: whatever privilege is retained by the dollar will not be sufficient to enable the United States to behave as it has in the past. Gone are the days when the United States could run an immense balance-of-payments deficit in order to support foreign commitments, to buy up foreign assets, and at the same time pursue a full employment policy at home. It will no longer be able to expand overseas at a relatively low cost to the American standard of living. Having already lost its technological superiority and technological rents, the United States will have to finance its economic and military position abroad through currency devaluation and a current account surplus. Thus the cost of any effort to maintain US political and economic hegemony will bear upon the American people themselves. The weight and popular appreciation of this cost will profoundly alter American attitudes toward America’s world role and toward its European and Japanese allies. These changes in political interests and perceptions cannot but help to push the world in a mercantilistic direction.

Implications for International Organization What then do these three models and their relative merits tell us about the future of international economic organizations? As a consequence of the relative decline of American power and of other developments treated in this article, there is little reason to believe that many new international institutions will be created, but it is likely that the nature and functioning of existing institutions will be profoundly altered. In a world of national states, international organizations tend to reflect the power and interests of the dominant states in the international system. From this perspective, the international organizations founded at the end of the Second World War reflected the then predominant states in the system. As the structure of the United Nations reflected the distribution of power between the United States and the Soviet Union, so the so-called Bretton Woods system and the institutions associated with it—the International Monetary Fund (IMF), the World Bank, and subsequently the General Agreement on Trade and Tariffs (GATT)—reflected the power and interests of the dominant world economy, the United States. In both cases, the relative decline of American power over the past several decades has led to profound modifications of these political and economic institutions. Thus, with the growth of Soviet power in the United Nations Security Council and of the so-called nonaligned bloc in the General Assembly, the United Nation’s role in American foreign policy and as an institution have been altered significantly. In terms of the major political issues of the world, the United Nations has moved from center stage to the sidelines. A similar transformation can be seen in the area of international economic institutions. This can be witnessed, for example, in the case of the IMF and the negotiations for the reform of the international monetary system which have taken place outside its aegis. The transformation of the IMF began in the late 1950s with the gradual weakening of the dollar as an international currency. After 1958 the American balance-of-payments deficit began to assume major proportions. The moderate deficits of the previous decade


516

became severe. A drain began on the large gold hoard the United States had accumulated before and during the Second World War. Between 1957 and 1963, US gold holdings fell from $22.8 billion to $15.5 billion and foreign dollar holdings (official and private) rose from $15.1 to $28.8 billion. By 1968, American gold holdings fell to $10.9 billion, and foreign dollar holdings rose to $31.5 billion. As Europeans and others began to turn dollars into gold, it became obvious that the United States could not continue to meet all gold claims. The immediate American response was to initiate numerous makeshift expedients—the gold pool currency swap arrangements, the General Arrangements to Borrow, etc.—to reinforce the position of the dollar. Additionally, the United States undertook unilateral measures such as the Interest Equalization Tax (1963), “voluntary” controls on the export of capital (1965), and, eventually, mandatory controls on foreign direct investment (1968) to stem the outflow of dollars. Despite these and other measures, monetary crises continued to mount throughout the 1960s. In response to these crises, demands mounted for a fundamental reform of the international monetary system. In the ensuing monetary negotiations, as in trade negotiations, the Western powers divided into three positions. On one side were ranged the United States and Great Britain. On the other stood France. In the middle was West Germany, which attempted to reconcile the Common Market and the Atlantic powers. Whereas the United States wanted a reform that would ensure the continued privileged position of the dollar, France under de Gaulle wanted a reform that would dethrone the dollar and thus would redistribute economic power in the West. This would allegedly be achieved if the world returned to what de Gaulle believed was the true measure of wealth and guarantor of political independence, namely, gold. A return to the gold standard would not only enhance the power of France, which had replenished its gold reserves, but the United States would have to expend real wealth in order to maintain and/or expand its hegemony. If other nations refused to accept any more dollars and demanded gold, the United States would be forced to bring its payments into balance and to liquidate its global economic and military position. In short, a shift from the dollar to gold as the world’s reserve currency would mean a retrenchment of American power in Europe, Asia, and round the globe. At the same time that the United States desired to maintain the privileged position of the dollar, the basic instability of the system was appreciated by all. An international monetary system and an expanding trade system that depended upon the deficits of the United States were prone to crisis. From the perspective of most countries, a return to gold was both politically and economically undesirable, however. In the late sixties, therefore, extensive IMF negotiations produced an international money called special drawing rights (SDRs). The United States had desired the SDRs to relieve the pressure on the dollar while preserving its ultimate reserve role. France wanted nothing less than the reimposition of monetary restraints on the United States. Between the two of them stood West Germany and its desire to hold together the European and Atlantic powers. Due largely to German initiatives, a compromise solution was finally reached, which gave the Americans their SDRs in exchange for greater European voting power in the International Monetary Fund. Thus, while the IMF would have the power to “issue” SDRs as an international


517

reserve on a limited scale, Europe (if it were united) could exercise a veto over American policy in the IMF. In short, the internal structure and functioning of the IMF was reconstituted to reflect the redistribution of world economic and monetary power. The United States no longer ran the organization. Control over it was now shared by the European powers. Similarly, one can anticipate that the immense growth of Arab monetary balances will lead to a further internal transformation of the IMF. By one method or other, this redistribution of monetary power will be given an institutional form. In the areas of trade and investment, the continuing redistribution of power among nation states will find a response in the nature and functioning of international economic organizations. In trade this has already begun to happen, as the United States and other industrial nations ponder the future of the GATT. Perhaps the German initiative at the Washington energy conference in calling for an international investigation of the oil multinationals presages what many have long advocated—a GATT for investment. If so, it too will reflect the changes that have taken place in the world’s distribution of economic and industrial power.

Conclusion In conclusion, what does this redistribution of world power imply for the future of the interdependent world economy? Today, the liberal world economy is challenged by powerful groups (especially organized labor) within the dominant economy; the dominant economy itself is in relative decline. With the decline of the dominant economic power, the world economy may be following the pattern of the latter part of the nineteenth century and of the 1930s and may be fragmenting into regional trading and monetary blocs. This would be prevented, of course, if the United States, as it is presently trying to do, were to reassert its waning hegemony over Western Europe, Japan, and the rest of the noncommunist world economy. In the wake of the decline of American power and the erosion of the political base upon which the world economy has rested, the question arises whether the wisest policy for the United States is to attempt to reassert its dominance. May not this effort in the areas of trade, money, investment, and energy exacerbate the conflicts between the United States, Western Europe, and Japan? If so, a future that could be characterized increasingly by benign mercantilism could well be transformed into its more malevolent relative. If this were to happen, the United States and its allies would be the losers. This admonition suggests that the United States should accept a greater regionalization of the world economy than it has been wont to accept in the past. It implies greater representation and voice for other nations and regional blocs in international economic organizations. While such a policy of retrenchment would no doubt harm the interests of American corporations and other sectors of the American economy, the attempt to hold on to rather than adjust to the shifting balance of world power could be even more costly for the United States in the long run. In a world economy composed of regional blocs and centers of power, economic bargaining and competition would predominate. Through the exercise of economic power


518

and various trade-offs, each center of the world economy would seek to shift the costs and benefits of economic interdependence to its own advantage. Trade, monetary, and investment relations would be the consequence of negotiations as nation states and regional blocs sought to increase the benefits of interdependence and to decrease the costs. This in fact has been the direction of the evolution of the international economy, from a liberal to a negotiated system, since the rise of large and rival economic entities in the latter part of the nineteenth century. Therefore, debate and policy planning today should not focus on economic independence or dependence but on the nature and consequences of economic interdependence. Economic interdependence may take many forms; it may affect the welfare of nations in very different ways. Some will emphasize security; others, efficiency, low rates of inflation, or full employment. The question of how these benefits and costs will be distributed is at the heart of the increasingly mercantilistic policies of nation states in the contemporary world.

Notes Robert Gilpin is a professor of politics and international affairs at Princeton University in Princeton, New Jersey. 1. Edward Hallet Carr, The Twenty Years’ Crisis 1919–1939 (London: Macmillan and Co., 1951), p. 117. 2. This theme is developed in Robert Gilpin, “The Politics of Transnational Economic Relations,” International Organization 25 (Summer 1971): 398–419. This article was part of a special issue of the journal, entitled “Transnational Relations and World Politics,” which was edited by Robert O.Keohane and Joseph S.Nye, Jr., and was subsequently published under the same title as a book by Harvard University Press in 1972. 3. Ernest Preeg, ‘Traders and Diplomats (Washington, D.C.: The Brookings Institution, 1970), p. 1. 4. C.Fred Bergsten, “The New Economics and U.S. Foreign Policy,” Foreign Affairs 50 (January 1972): 199–222. 5. Raymond Vernon, Sovereignty at Bay (New York: Basic Books, 1971). 6. Harry G.Johnson, International Economic Questions Facing Britain, the United States, and Canada in the 70’s, British-North American Research Association, June 1970, p. 24. 7. Samuel Huntington, “Transnational Organizations in World Politics,” World Politics 25 (April 1973): 361. 8. Edward Morse, “Crisis Diplomacy, Interdependence, and the Politics of International Economic Relations,” World Politics 24, supplement (Spring 1972): 123–50. 9. Keohane and Nye. 10. Hans Morgenthau, “Western Values and Total War,” Commentary, October 1961, p. 280. 11. Quoted in New York Times Magazine, 29 October 1972, p. 58. 12. John Diebold, “Multinational Corporations—Why be Scared of Them?,” Foreign Policy, no.12 (Fall 1973): 79–95. 13. “The Future of International Business,” The Economist, 22 January 1972. 14. Diebold, p. 87. 15. The literature on dependencia, or underdevelopment, has now become legend. One of the


16.

17. 18. 19. 20.

21.

22.

23.

24. 25. 26. 27. 28. 29. 30. 31.

32.

519

better statements of this thesis is Osvaldo Sunkel, “Big Business and ‘Dependencia’: A Latin American View,” Foreign Affairs 50 (April 1972): 517–31. For an excellent and critical view of the dependencia thesis, see Benjamin J.Cohen, The Question of Imperialism—The Political Economy of Dominance and Dependence (New York: Basic Books, 1973), chapter 6. “The Multinational Corporation and the Law of Uneven Development,” in Economics and World Order—From the 1970’s to the 7990’s, ed. Jagdish Bhagwati (New York: The Macmillan Co., 1972), p. 113 and passim. Ibid., p. 114. Quoted in Cohen, pp. 190–91. Monopoly Capital—An Essay on the American Economic and Social Order (New York: Monthly Review Press, 1966). Constantine Vaitsos, “Transfer of Resources and Preservation of Monopoly Rents,” Economic Development Report No. 168, Development Advisory Service, Harvard University, 1970. (Mimeographed.) See, for example, David Calleo and Benjamin Rowland, America and the World Political Economy (Bloomington, Ind.: Indiana University Press, 1973). Mercantilism is also the real theme of Ernest Mandel’s Europe vs. Americ—Contradictions of Imperialism (New York: Monthly Review Press, 1970). Vietnam-generated inflation was also a factor in the decline of American competitiveness in the late 1960s. But mercantilists and others (such as Richard Nelson and Michael Boretsky) respond that it is precisely because the US has lost much of its technological lead in products and industrial processes that pure competition has become so important. For an analysis of this debate, see Philip Boffey, “Technology and World Trade: Is There Cause for Alarm?,” Science, 2 April 1971. I have benefited very much from the as yet unpublished writings of Kendall Myers on this subject. Myers’s manuscript entitled “Appeasement and Nazi Germany—Regional Blocs or Universalism” was the basis of a seminar held at the Lehrman Institute in New York. See also the reflections of Simon Kuznets, Modern Economy Growth (New Haven, Conn.: Yale University Press, 1966). The issue, of course, is fundamental to the radical and Marxist critique of capitalism. Ernest Mandel, in his Europe vs. America, is more malevolent mercantilist than Marxist in his argument. Calleo and Rowland. For a recent analysis of this issue, see Ernest Preeg, Economic Blocs and U.S. Foreign Policy (Washington, D.C.: National Planning Association, 1974). This paradox is analyzed by Eugene Staley, World Economy in Transition (New York: Council on Foreign Relations, 1939), chapter 6, especially p. 15. For an excellent examination of this relationship, see Huntington. See Fred Bergsten, “The Threat from the Third World,” Foreign Policy, no. 11 (Summer 1973): 102–24. See for example, A.E.Safarian, Foreign Ownership of Canadian Industry (Toronto: University of Toronto Press, 1973). This point is developed in US Congress, Senate Committee on Labor and Public Welfare, The Multinational Corporation and the National Interest (report prepared for the Committee), 93rd Cong., 1st sess., 1973, Committee print. Karl Marx, “The Future Results of British Rule in India,” in Karl Marx on Colonialism and Modernization, ed. Shlomo Avineri (Garden City, N.Y.: Doubleday, 1968), pp. 125–31.


520

33. Cohen, chapter 6. 34. This distinction is developed by Keith Griffin, Underdevelopment in Spanish America (Cambridge, Mass.: The M.l.T.Press, 1969), p. 117. 35. For a more detailed analysis of the distinction, see J.D.Could, Economic Growth in History (London: Methuen and Co., 1972), chapter 1. 36. A forceful statement of this position is Raymond Vernon’s “Rogue Elephant in the Forest: An Appraisal of Transatlantic Relations,” Foreign Affairs 51 (April 1973): 573–87. 37. See Helmut Schmidt, “The Struggle for the Global Product,” Foreign Affairs 52 (April 1974): 437–51.


521

60 Supranationals and the State* Susan Strange

* Source: J.Hall, ed., States in History, 1986, ch. 11, pp.289–305.

For most of the last century people of many nationalities and a variety of political persuasions have been wishing, hoping, predicting—and sometimes even working—for the withering away of the state. Yet the state remains. It still looms large on the horizon of every individual life. It does not look like disappearing any more quickly in the next 50 years than in the last 50. And while the world economy has become ever more closely knit together, and while cheaper and easier transport and communications have created the beginnings of what one might reasonably call a world society, the international political system with which students of international relations have been mainly concerned has remained fundamentally unchanged. It still consists of a society of territorial states whose governments are united in their proclaimed deference to the principles of sovereignty and autonomy. True, its membership has grown very substantially. There were hardly more than two score states at the beginning of the century. There could be four times as many before its end. Yet the principles of the system have changed only very little. The exercise of political authority by its members is still for the most part bounded by territorial limits (or to be more precise, in view of the extension of authority to exclusive zones at sea, by cartographical limits). And the exercise of that authority requires the acquiescence (if not the formal recognition) of other similarly recognized political authorities. Between these authorities, the system allows both for conflict and for cooperation. Despite nearly a century of effort to limit the one and to nurture and extend the other, these attempts have not been nearly as successful as the idealists and internationalists have hoped. The essentially anarchic, quarrelsome society of states has not fundamentally changed. The argument presented in this chapter is an attempt to offer an explanation (or perhaps a political explanation) for the persistence of the state as the—or at least as a—major entity in the international political system, and for its continued enjoyment (in most cases) of predominant political authority within its own territorial limits. It will suggest and try to explain a paradox. For while the authority of the state has been under attack from two different directions, at the hands of two different kinds of supranational or transnational entities, at the same time both of these supranational entities have also been busily contributing to the power and authority of the state. The situation I propose to describe can best be understood by analogy. In a large and powerful river, where the river is broad and the water is shallow, sandbanks appear. At the upstream end, the river will tend to erode the sandbanks. At the downstream end, it will tend to build them up. Occasionally, some sandbanks will be washed away. New

Supranationals and the State

522

ones will appear. But the simultaneous process of erosion and deposition will go on all the time. That, so it seems to me, is just what has happened to international society, to what Bruce Miller calls the world of states; and what is still happening at the hands of the two major kinds of supranational entity that we observe in the world today, to wit, international organizations and transnational corporations. Both are building up the sandbank of state power at the same time as, in other ways, they are eroding it. It has to be added that the deposition of new powers with the state—like the silt or sand to the sandbanks—has been substantially aided by two historical phases that have directly or indirectly affected all the states in the society of states. Or rather, one should say four phases since each—both wars and slumps—have occurred twice in living memory: there have been two world wars—and, indeed a long-drawn-out ‘cold war’ with some especially acute periods—and two world economic depressions. All these systemic changes—like changes in the state of the river itself as when a spring flood might wash down an extra quantity of silt from the hills above—have tended vastly to increase the powers of the state, regardless of the political ideology professed by its government. This is a commonsensical point often lost sight of by students of comparative politics or comparative sociology. They become so interested in the many queer ways in which societies and political systems differ from one another that they sometimes overlook the ways in which their behaviour is strikingly similar. (In truth, they should not be called comparative studies at all but contrastive, for they concentrate on the contrasts, not the comparisons.) It is a tendency easily corrected by attention to international studies, and particularly to world history, both political, economic and societal. In our own times, for instance, Franklin Roosevelt was elected to balance the budget and restrain federal power. Yet the increase in federal spending and taxation and the proliferation of federal agencies had never before been so great as under the New Deal. At the same time, across the Atlantic, the increase of state power under the National Socialist government of Germany as compared with the Weimar Republic was powered with the same economic forces of slow economic growth, credit scarcity and consequent unemployment that occasioned the New Deal. And more recently, President Reagan when elected promised to reduce the power of the federal government and to turn the rising tide of state spending. Yet he has presided over fiscal deficits that broke all previous records, and (despite some peripheral deregulation) over an increase in the powers of some key agencies like the Federal Reserve Board and the Comptroller of the Currency and over more protective intervention on behalf of uncompetitive American industries than had been seen for many years. Nor is he alone. Notwithstanding the wave of deregulation in certain limited sectors started by the United States, it seems that the 1980s have witnessed an extension of the powers of all states over everyday life—over prices, wages, imports, credit, educational systems and factory conditions for instance— as states have struggled to respond to the vicissitudes of international debt, shrunken markets and rising unemployment. As for wars, it must be obvious that the powers assumed by governments at war over the life and death of individuals, over how much people eat (and drink), over where and how they work, over where they may and may not go and what they may and may not say were all vastly enhanced in the First World War, never entirely abandoned after it. They


523

were once again enhanced during the Second World War. Yet while wars and slumps tend to reinforce state power in a cyclical manner, international cooperation has tended to increase both secularly and cyclically. It tends to increase steadily as technical imperatives make adaptive cooperation (as in the organization of transborder railways, air transport, telegraph or radio systems) more necessary. And it tends to increase in a mildly cyclical pattern in the aftermath both of wars and of world economic depressions. After the First World War, the League of Nations, the International Labour Office and the Permanent Court of International Justice were established, each breaking new ground in the scope, the size and the aspirations of international institutions to supplant or restrain the power of the state in the world system. After the Second World War came the United Nations organization extending its concern for the preservation of peace and international order to a wide range of economic and social issues, including the preservation of human rights, the ending of colonialism and the assertion of common standards in all sorts of areas of state policy. Though it was far less marked, we can also see a similar reaction to world depressions. In the mid-1930s, the major states in the international monetary system laid a new basis of restored order in the Tripartite Agreement of 1936. Too late, the Bruce Commission three years later recommended that the League of Nations set up permanent machinery for international economic cooperation. Today, too, in a similar reaction to economic disorder and uncertainty, there are signs of a shift in opinion towards genuine monetary reform—as distinct from the phoney ‘reforms’ of the 1970s—and towards some strengthening of the machinery for managing the world economy. Much more powerful, and certainly more apparent, has been the secular trend towards the internationalization of markets and of production. The growth of transnational corporations—the misnamed ‘multinationals’—has so undermined the coincidence that used to prevail between state borders and economy borders that everyone today is aware of the effects this has had on the capacity of states to order and manage their national economies. The globalization of business—and the powerful response of enterprises, public as well as private, to it constitute the second great eroding force on the power of the state. It is if anything intensified by the sharpened competition between enterprises for shares in a world market shrunken by world depression. No wonder then that the onset of the second world economic depression has tended to revive academic interest in the perennial question of the twentieth century: is the state gaining power or losing it? The debate has intensified of late not only among sociologists but also in other branches of social science. Those economists still concerned with the real world (and not lost like so many of their fellows in a world of mathematical abstractions) are currently engaged in a series of discussions revolving around the relation of political authority to market forces. The whole question of fixed and floating exchange rates, for instance, which was thought by the late 1970s to have been a closed issue, has been reopened. There is discussion of the costs and benefits of protection by various means. And there are great debates ranging from the deeply philosophical to the abstruse and technically complex about the role of money and the machinery for its management and about the management of the credit-creating institutions, both banks and near-banks. Political scientists, too, are showing a rather marked lack of agreement among


524

themselves as to whether the state is or is not growing in power. The central theme chosen for the triennial world congress of political scientists in Paris in July 1985 was The changing state and its interaction with national and international society’. But predictably, whilst some discounted the extent of fundamental change, and others insisted upon it, the great majority were uncertain and hedged their bets. As one of them put it: There is no clearcut answer to the question of whether the state’s competence is widening or withering…the foreseeable future is likely to be one of states limping along, muddling through as it were, buffeted by internal and external forces that leave the norms, habits and practices relevant to their capacities for cooperation hovering endlessly on the brink of transformation yet managing to persist through time.1 It has to be said that the current uncertainty and confusion of the political scientist is evidently greater in the United States than in Europe, and greater too among the international relations people who have been interested in international organizations than in those who have been interested in, for example, other branches of the subject such as strategic studies or foreign policy analysis. The reason in both cases is that the exaggerated expectations of a few years ago have now been replaced by chilling disillusion. The Americans in the 1950s and 1960s exaggerated the power of the US government to solve the problems of the domestic economy and society and to dominate and shape the international system. The so-called behavioural revolution in the social sciences swept through the universities like an epidemic, replacing attention to the state and the realists’ ‘world of states’ with attention to systems and to the quantitative analysis of systems. The violence of the reaction that set in in recent years was one which Europeans and many other non-Americans found somewhat surprising. Similarly, enthusiastic students of international organization were once apt to exaggerate the changes in state behaviour likely to be brought about when governments agreed—often for mixed and contradictory reasons—to set up new intergovernmental institutions like the United Nations, the European Economic Community or the Organization of Petroleum Exporting Countries. Now that the limitations of each of these have been so often demonstrated that they are undeniable, there is growing disarray to be seen among the theorists who seek to find general theories to explain the growth and the decline of international Regimes’. All this is very different from the prevailing academic opinion 50 years ago—and perhaps suggests that ordinary people should exercise some caution in paying too serious attention to academic opinion. Then, Harold Laski was writing scornfully about the pretensions of the state, its boundaries ‘instinct with difficulty rather than implicit with meaning’.2 He poured scorn on its moral pretensions in the light of immoral behaviour and revelled in his own acumen in seeing the social skeleton and class structures beneath the flesh of political constitutions. His thoughts on international relations outlined in A Theory of Sovereignty were not too different from those of a contemporary, David Mitrany, still remembered as the father of functionalism—the notion that the habits of cooperation developed by technocrats in adaptive international organizations would slowly transform the behaviour of states towards each other in ‘more political’ matters. Yet even while they were writing in the


525

mid-1930s there was hardly a state—China was the major exception—in which the hold of government over society and economy was not stronger than it had been before the world depression. My purpose, however, being to shed some small ray of light on the current confusion, the next question is how best to demonstrate my contention that states are being sustained by two prevailing groups of supranational entities even while they are also being undermined by them. One way would be simply to give a selection of illustrative evidence of the ways in which this is being done. That seems to be unsatisfactory because it may be said that circumstantial evidence can always be selected that will back a particular argument. So, though examples will be useful, they will perhaps be more convincing if set in a broader historical perspective of change in the international political economy. By this means it is easier to identify the shifting bases of the power and authority of states—which are both social and economic—and thus the ability of both political institutions and economic enterprises to either reinforce and sustain or to challenge and diminish that power and authority. Broadly speaking, it may be said that, looking back over the past millennium, two changes relevant to the state have taken place. There has been a change in the basis of state wealth, and therefore power, from land and labour to capital and knowledge— including in ‘knowledge’ not only technology but also other kinds of information. And in the last century or less, there has been a change in structural power which from being contained within the state now spills out far beyond its frontiers, returning in some respects to an earlier pattern in which supranational institutions—the medieval church in Europe, for example—enjoyed certain kinds of structural power over secular governments whose authority was territorially bounded. Let us take first the change from land and labour as the basis of state power to capital and knowledge. The argument is simple. The exercise of political authority consumes resources. Though by imposing order it may make possible or easier the production of resources by others, the rulers and their servants, including generals and judges, are not themselves productive. Authority therefore has to tax the economy; it has to extract resources from society in order to support itself. In a pre-industrial economy with limited transport systems and therefore limited opportunities for trade, the major source of wealth was agriculture, with hunting perhaps contributing some marginal resources.3 The land is necessary for the production of food and other crops; and manpower labour—is necessary both to cultivate the land and to provide the means of military force to acquire more land. All the great empires of the ancient world, therefore, were built by the acquisition of more land by military conquest. Once built, great empires could exploit the manpower of their dominions to supplement their own resources of military manpower with auxiliaries, whether voluntarily recruited or compulsorily conscripted. The state in each case also enhanced its ability to tax the products of the land and therefore to increase the resources it could dispose of for political purposes, whether this was building pyramids or road systems or organizing elaborate games and entertainments. With certain differences, the economic basis of political power was still the same in medieval Europe as it had been in the ancient world. Weaker states and a more insecure political system meant a sharing of the power to tax the resources produced by land and labour. Security systems became more localized. There was no Pax Romana. Feudal


526

lords therefore exacted resources from the land which gave them political power. So too did the church and the great religious orders—the first capitalist enterprises whose ability to accumulate capital and invest it in long-term land improvement was not limited by the mortality of the individual. Between rulers, wars were fought for land and the rulers who could muster the largest military forces were those with command and authority over the largest feudal pyramid of landed barons and their vassals. And besides conquest there were the other means of acquiring land. As the old farming saying has it, there are three ways to acquire land: patrimony, matrimony and parsimony. Between rulers, therefore, the pursuit of daughters bringing substantial dowries of taxable land was a recognized form of interstate rivalry. And disputes over patrimonial rights and succession were a common cause of conflict and insecurity within and between states up until the eighteenth century. In this pre-industrial economy, only a very few states were able to exploit the possibilities of commercial exchange successfully enough to find in trade a secure basis of political power. Trade had to be highly profitable to provide the resources sufficient to stand up to states built on resources of land and labour. For a time the city-states of Italy aided by more advanced technology were able to exact enough tax from their traders and artisans to pay condottieri for their defence. But as the invasion of Italy by Charles VIII of France in 1494 showed only too painfully, no alliance of city-states was strong or firm enough to resist such a large territorial state coming from outside. Only Venice, sustained by sea power, proved less vulnerable. And it was only other ports with similar sea power and similar opportunities for accumulation through sea trade—towns such as Hamburg, Antwerp, Genoa, and later London—which were able to acquire power in the world of states by exploiting trade as well as land and labour. The nineteenth century can be seen as the period of balance—halfway between the old political economy which gave power to states with command over land and labour and the new political economy of today and of the future which gives power to states with command over capital and technology. In the nineteenth century and more or less until the First World War, the majority of the working population even in the most advanced industrialized countries still worked on the land. The productivity of agriculture was therefore still a major determinant of wealth which an efficient state machine could extract by taxation. And since mercenaries (like the condottieri before them) had been shown to be no match for armies imbued with nationalist fervour, rightly preferring to run away and live to fight another day, the European states now depend on their manpower to provide them with large conscript armies. But already technology in weaponry was proving more decisive. The French poilus overrun at Sedan in 1870 by a better equipped Prussian army were mown down again at Verdun in 1916, once again outclassed by more advanced military technology (tanks, poison gas, Big Berthas). And in the world beyond Europe, the domination of Europeans over non-Europeans was achieved not only through their advanced technology and therefore ability to conquer and rule, but also through their control of credit. Wealth gave them the ability to grant or to withhold the capital which even then was vital to other states with undeveloped economies which they hoped to modernize and make more productive. Tsarist Russia, for example, with more men and more land than France, was yet the junior partner through her dependence on French credit and French (and other foreign) technology. The


527

Ottoman Empire, similarly, was vulnerable to European states less because of military weakness—it performed well enough in the Dardanelles encounter after all—than because of its debts and its dependence on foreign capital. Other debtors like Egypt and Morocco actually lost their autonomy through dependence on foreign loans which they could neither repay nor service. And across the Atlantic, the domination exercised by the United States over Central America and the Caribbean at the turn of the century was only partly due to the superior naval power exhibited in the Spanish American war; it was also the result of the ability to lend capital to indigent or incompetent governments. If ‘parsimony’ is taken as a synonym for capital accumulation, it has certainly replaced matrimony as a useful tool of interstate relations and as a means of increasing power and influence over others. As late as the post-1945 period, the British Dominions, as they were then still called, were still looking to Britain to supply capital for development. Only when those hopes proved empty did the old imperial links weaken and break. By then, too, it was obvious to all that the one state still able to supply capital to others was the United States of America. It was the only state to come out of the war with an economy vastly stronger and more productive than it had had at the beginning, apart from India which was divided, poor and populous, and the neutrals like Sweden and Switzerland which were small and cautious. American domination of Western Europe, Latin America and the whole world market economy rested predominantly on its power to offer long-term loans or investments to countries whose appetite for imports to build, or rebuild, their economies vastly exceeded their capacity to earn foreign exchange by producing for export. By 1945, however, another major change had taken place in the relative importance of the four factors of production—land, labour, capital and technology—as bases for the power of states. The United States was ahead of all others in producing an atom bomb. If the Russians, the Germans or the Japanese or the Chinese had got there first, the outcome of the war and the post-war world would have been very different. And it was a military advantage achieved by recruiting a small, multinational band of rather pacific-minded scientists, not by any of the conventional sources of military might. Hiroshima and Nagasaki, then, were a turning point in the evolution of states. For these terrible explosions left the United States with a brief monopoly of atomic weapons of destruction sufficiently threatening to balance the power of the only other victorious state still able to keep large numbers of what now became ‘conventional’ forces under arms. The opportunity costs of doing so must have slowed Soviet economic recovery and consequently widened the gap still further between American and Soviet GNP. By the end of the 1950s, and as demonstrated in the Cuban crisis of 1962, the monopoly had effectively become a duopoly, as the Soviet Union matched the American power not only to make nuclear weapons but to deliver them with intercontinental ballistic missiles. The significance of this for the world of states was clear enough: armies, as traditionally understood, had become less important in the competition of states than technical superiority in the development of new destructive weapons. For thousands of years, armies had been necessary to conquer and to occupy territory—in order to acquire land. Now land was not so necessary to power, and the power to destroy cities in seconds was a more powerful threat than that of invasion, occupation and annexation. States, including the superpowers, still behaved as though land were the chief base of state


528

power. John Foster Dulles’s policy of containment aimed to confine the USSR within its territorial limits, and the Soviet Union in 1956, in 1968 and in 1981 fiercely resisted any internal challenges from Hungarians, Czechs and Poles that would have weakened its occupation of East European territory. But this was more because, as always, internal social cohesion was a necessary condition for the exercise of external power than because a territorial empire was a source of power and wealth. By 1984, with Ronald Reagan’s announcement of a Strategic Defence Initiative (SDI), the shift of interstate competition from competition for land to competition for technology became yet more pronounced. Whoever could first devise a defensive system against the delivery of nuclear warheads would enjoy a monopoly of immunity. The balance of nuclear power aptly described by the acronym MAD (Mutual Assured Destruction) would be gone, and the threat to survival posed by the proliferation of nuclear weapons among an ever-growing number of middle and even small states (like Israel or Spain) would have been averted. The dangers and uncertainties opening up before us as a result of SDI becoming more than a pipedream are matters of continuing debate, not least for the Europeans whose long experience of interstate conflicts have taught them that peace is often just as much in jeopardy when a major state is materially weakened as when one suddenly perceives itself as materially stronger. The shifting of interstate competition to technology has a very direct bearing on the question before us: how do supranational entities add to or subtract from the power of states? The point is that neither international organizations nor transnational corporations occupy land. But both do dispose of technology, and that, together with their disposition over capital, is the major reason why in any study of states, the relation of a state to either international organizations or transnational corporations must make up an important part of the analysis. It would not be so necessary to emphasize the point by setting it in longer historical perspective if the conventional wisdom in the study of international relations had not been so slow to take the point. Textbooks still take an essentially territorial view of the bases of state power and the focus of interstate competition. Yet as long ago as 1968, Robert Gilpin, observing the importance attached by General de Gaulle to the organization and application of science in France, had written: Just as the nation-state with its superior organization and resources, supplanted the city-state and feudal society after 1500, so today the organization of political life is being transformed by the role of science in human affairs. Increasingly, international politics will be dominated by large-scale political entities with substantial scientific and technological resources at their disposal.4 Gilpin’s argument was that only the two superpowers still had the scientific and technological resources to enjoy full economic, military and political independence. Western Europe was caught between them and faced three dangers: that US corporations would pre-empt decision-making power over European industrial development; that the United States would control all the defence-related industries; and the there would be a continued brain drain westwards across the Atlantic which would rob European countries of their capacity for autonomous innovation. He ended by quoting a British European


529

Movement report on European cooperation published in 1965 which wrote prophetically: The conclusion forced upon us is that the European countries must pool their efforts if they wish to play a continuing part in world economic development. Otherwise, it will not be many years before they are so far behind the United States and the Soviet Union in advanced industrial fields that relatively they will have fallen to the level of underdeveloped countries.5 Just 20 years later, Dr W.Dekker, president of Philips, was repeating the warning. The Untergang des Abendlander predicted by Spengler would after all be their fate if European countries failed to realize that the only things they could achieve in the field of industrial innovation were the things they could achieve together. Referring to the world market for micro-electronics, he said: If we wait too long for a European initiative, the result could be that we, in Europe, lose the battle [i.e. with the US and Japan]. Believe me, I am European enough to wait for the last possible moment and literally do everything possible to prevent that happening. If Europe is neither willing nor able to develop its own unique economic structure…industrial innovation will pass Europe by.6 The scramble by Mitterrand to get EUREKA, a rival research programme to SDI, off the ground and by other European governments to exert a counterforce pulling scientists and corporations back from contracting to work for the US programme, were further indications of the importance attached to SDI. If Gilpin was somewhat ahead of the times with the book on French science policy, he was somewhat behind with a later and better-known one, US Power and the Multinational Corporation: The Political Economy of Foreign Direct Investment.7 That is to say, he was the initiator for many readers of the notion that big transnational corporations were inimical to the interests not only of developing countries but also of developed ones. Beware, he told his fellow-Americans, of the example of the British who allowed so much of their wealth to be poured into foreign investments that their own industrial base degenerated and went into irreversible decline. This was the exact complement of a view of transnational corporation activity being propounded throughout the Third World, and at UNCTAD and the United Nations, and expounded by André Gunder Frank, Johann Galtung and a host of Latin-American dependencia writers. All of them saw the large foreign corporation as a threat to the autonomy and often the welfare—broadly defined—of the less developed countries. It is my contention that both Gilpin and Galtung et al. were right and that both were also wrong. They were both right in perceiving the power exercised in the system by TNCs, through their influence on the structure of production by the disposition of capital and of technology. It was true, as Gilpin said, that US electronic or clothing firms, for instance, that relocated production abroad either in what Michalet has called ‘relay affiliates’ or else in ‘workshop affiliates’8 were taking jobs from American workers and reducing manufacturing production in the United States. It was also true, as the dependistas said, that TNCs often kept control over research and development and used


530

the proceeds of their sales in LDCs to finance more research and development. But it seemed to stretch credibility to accept both views. Could it be that TNCs were both undermining the power of the US for the benefit of the rest of the world, and that they were also undermining the power of the LDCs for the benefit of the US? That paradox could only be true if it could be argued that the TNCs were both a help and a hindrance, both antipathetic and sympathetic to the interests of both their home countries and their host countries. That, essentially, is my argument: that there is a symbiosis between state and transnational corporation from which both benefit, that they are allies as well as competitors or opponents. That is why the one-sided interpretation of the role of transnational put forward by Gilpin for the United States and by Galtung, UNCTAD and others for the developing countries is behind the times. It is denied every day by the everincreasing structural power of the United States to which US corporations have markedly contributed, and by the ever-increasing collaboration of foreign corporations in the industrial development of at least some developing countries. That this should be so—and I shall suggest in a moment where the evidence for it is to be found—should not surprise anyone with some historical perspective. Take the relations between church and state from say the thirteenth to the eighteenth centuries. All the arguments over papal authority, and over the respective domains of canon and secular law, and all the use of ‘ultimate weapons’ by both sides—excommunication by the church, dispossession and death by the prince—bear witness to a perennial conflict between them. Yet both were well aware of the useful support forthcoming from the other. For the state, the church would legitimize the social order, sanctifying with divine blessing ‘the station in life to which it has pleased God to call us’, legitimizing foreign policy aims in the world beyond Europe, and anathematizing people whom the state regarded as dissidents as well as people whom the church regarded as heretics. For the church, the security necessary to communication between bishops and religious orders, the extension of the faith to the heathen, and resistance to the inroads of Islam on Christianity all required the cooperation of the state. It was only over the rights to tax and the rights of property that the really serious conflicts arose between the state and church, just as today it is over the same issues (in different disguise) that state and corporation come into conflict with each other. For the rest, now as then, it often suits both to pretend that each is totally opposed to the other. Le Rouge et le noir was about a contest already obsolete when Stendhal wrote it, just as Disestablishment was a dead issue when Granville Barker wrote Waste. In the same way, political leaders continue to inveigh against the wickedness of the multinationals, and corporate executives to castigate the restrictions and the excessive nationalism of governments in public, while each—big business and big governments—is in the other’s pocket in private. The best illustration of this symbiosis is to be found in the multitude of joint production arrangements, one of which is almost certain to figure in the financial press on any chosen day of the week. The foreign corporation agrees to provide the technology, to open up access to the coveted export markets and to help arrange the finance; while the host state or one of its major state enterprises guarantees a preferred place in the domestic market, a good share of the output and the profits, freedom to repatriate them and in many cases a slice of the equity. It is by these means that so much of the world’s


531

manufacturing capacity has moved from north to south, from developed to developing countries. It could not have been done without the collusion of the multinationals. It was the Japanese corporations (and their banks) which helped put South Korean manufacturers—ships, steel, shoes and cars—on to Asian export markets. It was United States corporations that built up Mexico’s industrial capacity. It was French companies that made possible industrial development in, for example, the Ivory Coast, just as it was the Germans in Brazil and the British in India. All these newly industrializing countries (NICs) have found an easier and quicker way to gain access to the technology required for rapid industrialization. Japan’s method of buying patents wholesale seems cumbersome and slow by comparison. As a result the great issue of the 1960s—nationalization—is rarely mentioned. As Edith Penrose pointed out many years ago, it was unnecessary and wasteful to buy 51 per cent (let alone 100 per cent) of a company if by buying 5 per cent of the shares a state could gain access to the information needed to negotiate a better bargain whether over profits, wages, exports or training.9 The big issue today is taxation. A good example of how crucial it is can be seen in Indonesia. From barely a dozen in 1973, the number of multinationals with regional headquarters in Manila rose in ten years to over 375. But when Indonesia was obliged to seek IMF help to service its debts, a condition of the $9.7 billion package was that the government reduce its fiscal deficit and tax the multinationals. Over 200, mostly but not all American, very promptly moved their offices elsewhere.10 The moral of the tale is simply that big business and big government are now competing as tax authorities. One exacts tax by using coercive power—the mafias and religious authorities do the same— while the other exacts tax by ‘internalizing the market’ as the business school literature has it; in other words, by taking the tax out before the profit is declared instead of afterwards. The effect is the same, inasmuch as state and corporation then take political (i.e. non-market, non-profit-maximizing) decisions as to how and on whom the proceeds should be spent. As for the international organizations, the ways in which these supranational entities sustain the state is rather different. The similarity of their role with that of the TNCs is only that their sustaining effect has been, on balance, far more effective than their undermining effect. For a great many new states in the world, recognition as members of the United Nations has been the major basis of their claim to legitimacy. For a start, it is necessary to make the point that international organizations fulfil these three functions in the world of states: adaptive, strategic, and symbolic. That is to say, they are sometimes set up in order to allow autonomous political authorities bounded by territorial limits to benefit from the economic opportunities offered by new systems of transport, communications or commerce. Thus, they must agree to adopt standard practices or rules and to resolve differences before they can make use of new technology, whether it is the electric telegraph or the communications satellite. In each case, states have been careful never to let the powers of an international bureaucracy extend beyond what is strictly necessary for the system to be workable. Alternatively, international organizations have been set up by a state or group of states essentially as tools of foreign policy. The cost is small, the PR effect is great, and the achievement (whether it is the administration of Marshall Aid by the OEEC in the 1950s, the coordination of allied forces by NATO or


532

the Warsaw Treaty Organization or the prevention of default and ending of a debtors’ profligacy by the IMF) is very much more impressive than anything that could be achieved through the exercise of bilateral diplomacy by the rich and strong acting on the poor and weak. Finally, some of the best-known international organizations in which the highest hopes reposed have turned out to be primarily symbolic. As Conor Cruise O’Brien, writing about the UN’s Congo operations in 1960, once described it, they are ‘ritual drama’. They express an aspiration; they acknowledge, as by genuflexion, a moral principle; and they profess faith in the reality of a society of mankind—but without accepting for one moment the costs of putting any of these high-minded objectives into practice. It is only necessary to think of the UN’s record on human rights, on sanctions, on genocide, on child welfare and wildlife to see that this is so. Given that many international organizations fulfil more than one of these functions at the same time, this way of looking at them explains better than looking at the subject or issue-area of their operations trade, money, security or whatever—why it is that some international organizations do appear to exercise an influence on states while others have reached an impasse of tedium and irrelevance from which even those most moved by their symbolism cannot save them. Those that are primarily symbolic in function have had least effect on state behaviour. Those that serve some powerful state’s strategic purpose—counting the preservation of an open market system and the property rights of corporate capital as a strategic purpose—have had most effect on state behaviour. But the effects are uneven, as in any political system. The rulers do not abide by the rules they impose on the ruled. Just think of the fiscal profligacy practised by the United States with such impunity and the costly demands made by the IMF on Indonesia. And the organizations that are primarily adaptive do somewhat modify state behaviour, but only within strict limits set by the state’s own definition of its national interest. The result is that generalization about international organizations and their relation to states is increasingly impossible; a general theory is a chimera. Organizations are polarized between those which are irrelevant and dispensable, and those which serve an important political purpose. On their side, states too are polarized. International organizations have been fairy godmothers to weak and small states that posed no threat or that offered no tempting prize to more powerful neighbours. But they were little use to Hyderabad or Goa, to Nepal or Tibet, to the would-be nation-states like Biafra, West Irian, Kurdistan or the Polisario. They were of enormous legitimizing value to Tuvalu or Zimbabwe, to Bangladesh, or to the mini-states of the Caribbean. They are useful tools for strong states and useful sanctifiers and benefactors for small ones. Some, like the World Bank, can offer access to capital and technology which otherwise would be more costly or more inaccessible; state powers are thereby reinforced. Once again, a symbiosis is apparent. The fact that the Finance Minister of Upper Volta can spend up to half his working days receiving visiting aid missions from a galaxy of multilateral organizations and some bilateral donors when he should be putting his country’s finances in some sort of order is just one indication of the mutual interest shared by the IMF and its Victims’, by the World Bank and its mendicant dependants. Broadly speaking, it is the middle powers whose attitude to international bureaucracies is understandably ambivalent. They are too big to need either the sanctification or the largesse of international organizations but too small to be able, alone, to muster either the


533

capital resources or the scientific and technological ones to keep up with the great continental superpowers. This, and not simply age and a sclerotic alliance of conservative interests opposed to change (as described by Mancur Olson11) is part of the reason for the weaknesses of the European Community and the European dilemma as perceived by industrialists like Dr Dekker. So acutely do the industrialists of Europe perceive the vulnerability of their home states to American and Japanese domination in advanced technologies that they have actually formed an association which, in conscious imitation of inter-governmental ones, they call the Group of 22. The 22 are the leading European TNCs like Siemens, Philips, Nestlé, Thyssen, Olivetti, Unilever, Ciba-Geigy and Volvo. Their purpose is to create a new industrial élan in Europe by funding new joint ventures in advanced technology, by combining research efforts and by influencing governments to act more positively—for all the reasons explained by Dr Dekker. This is but one more illustration of a significant new phase in the development of the international political economy in which the association of corporate enterprises is becoming more like the association of political authorities—international organizations— and in which the competition between states for market shares, for capital and for technical leadership is becoming more and more like the competition of corporations for those very same objectives. It is hardly to be wondered at that, as these two global chess games merge with one another, there should be both more cooperation and more conflict between the two groups of players.

Notes 1. J.N.Rosenau, ‘The state in an era of cascading politics, wavering conceptions, widening competence for weathering change’ (mimeo), IPS A XIII World Congress, 1985. 2. H.J.Laski, ‘International government and national sovereignty’ in Problems of Peace, London, Geneva Institute of International Relations, 1927, p. 289. 3. Hunting in northern Europe would soon have ceased through the depletion of species to be a useful source of scarce protein in winter if everyone had been free to do it. Hence its restriction, backed by political power, to members of a ruling class. 4. Robert Gilpin, France in the Age of the Scientific State, Princeton, Princeton University Press, 1968, p. 4. 5. United Kingdom Council of the European Movement, European Cooperation in Advanced Technology, Report of a conference, July 1965, p. 4, quoted in Gilpin, France in the Age of the Scientific State, p. 458. 6. Dr W.Dekker, ‘The importance of innovation on the economic structures of Europe’, Paper delivered at a Financial Times-Institute for Research in Multinationals conference, Munich, April 1985. 7. New York, Basic Books, 1975. 8. C.A.Michalet, Le Capitalisme mondiale, Paris, Presses Universitaires de France, 1976. A relay affiliate was one which reproduced the production process developed for the TNC’s own home state. A workshop affiliate was one in which part of the process was ‘farmed out’ by the parent company to an enterprise in another country. 9. E.Penrose, “‘Ownership and control”: mulinational firms in less developed countries’, in G.Helleiner, ed., A World Divided, Cambridge, Cambridge University Press, 1976.


534

10. International Management, January 1985, p. 9. 11. See M.Olson, The Rise and Decline of States, New Haven, Yale University Press, 1983.


535

61 Who Is “Us”?* Robert B.Reich

* Source: R.Reich, The Work of Nations: Preparing Ourselves for 21st Century Capitalism, 1991, ch. 25, pp. 301–315.

It is right to prefer our own country to all others, because we are children and citizens before we can be travelers and philosophers. George Santayana, The Life of Reason (1905)

1 What is the role of a nation within the emerging global economy, in which borders are ceasing to exist? My answer has, I hope, been clear. Rather than increase the profitability of corporations flying its flag, or enlarge the worldwide holdings of its citizens, a nation’s economic role is to improve its citizens’ standard of living by enhancing the value of what they contribute to the world economy. The concern over national “competitiveness” is often misplaced. It is not what we own that counts; it is what we do. Viewed in this way, America’s problem is that while some Americans are adding substantial value, most are not. In consequence, the gap between those few in the first group and everyone else is widening. To improve the economic position of the bottom four-fifths will require that the fortunate fifth share its wealth and invest in the wealthcreating capacities of other Americans. Yet as the top becomes ever more tightly linked to the global economy, it has less of a stake in the performance and potential of its less fortunate compatriots. Thus our emerging dilemma, and that of other nations as well. History rarely proceeds in a direct line, however. Those who project that today’s steady improvement (or deterioration) over yesterday’s will become even more pronounced tomorrow often end up embarrassed when the future finally arrives. In the intervening moments there will occur an earthquake, a potent idea, a revolution, a sudden loss of business confidence, a scientific discovery—reversing the seemingly most intransigent of trends and causing people to wonder how they could ever have been deluded into believing that any other outcome was ever remotely possible. The predictable failure of all prediction notwithstanding, the public continues to pay attention to stock analysts, trend spotters, futurologists, weather forecasters, astrologers, and economists. Presumably, such respect is due less to the accuracy of their prophecies than to the certainty with which they are delivered. The reader of these pages is duly warned. An all-too-simple extrapolation of the past

Who Is “Us”?

536

into the future would show a continuing rise in the fortunes of symbolic analysts and a steady decline in the fortunes of almost everyone else The costs of worldwide transportation and communications will continue to decline—creating an ever larger market, and burgeoning demand, for the services of America’s problem-solvers, identifiers, and brokers, but simultaneously generating an ever larger supply of unskilled workers. In consequence, America’s symbolic analysts will become even wealthier; routine producers will grow poorer and fewer in number; and, with the enhanced mobility of world labor and the versatility of labor-saving machinery, in-person servers will become less economically secure. The fortunes of the most well-off and the least will thus continue to diverge. By 2020, the top fifth of American earners will account for more than 60 percent of all the income earned by Americans; the bottom fifth, for 2 percent. Symbolic analysts will withdraw into ever more isolated enclaves, within which they will pool their resources rather than share them with other Americans or invest them in ways that improve other Americans’ productivity. An ever smaller proportion of their incomes will be taxed and thence redistributed or invested on behalf of the rest of the public. Government spending on education, training, and infrastructure will continue to decline as a proportion of the nation’s total income; any savings attributable to a smaller defense budget will result in further tax reductions and the diminution of the fiscal deficit. Poorer cities, townships, and states will be unable to make up the difference. Distinguished from the rest of the population by their global linkages, good schools, comfortable lifestyles, excellent health care, and abundance of security guards, symbolic analysts will complete their secession from the union. The townships and urban enclaves where they reside, and the symbolic-analytic zones where they work, will bear no resemblance to the rest of America; nor will there be any direct connections between the two. America’s poorest citizens, meanwhile, will be isolated within their own enclaves of urban and rural desperation; an ever-larger proportion of their young men will fill the nation’s prisons. The remainder of the American population, growing gradually poorer, will feel powerless to alter any of these trends. It is not that simple, of course. Other events will likely intervene to deflect this trajectory. Not the least is the inability of symbolic analysts to protect themselves, their families, and their property from the depredations of a larger and ever more desperate population outside. The peace of mind potentially offered by platoons of security guards, state-of-the-art alarm systems, and a multitude of prisons is limited. There is also the possibility that symbolic analysts will decide that they have a responsibility to improve the well-being of their compatriots, regardless of any personal gain. A new patriotism would thus be born, founded less upon economic self-interest than upon loyalty to the nation.

2 What do we owe one another as members of the same society who no longer inhabit the same economy? The answer will depend on how strongly we feel that we are, in fact, members of the same society.


537

Loyalty to place—to one’s city or region or nation—used to correspond more naturally with economic self-interest. Individual citizens supported education, roads, and other civic improvements, even when the individual was likely to enjoy but a fraction of what was paid out in the short term, because it was assumed that such sacrifices would be amply rewarded eventually. Civic boosterism, public investment, and economic cooperation were consistent with Tocqueville’s principle of “self-interest rightly understood.” As our fellow citizens grew wealthier and more productive, we benefited by their ability to give us more in exchange for what we offered them. And as we resisted opportunistic behavior, so did they, with the result that we benefited all the more. The resulting networks of economic interdependence induced the habits of citizenship. Between 1950 and the early 1970s, the American economy as a whole began to exemplify this principle. Labor, business, and the broader public, through its elected representatives, tacitly cooperated to promote high-volume production; the resulting efficiencies of scale generated high profits; some of the profits were reinvested to create even vaster scale, and some were returned to production workers and middle-level managers in the form of higher wages and benefits. As a result, large numbers of Americans entered the middle class, ready to consume the output of this burgeoning system. But as the borders of cities, states, and even nations no longer come to signify special domains of economic interdependence, Tocqueville’s principle of enlightened selfinterest is less compelling. Nations are becoming regions of a global economy; their citizens, laborers in a global market. National corporations are turning into global webs whose high-volume, standardized activities are undertaken wherever labor is cheapest worldwide, and whose most profitable activities are carried out wherever skilled and talented people can best conceptualize new problems and solutions. Under such circumstances, economic sacrifice and restraint exercised within a nation’s borders is less likely to come full circle than it was in a more closed economy. The question is whether the habits of citizenship are sufficiently strong to withstand the centrifugal forces of the new global economy. Is there enough of simple loyalty to place—of civic obligation, even when unadorned by enlightened self-interest—to elicit sacrifice nonetheless? We are, after all, citizens as well as economic actors; we may work in markets but we live in societies. How tight is the social and political bond when the economic bond unravels? The question is relevant to all nations subject to global economic forces—forces which are reducing the interdependence of their citizens and separating them into global winners and losers. In some societies, where national allegiances are stronger than in others, the balance between societal ties and economic ties tilts toward the former. The pull of the global economy notwithstanding, national allegiances are sufficiently potent to motivate the winners to continue helping the losers. The “we’re all in it together” nationalism that characterizes such places is founded not only on enlightened self-interest but also on a deeply ingrained sense of shared heritage and national destiny. The Japanese, Swedes, Austrians, Swiss, and Germans, for example, view themselves as cultures whose strength and survival depend, to some extent, on sacrifices by the more fortunate among them. It is a matter of national duty and pride. Partly as a result, the distribution of income within these nations has been among the most equal of any countries (although the global

Who Is “Us”?

538

division of labor is beginning to drive a wedge between their rich and poor and testing their commitment to economic equality). These nations, incidentally, experienced during the decades of the 1960s and 1970s some of the most spectacular records of growth of all industrialized nations.1 Could such sentiments be nurtured in America? Should they be?

3 Nationalism can be a hazardous sentiment. The same “we’re all in it together” attitude that elicits mutual sacrifice within a nation can easily degenerate into jingoistic contempt for all things foreign. Indeed, the two emotions tend to reinforce one another. It is a commonplace notion in Britain that the nation’s citizens have never since displayed such virtue and solidarity as when they fought Hitler. America’s Cold War with the Soviet Union inspired, and provided justification for, billions of dollars of public expenditure on highways, education, and research. The willingness of talented Japanese citizens to work long hours and receive relatively low incomes for the honor of doing their part for their country is fueled by the same set of emotions that makes it difficult for the Japanese to open their borders to foreign products and immigrants. History offers ample warning of how “zero-sum” nationalism—the assumption that either we win or they win—can corrode public values to the point where citizens support policies that marginally improve their own welfare while harming everyone else on the planet, thus forcing other nations to do the same in defense. Armaments escalate; trade barriers rise; cold wars turn hot. The same social discipline and fierce loyalty that have elicited sacrifices among Germans and Japanese have also, in this century, generated mind-numbing atrocities. Unbridled nationalism can also cause civic values to degenerate at home. Nations grow paranoid about foreign agents in their midst; civil liberties are restricted on grounds of national security. Neighbors begin to distrust one another. Tribal allegiances can even tear nation-states apart. The violence that periodically erupts between Greek and Turkish Cypriots, Armenians and Azerbaijanis, Albanians and Serbs, Flemings and Walloons, Vietnamese and Cambodians, Israelis and Palestinians, Sikhs and other Indians, Tamils and Sinhalese, and Lebanon’s Christian and Moslem sects is grim evidence of the costs of tribal loyalty. The argument against zero-sum nationalism, and in favor of a larger and more cosmopolitan perspective, seems especially strong in light of the growing inequalities in the world. The gap between the top fifth and bottom four-fifths of incomes within the United States is negligible in comparison with the gap between the top fifth and bottom four-fifths of the world’s population. North America, Western Europe, and East Asia— together comprising the top fifth—account for three-quarters of the gross world product and 80 percent of the value of world trade. As these wealthy regions have become uncoupled from the rest of the world, much of what remains behind is sinking precipitously into hopeless poverty. Between 1970 and 1980, the number of undernourished people in developing nations (excluding China) increased from 650 million to 730 million. Since 1980, economic


539

growth rates in most of these nations have slowed and real wages have dropped further. In Africa and Latin America, per capita incomes were substantially lower in 1990 than in 1980. Commodity prices have plummeted; indebtedness to global banks has crippled many underdeveloped economies, as more than $50 billion is transferred each year to advanced nations. The ravages of deforestation, erosion, and large-scale farming have taken their toll on much of the Third World. Meanwhile, the world’s poor populations are giving birth at a much higher rate than are the world’s rich. Sixty percent of the 12,000 children born into the world each hour join families whose annual per-person incomes are lower than $350. In 1990, just over 5 billion human beings lived on the planet; their number is expected to reach 8 billion by 2025, and 16 billion by the end of the twentyfirst century. The number of impoverished people has grown dramatically in Brazil, Chile, Ghana, Jamaica, Peru, and the Philippines. Life expectancy has declined in nine sub-Saharan African countries; deaths from malnutrition among infants and children have increased.2 A focus on national well-being is also dangerously narrow in relation to other problems on which global cooperation is essential: acid rain, the depletion of the ozone, the pollution of oceans, the use of fossil fuels and global warming, the destruction of speciesrich tropical rain forests, the proliferation of nuclear weapons, the drug trade, the spread of AIDS, international terrorism. Narrowly nationalistic attitudes render solutions to these and other transnational crises all the more difficult to achieve.

4 Zero-sum nationalism also endangers global economic prosperity. The neomercantilist premise that either they win or we win is simply incorrect. As one nation’s workers become more insightful and educated, they are able to add more wealth to the world. Everyone on the planet benefits from smaller and more powerful semiconductor chips regardless of who makes them. It is true, of course, that the nation whose workers first gain the insights are likely to benefit disproportionately. This advantage may cause other nations’ citizens to feel relatively poorer notwithstanding their absolute gain. Sociologists have long noted the phenomenon of “relative deprivation,” whereby people evaluate their well-being in light of others’ wealth. The average citizen of Great Britain is in absolute terms far better off than twenty years ago but feels poorer now that the average Italian has pulled ahead. When I ask my students whether they would prefer living in a world in which every American is 25 percent wealthier than now and every Japanese is much wealthier than the average American, or in one in which Americans are only 10 percent wealthier than now but ahead of the average Japanese, a large number usually vote for the second option. People may be willing to forgo absolute gains to prevent their perceived rivals from enjoying even greater gains. While understandable, such zero-sum impulses are hardly to be commended as a principle of international economic behavior. Since economic advances rarely benefit the citizens of all nations in equal proportion, such an approach, if widely adopted, would block most efforts to increase global wealth. Economic interdependence runs so deep, in fact, that any zero-sum strategy is likely to

Who Is “Us”?

540

boomerang, as the members of the Organization of Petroleum Exporting Countries discovered in the 1970s when their sky-high oil prices plunged the world into recession and reduced the demand for oil. Today, no nation’s central banker can control its money supply or the value of its currency without the help of other nations’ central bankers, nor can a nation unilaterally raise its interest rates or run large budget surpluses or deficits without others’ cooperation or acquiescence. These days, every advanced nation depends on others as a market for, and source of, its goods. The Japanese need a strong and prosperous America as a market for their goods and a place to invest their money. If any step they might take were to precipitate steep economic decline in the United States, the results would be disastrous for the Japanese as well. But what if foreigners dominate a major technology, as it seen likely the Japanese soon will with advanced semiconductors, high-definition television, and dozens of other gadgets? Again, we should beware of zero-sum assumptions. The Japanese mastery of particular technologies will not foreclose technological progress in the United States or elsewhere. Technologies are not commodities for which world demand is finite, nor do they come in fixed quantities that either they get or we get. Technologies are domains of knowledge. They are like the outer branches of a giant bush on which countless other branches are growing all the time. While Americans need direct experience in researching, designing, and fabricating technologies on outer branches if they are to share in future growth, these need not be exactly the same branches that are occupied by another nation’s work force.

5 The cosmopolitan man or woman with a sense of global citizenship is thus able to maintain appropriate perspective on the world’s problems and possibilities. Devoid of strong patriotic impulse, the global symbolic analyst is likely to resist zero-sum solutions and thus behave more responsibly (in this sense) than citizens whose frame of reference is narrower. But will the cosmopolitan with a global perspective choose to act fairly and compassionately? Will our current and future symbolic analysts—lacking any special sense of responsibility toward a particular nation and its citizens—share their wealth with the less fortunate of the world and devote their resources and energies to improving the chances that others may contribute to the world’s wealth? Here we find the darker side of cosmopolitanism. For without strong attachments and loyalties extending beyond family and friends, symbolic analysts may never develop the habits and attitudes of social responsibility. They will be world citizens, but without accepting or even acknowledging any of the obligations that citizenship in a polity normally implies. They will resist zerosum solutions, but they may also resist all other solutions that require sacrifice and commitment. Without a real political community in which to learn, refine, and practice the ideals of justice and fairness, they may find these ideals to be meaningless abstractions. Senses of justice and generosity are learned. The learning has many roots, but significant among them is membership in a political community. We learn to feel


541

responsible for others because we share with them a common history, we participate with them in a common culture, we face with them a common fate. As the social philosopher Michael Ignatieff has written, “We think of ourselves not as human beings first, but as sons, and daughters…tribesmen, and neighbors. It is this dense web of relations and the meanings which they give to life which satisfies the needs which really matter to us.”3 That we share with others nothing more than our humanity may be insufficient to elicit much sacrifice. The management consultant living in Chappaqua and commuting to a steel-and-glass tower on Park Avenue, and dealing with clients all over the world, may feel slightly more responsibility toward a poor family living 3,200 miles away in East Los Angeles than to a poor family of Mexicans living 3,200 miles away in Tijuana, but the extra measure of affinity may not be enough to command his or her energies or resources. A citizen of the world, the management consultant may feel no particular bond with any society. Cosmopolitanism can also engender resignation. Even if the symbolic analyst is sensitive to the problems that plague the world, these dilemmas may seem so intractable and overpowering in their global dimension that any attempt to remedy them appears futile. The greatest enemy of progress is a sense of hopelessness; from a vantage point that takes in the full enormity of the world’s ills, real progress may seem beyond reach. Within smaller political units like towns, cities, states, and even nations, problems may seem soluble; even a tiny improvement can seem large on this smaller scale. As a result, where the nationalist or localist is apt to feel that a sacrifice is both valorous and potentially effective, the cosmopolitan may be overcome by its apparent uselessness.4 The reader has only to reflect upon personal experience. Nothing more surely stills reformist zeal than a faithful reading of The New York Times or other great newspaper of the world, in which the global dimensions of hunger, disease, racism, environmental depredation, and political injustice are detailed daily. It should not come as a surprise, then, that all great social movements have begun locally. Those who aim to reform the world in one great swoop often have difficulty signing up credulous recruits. In short, while a cosmopolitan view provides a useful and appropriate perspective on many of the world’s problems and avoids the pitfalls of zero-sum thinking, it may discourage the very steps necessary to remedy the problems it illuminates. It is not clear that mankind is significantly better off with an abundance of wise cosmopolitans feeling indifferent or ineffective in the face of the world’s ills than it is with a bunch of foolish nationalists intent on making their particular society Number One.

6 But must we choose between zero-sum nationalism and impassive cosmopolitanism? Do these two positions describe the only alternative modes of future citizenship? Unfortunately, much of the debate we hear about America’s national interest in the global economy is framed in just these dichotomous terms. On one side are zero-sum nationalists, typically representing the views of routine producers and in-person servers, urging that government advance America’s economic interests—even at the expense of others around the globe. In their view, unless we become more assertive, foreigners will

Who Is “Us”?

542

continue to increase their market shares at America’s expense in industry after industry— exploiting our openness, gaining competitive advantage over us, ultimately robbing us of control over our destinies. On the other side are laissez-faire cosmopolitans, usually representing the views of symbolic analysts, arguing that government should simply stay out. In their view, profit-seeking individuals and firms are far better able to decide what gets produced where; governments only mess things up. Free movement of all factors of production across national boundaries ultimately will improve everyone’s lot. What is being lost in this debate is a third, superior position: a positive economic nationalism, in which each nation’s citizens take primary responsibility for enhancing the capacities of their countrymen for full and productive lives, but who also work with other nations to ensure that these improvements do not come at others’ expense. This position is not that of the laissez-faire cosmopolitan, because it rests on a sense of national purpose—of principled historic and cultural connection to a common political endeavor. It seeks to encourage new learning within the nation, to smooth the transition of the labor force from older industries, to educate and train the nation’s workers, to improve the nation’s infrastructure, and to create international rules of fair play for accomplishing all these things. The objectives of such investments are unambiguously public. Neither is this the position of a zero-sum nationalist: Here the overarching goal is to enhance global welfare rather than to advance one nation’s well-being by reducing another’s. There is not a fixed amount of world profit to be divided or a limited market to be shared. It is not “their” corporations against “ours” in a fight for dominance in world commerce. We meet instead on an infinitely expanding terrain of human skills and knowledge. Human capital, unlike physical or financial capital, has no inherent bounds. Indeed, these nationalist sentiments are likely to result in greater global wealth than will cosmopolitan sentiments founded upon loyalty to no nation. For like villagers whose diligence in tending to their own gardens results in a bounteous harvest for all, citizens who feel a special obligation to cultivate the talents and abilities of their compatriots end up contributing to the well-being of compatriots and noncompatriots alike. One nation’s well-being is enhanced whenever other nations improve the capacities of their own citizens. To extend the metaphor, while each garden tender may feel competitive with every other, each also understands that the success of the total harvest requires cooperation. While each has a primary responsibility to tend his own garden, each has a secondary responsibility to ensure—and a genuine interest in seeing—that all gardens flourish. Thus positive economic nationalism would eschew trade barriers against the products of any work force as well as obstacles to the movement of money and ideas across borders. Even were such obstacles enforceable, they would only serve to reduce the capacity of each nation’s work force to enjoy the fruits of investments made in them, and in others. But not all government intervention would be avoided. Instead, this approach would encourage public spending within each nation in any manner that enhanced the capacities of its citizens to lead full and productive lives—including pre- and postnatal care, childcare and preschool preparation, excellent primary and secondary education, access to college regardless of financial condition, training and retraining, and good infrastructure. Such investments would form the core of national economic policy. Positive nationalism also would tolerate—even invite—public subsidies to firms that


543

undertook within the nation’s borders high-value-added production (complex design, engineering, fabrication, systems integration, and so forth), so that the subsidy-granting nation’s work force could gain sophisticated on-the-job skills. But it would draw no distinctions based on the nationalities of the firm’s shareholders or top executives. To ensure against zero-sum ploys in which nations bid against one another to attract the same set of global firms and related technologies, nations would negotiate over the appropriate levels and targets of such subsidies. The result would be a kind of “GATT for direct investment”—a logical extension of the General Agreement on Tariffs and Trade that the United States sponsored after World War II—setting out the rules by which nations could bid for high-value-added investments by global corporations. Barred would be threats to close the domestic market unless certain investments were undertaken within it, for such threats would likely unravel into zero-sum contests. Instead, the rules would seek to define fair tactics, depending upon the characteristics of the national economy and the type of investment being sought. For example, the amount of permissible subsidy might be directly proportional to the size of the nation’s work force but inversely proportional to its average skills. Nations with large and relatively unskilled work forces would be allowed greater leeway in bidding for global investment than nations with smaller and more highly skilled work forces. Other kinds of subsidies would be pooled and parceled out to where they could do the most good, as the European Community has begun to do regionally. For example, nations would jointly fund basic research whose fruits are likely to travel almost immediately across international borders—projects such as the high-energy particle accelerator, the human genome, and the exploration of space. (Single governments are unlikely to support many such projects on their own, given that the entire world so easily benefits from them.) How such funds were apportioned, and toward what ends, would, of course, be subject to negotiation. Positive economic nationalism also would ease the transition of a work force out of older industries and technologies in which there was worldwide overcapacity. This might take the form of severance payments, relocation assistance, extra training grants, extra unemployment insurance, regional economic aid, and funds for retooling or upgrading machinery toward higher-value-added production. Since every nation benefits when overcapacity anywhere is reduced, these subsidies might come from a common fund established jointly by all nations. Payments into the fund could be apportioned according to how much of that particular industry’s capacity lay within each nation’s borders at the start. Finally, positive economic nationalism would seek to develop the capacities of the work forces of the Third World—not as a means of forestalling world communism or stabilizing Third World regimes so that global companies can safely extract raw materials and sell products within them—but as a means of promoting indigenous development and thereby enhancing global wealth. To this end, the shift of high-volume, standardized production to Third World nations would be welcomed, and markets in advanced economies would be open to them. Advanced nations would reduce the Third World’s debt burden, make new lending available, and monitor the loans more carefully than in the past.

Who Is “Us”?

544

7 The pressures of global change have fragmented the American electorate. Routine producers and in-person servers—tending toward zero-sum nationalism—fear that foreigners, the Japanese in particular, are taking over the nation’s assets and secretly influencing American politics. They resent low-wage workers in Southeast Asia and Latin America who are inheriting many of America’s routine production jobs and seem in addition to be swarming into American cities. Many symbolic analysts—tending toward laissez-faire cosmopolitanism—feel no particular urgency about the economic plight of other Americans, on the one hand, and, on the other, feel ineffectual and overwhelmed with regard to many of the larger problems facing the rest of the world. Neither constituency, in other words, is naturally disposed to positive nationalism. Those who are threatened by global competition feel that they have much to lose and little to gain from an approach that seeks to enhance world wealth, while those who are benefiting the most from the blurring of national borders sense that they have much to lose and little to gain from government intervention intended to spread such benefits. The direction we are heading is reasonably clear. If the future could be predicted on the basis of trends already underway, laissez-faire cosmopolitanism would become America’s dominant economic and social philosophy. Left to unfold on its own, the worldwide division of labor not only will create vast disparities of wealth within nations but may also reduce the willingness of global winners to do anything to reverse this trend toward inequality—either within the nation or without. Symbolic analysts, who hold most of the cards in this game, could be confident of “victory.” But what of the losers? We are presented with a rare historical moment in which the threat of worldwide conflict seems remote and the transformations of economies and technology are blurring the lines between nations. The modern nation-state, some two hundred years old, is no longer what it once was: Vanishing is a nationalism founded upon the practical necessities of economic interdependence within borders and security against foreigners outside. There is thus an opportunity for us, as for every society, to redefine who we are, why we have joined together, and what we owe each other and the other inhabitants of the world. The choice is ours to make. We are no more slaves to present trends than to vestiges of the past. We can, if we choose, assert that our mutual obligations as citizens extend beyond our economic usefulness to one another, and act accordingly.

Notes 1. Much has been written about the “developmental state” of modern Japan. South Korea and Hong Kong, once touted by orthodox free marketeers as models of laissez-faire economic individualism, on closer inspection look remarkably like their more advanced neighbor to the north. See, for example, Alice Amsden, Asia’s Next Giant (New York: Oxford University Press, 1989), and M.Castells and L.Tyson, “High Technology and the Changing International Division of Production,” in R.Purcell (ed.), The Newly Industrializing Countries in the World Economy (Boulder, Colo.: Lynne Rienner, 1989). Austria, Switzerland, and Sweden represent


545

a different path, of course, but these nations also are characterized by systems of internal bargaining which soften adjustment for their least fortunate citizens and elicit sacrifices from their most fortunate. See, for example, Peter Katzenstein, Corporatism and Change: Austria, Switzerland, and the Politics of Industry (Ithaca, N.Y.: Cornell University Press, 1989). 2. See Global Outlook 2000 (New York: United Nations 1990), pp. 202–21, 285–97. 3. Michael Ignatieff, The Needs of Strangers (New York: Viking Penguin, 1985), p. 29. 4. Jonathan Glover, “It Makes No Difference Whether or Not I Do It,” Supplemental Proceedings of the Aristotelian Society, New York, 1975.


546

62 Beyond the Nation-State: The Multinational State as the Model for the European Community* M.Rainer Lepsius

* Source: Telos, 1992, no. 1, pp. 57–76.

The Future of the European Nation-State Closer economic and security cooperation between European nation-states after 1945 obtained not only as a result of the desire for European unity but also as a result of active US sponsorship. The Organization for European Economic Cooperation (OEEC) and the Brussels Pact of 1948 evolved into NATO the following year. In 1952 Jean Monnet’s principle of supranationality became concrete in the European Coal and Steel Community (ECSC). Along with the Rome Treaty that created the European Economic Community (EEC) and the European Atomic Community (Euratom), the ECSC was the first attempt to integrate European nation-states into a larger political and economic system. Participating states agreed to transfer partial sovereignty to the European Community (EC), which meant the direct subordination of individual citizens, national institutions and courts to supranational agencies. Although international treaties were to bind only the governments of the contracting states, which would retain unlimited jurisdiction in their own spheres as well as the option to withdraw at any time, this is no longer the case in the EC. The expansion of EC jurisdiction through its regulative resolutions further eroded the autonomy of members. Thus a process was set in motion which in principle may spell the end of the European nation-state. What is the future of the European nation-state now that its room for maneuver has been limited, its ability to compete in the international arena is in doubt, and its complete subsumption within a European Union (EU) is on the agenda? This state of affairs has raised again the old question of a European federation and the new question of whether the nation-state is compatible with federalism. One thing is certain: European unification cannot proceed independently of the nation-state as a given historical fact. One option, a “United States of Europe” modeled after the USA, would replace existing nation-states with the EC as a unitary state. Presently, the EC shares sovereignty with its member states, but its growing independence from them has set it on a trajectory of becoming a unified state. Indicative is the fact that the European Supreme Court has authority to issue sanctions. While a Council of Ministers controlled by the national governments of the member states remains the ultimate EC authority, the Commission’s

Beyond the Nation-State 547 prerogatives are becoming increasingly dependent on the recommendations of the European Parliament. Even though the European Parliament’s power is limited, its freedom of action is expanding irreversibly. While the EC institutional order is inherently complex, it is nonetheless the core of a new political entity analogous to the 1871 German Reich. Then, as now, sovereignty lay with the contracting states, even though the people elected a parliament limited in its ability to initiate legislation and to control the budget. Although the Reich had a limited jurisdiction and was dependent on the Länder in budgetary matters, it constituted an economic and legal community. The analogy, however, should not be carried too far. The Reich had wider functions in military matters and foreign policy than the EC today. The similarity lies more in the dual structure of legitimacy, in the tension between the member states and popular representation. In the Reich the result was internal political paralysis. Something similar is also developing within the EC. Potential conflict between the two principles of legitimacy can only be overcome in the short run. Another option is a federal system, either as a confederation or as a federal state. This tendency is evident in organizations such as the West European Union (WEU) and European Political Cooperation (EPC). Whereas the first is conceived in terms of strengthening the European pillars of NATO, the second seeks to coordinate European foreign policy. Both lack supranational organizational and legislative powers and are only based on voluntary cooperation and international agreements. They are, however, the basis of a European confederation. The EC and the WEU are far apart, since the tendency of one is supranational and the tendency of the other is national or at best international. In the former, the heads of state and government function as a European Council with a kind of extra-constitutional authority; in the latter, they are merely representatives of the member states. In like manner, the foreign ministers who make up the EC Council of Ministers merely function unilaterally or cooperate multilaterally in the EPC. Thus presently the organizational principles of both the nation-state and federalism coexist without either gaining predominance. This duality and the various options available for a EU have not yet received the public attention they deserve. Not only do national governments still command considerable loyalty, but the nation-state is still the primary locus of political legitimacy. Despite its inability to defend its borders and its dependence on international economic forces, domestically it still retains full authority. In addition, there is a peculiar distribution of responsibility regarding certain political matters. For example: if the domestic economy is good, that is hailed as the result of a successful governmental policy; if it is bad, it is blamed on world economic conditions over which the government has no control. This type of rationalization—giving the nation-state credit for its political decisions at times and absolving it for them at others—upholds the nation-state’s legitimacy while diminishing its authority. This division of legitimacy and authority is typical of the process of European integration. Though legitimacy is located primarily in the member states, the EC is staking an independent domain beyond their control. The accumulation of legal ordinances and their enforcement have strengthened the Commission and the decisions of the European Supreme Court. The power of the European Parliament expands as the


548

“legitimation gap” of the Commission’s “bureaucratic autocracy” becomes increasingly problematic. But this also encourages the development of a European centralized state. National constitutional values are gradually transcended through a universalization of human rights that becomes the basis of EC legitimacy. Yet the EU remains bound to the nation-state. Therefore, attempts to develop a structure of legitimacy above and beyond the nation-state cannot simply begin with an ideal “European nation.” Existing nation-states must be incorporated systematically in the allocations of authority and legitimacy. That amounts to the conscious federalization of a united Europe. Since EC jurisdictions are still closely defined, it has been possible to ignore the problem of a European “supranational state.” This is no longer the case and after 1992 the problem will have to be confronted. A “Europe of fatherlands,” such as Charles de Gaulle once foresaw, must be placed on the agenda. Should the system of European nation-states evolve into one Europe, the present “nations,” i.e. politically autonomous entities defined in terms of state citizenship, will become “nationalities,” i.e., minorities within a federation. In this context, however, it is worth recalling that the Austro-Hungarian Empire came to ruin because of its failure to solve the “nationality question” and its inability to construct a political order giving minorities sufficient autonomy while maintaining a common will at the federal level. The universalization of human and civil rights allows for the autonomy and participation of individuals but not of corporate entities. “Nationalities” require collective autonomy and participation, which is only possible in a federation. Hitherto the EC has proceeded to develop a federal state and ruled out the alternative of a confederation. While this alternative may create the same problems with democracy encountered by “cooperative federalism” in Germany, federalism remains a viable alternative for the integration of European nation-states. In a nutshell, the nation-states’ obvious limitations in conducting their own affairs require new forms of political integration. In principle this could be accomplished by imperialist means. In Western Europe, however, there are no imperialist states with enough power to do so. It could also be accomplished through the revival of a federal system similar to the one which has recently resolved regional tensions in, e.g., Spain and Belgium. If the EU continues to develop along the lines of a “supranational state,” the result will be the escalation of “conflicts between nationalities.” In this context it becomes crucial to find new ways to link the nation-state with the federal system. This would create new challenges for the nation-state, which will have to confront tensions within supranational as well as subnational political entities. To differentiate the equality of citizens from that of regions, religions, cultures, ethnicities and social formations allows for the grounding of European institutions in transnational values, but it does not require the dissolution of the nation-state as such. The latter is heir to a dual tradition: the right of social entities to independence and the development of civic equality irrespective of regional, cultural and ethnic differences. Both traditions exist in a relation of tension within the history of the European nationstate and will determine European integration. The future of this project depends on the successful management of this tension.

Beyond the Nation-State 549

Nation-State or Multinational State? The EC may have reached a stage requiring institutional innovations. This is the result of the constant expansion of EC jurisdictions and the lack of sufficient direct democratic participation in and control over EC decisions. The two cannot be separated: the greater the scope of regulations, the more necessary a direct legitimation of decisions. During the last thirty years the EC institutional structure has changed only slightly. Decision-making procedures established in the original treaties were reiterated in the expansion of the EC from six to twelve members. They have proven to be exceptionally viable. Of course, this has not always been the case. But the efficiency with which the domestic market has been created is indicative of this viability.1 It is based on the jurisdictions (including the right of legal sanctions) the EC has acquired over the years as well as the political will of the member states. The European Parliament has decision-making power and the potential for selflegitimation. Composed of the delegates of the parliaments of the member states, it has grown beyond a consultative assembly through direct elections (since 1979) and expanded rights of participation (via the unified European acts of 1986) in “cooperative procedures” with the Commission and the Council of Ministers. Yet it has not assumed the function of a parliament. While its basis of legitimacy is the democratic representation of the citizens of the member states, its authority is less than that of a normal parliament. Since, however, it is not a “governing” parliament, no European party system has developed. Political parties form loose factional alliances and compete for power only on the national level. Citizens see European elections as secondary phenomena mediated by their national political systems. To this extent, the European Parliament’s legitimacy is purely formal.2 The EC shares with the Commission peculiar planning, fiscal and executive functions, although the former exercises legislative initiatives and enforces European court decisions. Since there is no responsible parliamentary government, a political “regime”3 creates tasks for EC as well as national representatives and experts. Their “interfacing” is the basis of the Commission’s planning and steering capacity. Cadres structure and participate in the decisions of the Council of Ministers. The EC institutional order is reminiscent of the constitution of the 1871 German Reich, which also came about through an alliance of princes and Hanseatic cities rather than through the people’s constituent act. In both cases a kind of “supplementary representation” of the people was instituted above a parliament with limited jurisdiction. In due course the German Reich developed into a nation-state without a suitable constitution. The chancellor was still appointed by the Kaiser and was primarily Prussia’s prime minister. The various state offices became governmental ministries whose bureaucratic elites cooperated with organized interest groups without adequate parliamentary controls. While the Reichstag retained budgetary powers, it was never a “governing parliament.” Parties functioned primarily as opposition groups and did not develop into political organs able to govern or constitute a national will. A comprehensive constitutional development was blocked not only by antipathy to


550

democracy but also by the unresolved tension between the federal units and the “imperial center.”4 Despite the constitution, the political center continued to accumulate power until the 1918 revolution. While the EC is not the same as Imperial Germany, there is a basic conflict in its institutional order which is not likely to be resolved naturally. The EC is both a contractual “functional alliance”5 and a supranational organization which can restrict its members’ activities. Strengthening the EC’s direct legitimacy, rather than the indirect legitimacy obtained through the member states, has expanded the scope of EC regulations. Similarly, once the European Parliament achieves full jurisdiction, its democratic convictions will legitimate the EC as a sovereign entity. This is the goal of various reform proposals. They seek to establish parliamentary control over the various commissions as if it were a parliamentary government, relegating the Council of Ministers to the role of a second chamber. Here the model is that of the 19th century European nation-state.6 Such an expansion of the European Parliament’s power neither resolves the problems of jurisdiction between the Council of Ministers and the Commission nor defines the European Parliament’s jurisdiction. The fundamental conflict lies in the dual basis of EC legitimacy, which derives partly from the treaty of the members (the legitimate governments of the respective nation-states are the “dominant agents of the treaty”) and partly from direct representation of the “people of the EC” in the European Parliament. Yet without a European citizenry EC membership complete with new passports remains purely symbolic. The rights of European citizens are beginning to develop in supranational terms of freedom to live, travel and study, settlement rights, recognition of degrees and the like. Exclusively national rights will give way as soon as the active and passive franchise is established within all member states. Expansion of EC legislation increasingly subordinates the courts and the citizens of the member states to legal norms originating outside national legislation. Its various legal systems are not determined democratically. The people are “sovereign” in the national state and the “subjects” of the EC. Thus the development of the EC into an EU is not merely a question of institutional reform but rather of mediation between different principles of legitimacy. According to the federal state model, dual legitimacy will be suspended by transferring sovereignty to the center. As the member states are subordinated to the federation, the latter assumes not only the most essential functions but the “authority to allocate jurisdiction” (Kompetenzkompetenz) as well. Indicative is the “cooperative federalism” of the German constitutional system in which member states retain only limited autonomy. They have the monopoly of administration, their own revenues and participation in federal legislation, but no veto power. Only by securing a majority in the Federal Council can a member state prevail against a federal parliamentary majority in the so-called consensual legislation. Federal law supersedes state law, and citizenship is bestowed by the federal state. The transformation of the EC into an EU along federal lines would destroy the dual basis of legitimacy. This model appears not only plausible but self-evident. What remains open is the mediation between the “Chamber of States” and the “House of Representatives.” The problem concerns respective jurisdictions and compromise

Beyond the Nation-State 551 regulations in case of legislative differences. These are decisive in determining the political process. In this case, the principle of subsidiarity really does not limit jurisdiction. The distribution of jurisdictions, according to the rule that each should do what he can do best, presupposes the notion of a hierarchy of tasks and agencies. The highly functional inter-weaving of political spheres introduces a political entanglement into the federal system and over time also a mixed financing of programs—a pastiche of what were once clearly distinguishable jurisdictions.7 The result is an increase in regulative arrangements. The expansion of jurisdictions has given power to those who allocate jurisdictions and this higher level needs a stronger basis of legitimacy. Present internal dynamics lead to the conclusion that European unity will culminate in a federal state. Once in motion, this process can be slowed down or speeded up, but its direction cannot be altered. With respect to relatively concrete tasks, the EC has expanded its jurisdictions through treaties. The four so-called basic freedoms which determine EC goals—freedom of travel, freedom of trade, freedom of services, and freedom of movement of capital within EC borders—have increased the political space for regulation: rights of cooperation of employees in the European joint stock exchange, the packaging of goods, students’ exchanges, automobile pollution regulations, and the regulation of television advertising. Open competition seems to be both instrumental and value-neutral but requires intervention in the social, fiscal, environmental, research and cultural policies of member states. Such regulations also come under the jurisdiction of the European Supreme Court and thus directly affect member states without the mediation of national parliaments. EC jurisdiction is buttressed by the formally equitable standards of the unified domestic market. Given the interdependence of the various political spheres, EC jurisdiction will continue to expand and include foreign economic policies as well as external and internal European security. The jurisdictional claims of an EU eventually would include all the tasks and duties of traditional nation-states. Problems are increasing with the expansion of regulations and are having a growing impact on the member states’ political structure. The best example is the proposed currency union and the establishment of a European central bank. These goals appear instrumentally plausible and necessary for a unified domestic market. Fixed exchange rates and standard monetary and credit policies are needed to control inflation and discipline the member states’ economies. Without them it is impossible to steer the member states’ economic and social development. The whole national system of interest mediation and compensation would be subject to the European central bank’s monetary and credit policies. Leaving aside the economic consequences, such a European central bank would become the main institution for steering EC economic policy. Given its autonomy, it would require legitimation through both the Commission and the member states’ governments and would have to work closely with both. A central bank must be recognized by the whole economic and political system on more than merely legal grounds. Thus the creation of a European central bank would be a decisive step not only in currency unification but also in creating a federal state. Although fundamental reform of the organizational structure is not initially required, the internal dynamics of the process will eventually make this appear imperative. There are already plans for a European government responsible to parliamentary control based


552

on the EC’s democratic legitimacy, the transformation of the Council of Ministers into a second chamber, and the establishment of an elected head of state. A fully sovereign European federal state is projected on the basis of an aggregation of the tasks of existing nation-states. The EU should evolve a comprehensive concept of the state along with a concept of European citizenship, its own constitutional rights and modes of parliamentary participation. Such a process would gradually lead to the functional displacement of the member states. The latter would acquire the status of the German Länder, while in Germany the Supreme Court would lose its authority and the Federal Office of Cartels would become “residual.” Something of the kind will also happen to all other member states. To assume that the system of interest mediation, decision-making and legitimate consensus in modern democratic societies is guaranteed by central government agencies is to minimize the state. This is not the case in any of the German Länder and it would not be the case in a united Europe. The project of an EU, reiterated in June 1990 and further negotiated in December of the same year by the heads of the member states, limits the problem to a reallocation of jurisdictions of the central agencies and so proceeds along the lines of the 19th century nation-state. On the one hand, the EU is seen as a comprehensive authority with reserve powers for the member states. Decisions with respect to currency, monetary and credit policy, foreign policy and security, social policy, regional policy, environmental and fiscal policy should be centralized. On the other hand, the rights of parliament should be limited and decisionmaking power should remain with the Council of Ministers. The governments of member states should continue to be “sovereign,” as were the constitutional monarchies in the mid-19th century. The European Parliament ends up with the role of all underprivileged chambers. It not only wants more participation but also decision-making power and control of the government. The more concrete the project of European unification becomes, the greater the possibility of a “constitutional conflict.” This can be seen in the Luxembourg government’s reactions to the treaty outlined on April 17, 1991. The more comprehensive the authority of European agencies, the greater the claim to complete democratization of these agencies, i.e., majority decision, parliamentary control of government, and parliamentary legislation. The “constitutional conflict” between the European Parliament and the Council of Ministers is the result of EC “dual sovereignty.” The community of sovereign states creates a supranational regime which gradually emancipates itself from the member states and constructs a “supranational sovereignty.” In principle this conflict is difficult to institutionalize so that “dual sovereignty” can be maintained indefinitely. The tendency is for one to either bloc or absorb the other. Today a European state is being created which presumably will prefigure a European nation.8 The homogeneity and identity deriving from the rights of a European citizenry could then also lead to the development of an institutional system that eventually would pave the way for supranational European decisions. According to the model of the European nation-state, that is the ultimate goal of an EU. It remains to be seen whether European integration can create a European nation modeled after the 17th and 18th century monarchies—whether it can create a homogeneous nation with sovereignty over a given territory. The US was able to constitute a nation by integrating immigrants who could not create their own institutions.9 In Europe, however, there are nation-states which for centuries have had their own

Beyond the Nation-State 553 cultural, social and political histories and institutional orders. In order to form a European nation out of these individual states they would have to be dismantled and reconstructed. In this respect, there should be no ambiguity about an EU, even if this is not realizable in the near future. It is counterproductive to occlude the implications of such an undertaking, even if this may be tactically useful. The federal option should be considered if there is to be a European state without a European nation. Not only should the nation-states be retained, but so should their respective jurisdictions in accordance with the principle of subsidiarity. Cultural affairs should be left to particular nationalities, as Germany does with the Länder. Nine national languages are spoken in the EC. A tenth, Catalan, has been recognized even though it is not the language of any particular state. This alone is a significant distinction between the EC and the US. Given such linguistic difference, which will no doubt increase, there is no possibility for the EC to develop the kind of cultural unity forcibly imposed by European nation-states over several centuries. On the one hand, such languages as English, French, Spanish and Portuguese are internationally recognized; on the other, English, which is the lingua franca of economic, technical and scientific commerce, is a foreign language to more than 80% of the EC population. If one assumes that in the future some 20% of this population will become competent in English or French, that still would exclude 50% from a common “European communication.” Thus Europe will still rely on national languages. Here the cultural autonomy of nation-states does not concern regional rights nor the preservation of dialects and folklore but rather the maintenance of communicative structures. The recognition of one’s own linguistic competence for all everyday communication was and is one of nationalism’s most powerful attractions shared by the cultural elite and the population at large. Full linguistic competence is the precondition for information and participation, i.e., a fundamental democratic prerequisite. Schooling is also an aspect of cultural autonomy because it facilitates an intergenerational transmission of cultural traditions. Since education imparts professional qualifications, it is tied to the labor market. For this reason, coordination of the European labor market with respect to freedom of movement has led the EC to regulate educational policies. The effects are still modest. Yet there are already conflicts between national autonomy and European coordination concerning educational charters and exchanges of students and scholars. Nevertheless, financial incentives have changed national goals. Because linguistic and educational autonomy are inseparable, no higher authority should interfere with either, despite the seeming implications of competition and freedom of movement. The nation-state is not merely a cultural entity but a complex system of structured interests, institutionalized conflicts and integrated distribution. National stability is based on the functioning of this system, which is only in part represented by governing agencies. This includes systems of wage and price agreements, trade unions, industrial associations, national tax and welfare policy, regional resource administration, forming public opinion, and legal and security systems. In Germany the Länder’s stability is based on their systems of interest formation and mediation. The goal of a common European currency to coordinate the member states’ budgetary and labor policies by financial-political means hinges on the “distinct national systems of wages and the distinct organizational structures of trade unions.”10 More generally, the


554

coordination of an economic and social order hinges on the structure of intermediate institutions which have hitherto been barely affected by EC integration. Political parties, trade unions, professional associations and the mass media are all organized along national lines and dependent on the nation-state. An intermediary “Europeanization” of this network of organizations would last decades, lead to substantial conflicts within the EC and prevent the development of a centralized federal regime. Many expect the EC to substitute for the inadequate intermediary as well as central political institutions of the nation-state. Dissatisfaction with national political systems and the seeming impossibility of reforming them (e.g., Italy) has led to demands for greater European authority supporting regionalism. Thus the EC is seen as a potential ally of attempts to reform inadequate national institutions. But were the EC to adopt such a role, there would be less internal political pressure to reform. The decision to substitute European for national institutions should remain at the national level, which would allow the international drive toward efficiency and increased competitiveness to encourage reform from within. National institutions generate a great deal of political participation and legitimation. Actual or potential participation which legitimates political systems is not just the result of political parties, elections and parliaments. This is also true in Germany, which requires a wider network of mediations. It would be impossible for some 500 deputies of the Bundestag to fulfill such a function in a democratic state. This is even truer of the some 500 deputies of the European Parliament. To attempt to mediate European politics vis-à-vis the citizenry and to represent this citizenry in European politics would overtax their abilities. The European Parliament can gain legitimation through direct election, but it can sustain a common European legitimation only through a complete unification of the structures of participation in nation-states. European legitimation must also involve the whole system of intermediary interests presently handled by economic and social managers outside democratic control mechanisms. A cooptation of experts leads to a coordination of functional elites but does not create political legitimacy. The argument for increasing European legitimacy with respect to decision-making is not only a matter of direct legitimation through representative institutions but also one of interest formation. In this respect, the allocation of jurisdictions and resources is decisive for political parties, interest groups and agencies of public opinion. Public opinion is constituted on various levels with a variety of procedures for mediating interests rather than simply through an internal principle of organization.11Thus German federalism is also a matter of party organizations, the formation of interest groups, autonomy in wages and prices, and the media. The Länder also play a role since they have the means and expertise to mediate interests, participation and distribution. What is crucial, however, is the development of the centralized mediation of vertically organized interests and authorities. This takes place at the level of the nation-state. Thus in order to create a system of legitimation in a united Europe, the balancing of interests must be a central function of the nation-states. Given the gap between various interests, which are promoted by certain bureaucratic organizations, the differences among interests and authorities vary vertically. However, the mediation and confrontation between interests and the reasons behind them must remain on the national level because this function cannot be duplicated on the European level. To date, the success of the EC

Beyond the Nation-State 555 has been a function of the fact that its political jurisdiction has been limited. It could vertically isolate itself from politics and thus guard against conflicts of interest, while the horizontal mediation of interests took place within the member states. The more the EC enters the political sphere, the greater its dependence on the member states. Thus there are limits to the evolution of a horizontal mediation of interests on the European level. A devolution on the regional level is also problematic. At times the centralization of European decision-making is seen in terms of a “Europe of regions,” even if some regions coincide with smaller EC nation-states: Ireland, Denmark, the Netherlands, Portugal, Greece and Luxembourg. All have relatively homogeneous internal economic and social systems. Thus the issue of regional sovereignty concerns only the larger states with regions modeled after the German Länder. The concept of a “Europe of regions” is thus essentially a program for the devolution of larger nationstates. But it is unlikely that these larger states can be divided into relatively homogeneous socio-economic entities such as Baden-Württemberg and Bavaria. This devolution could not link political spheres horizontally and democratically because the smaller allocation of resources and authority on the regional level would no longer allow the centralized horizontal mediation of differing standards and their political expression. A “Europe of regions” as a substitute for the traditional nation-states would mean that each region would have to have sufficient jurisdiction, revenue, and a centralized system of establishing interests to mediate conflicts and acquire legitimation. If this is not guaranteed, then there is no structural basis for a centralized democracy in an EU. In other words: a Land such as Bavaria would be the model for the devolution of the large nation-states into a reconstituted Europe of regional states. Sill unexplored is another concept of regions, namely those relatively homogeneous socio-economic areas of planning and development. In the 12 EC member states today there are 150 such regions that could be split into smaller units, yet they could not be the only supporting entities of European unity. Despite their independent jurisdiction, they would require nation-states to provide financial support and the necessary legitimacy. Suprastate agencies have developed precisely because European nation-states had difficulty in dealing separately with matters such as security, environment, the infrastructure, foreign policy, underdevelopment and economic policy.12 EC security interests are still not dealt with on a European level but rather within NATO. This is also true of states outside the EC, with the exception of Ireland, notwithstanding France’s special position in NATO. Thus security policy is already outside the EC proper. Similarly, if new EC members such as Austria, Switzerland, Finland and Sweden maintain their neutrality, they will have to be excluded from EC jurisdiction for security reasons. Since NATO includes both the US and Canada, it transcends Europe. Future security tasks are not likely to shrink. It is therefore unlikely that an EU can take over security policy. The tying of former Eastern bloc countries (including the former Soviet Union) to European security institutions enhances the significance of discussions in the CSCE. Lately they have begun to function as a “European Security Council” for conflict resolution. The EC is not in a position to determine autonomously European security policy because it is tied to a multiplicity of “governmental regulations.” Nation-states remain the agencies deciding security and foreign policy. As


556

long as they remain members of the UN, CSCE, NATO and other broader organizations, and as long as Great Britain and France are members of the UN Security Council, there can be no binding EC majority decisions in security and foreign policy. There are good reasons why foreign policy proper, unlike foreign economic policy, has not been coordinated in the EC. Along with the EC, the EPC is only a loose system surpassed by bilateral consultations, regular foreign ministers’ meetings, political exchanges within NATO, CSCE agreements, disarmament conferences, the so-called “10 plus” consultations on the UN level, and similar organizations. While the EC may coordinate its members’ foreign policy domestically, in foreign relations only the member states are sovereign subjects of international law. For this reason, internationally the representation of European interests falls to the nation-states. In environmental and developmental policy as well as foreign and security policy, the ability of an EU to act would be as limited in some areas as is that of the nation-states because some problems transcend EC territory as well as its ability to handle them. In view of the internal elasticity of electoral situations and in particular the possibility that large problems can be tackled within a number of smaller domains, there are advantages to a multiplicity of political regimes linked but not institutionally bound. It would be disadvantageous, however, if Europe did not eventually “speak with one voice.” If it does not transform its economic power into political influence, it will not be able to respond promptly to crisis situations. But this is not an end in itself. The EC is certainly not an “independent state” in foreign policy. There are various “external limits” to its autonomy. In view of the direct or indirect expansion of the EC to 20–25 full or associate members, the complexity of domestic interests is likely to increase these limitations.13 As things stand, Western Europe cannot be isolated from international developments. This situation would not change nor would the problems be simplified if the member states’ foreign policy and other political concerns were centralized within EC agencies. It does not follow, however, that most member states should not seek internal unanimity, but only that this will not be generated by EC mechanisms such as the European Parliament or the Council of Ministers. An accumulation of EC jurisdictions reduces its ability to function, increases its means of action (looking to the day when it has its own right of taxation) and sharpens legitimation problems. Broadening jurisdictions expands resources and increases legitimacy. EC success thus far has been predicated on a limitation of jurisdictions and on the coordination of competitive conditions and economic policies within the Common Market. This broad political sphere is crucial for EC economic development and that of its member states. It involves the deregulation of particular arrangements on the state level, their re-regulation on the European level, and the coordination of various particular national regulations. In this context, the development and above all the direct administration of individual programs of inter-European regional and social policy are problematic. They lead to the construction of an independent bureaucracy which, in view of limited resources, is unable to implement structural changes. Direct economic relations have more impact on regional structures and labor markets than bureaucratic interventions. Finally, the EC Grossraum is not economically self-sufficient in the world economy. Multinational industrial concerns bind international blocs and generally determine the technological-economic rate of integration.

Beyond the Nation-State 557 Although the powers of the nation-states in the EC have been substantially diminished, the latter has not displaced the former. Moreover, the tension between “necessary alliance” and “supranational regime” are not likely to disappear with the evolution of the EC into an EU. It is necessary to recognize this tension in undertaking institutional reform since such relations always bring into sharp focus various standards. Since the task is to develop a political regime that unites existing nation-states into a single entity, the sovereign nation-state cannot be a model for an EU. The multinational state could be such a model, although it lacks political traditions and an institutional order. In multinational states the democratic majority principle is not enough to deal with the interests of individual nationalities. Structurally nationalities are politically organized minorities isolated and unable to become majorities. This is why they need their rights. Both as citizens of a united federation and as part of it, the members of a multinational state have a dual interest in representation. While the part of the federation cannot represent a subordinate area, it represents the nationality vis-à-vis the federation as a whole. A principle of equality must obtain between nationalities, irrespective of size. 5 million Danes must be equal to 75 million Germans. There can be no such principle of equality in either a one- or two-chamber parliament without the veto right for every member state. The nationalities’ individual rights must be organized on the basis of autonomous national institutions requiring an allocation of jurisdictions between the nationalities within the federation and the federation of nationalities.14 EC members are not entering into an EU as nation-states, i.e., as subjects of international law; they merely cooperate in various organizations with different tasks— the UN, the CSCE, NATO, a European environmental agency, a European council on religion, etc. The coordination of these activities is the task of the European Council, which is not an EC agency but one in which all member states participate with equal voting rights and no majority rule. The European Council exercises its influence on the EC through the Council of Ministers. Its legitimation is based on the system of nationstates. Its parliament coordinates the various political spheres and mediates European politics with the nation-states, thus guaranteeing the rights of nationalities. In order to fulfill this function, national parliaments must be “Europeanized,” i.e., national interests must be defined within a European context. The central task of these parliaments is the horizontal mediation of political spheres vis-à-vis vertically organized bureaucracies and interest groups. Consequently, their European orientation should be strengthened and this task should not be left to the European Parliament. The EC is and remains a functionally limited “alliance.” In principle it is not a political regime above the nation-states. Thus it is not a “federal state” and does not need representative officials such as a European president. The Commission’s chairman suffices. Similarly, it does not need direct administration to implement programs such as regional or social funds. The monitoring of living standards in EC member states is the exclusive task of the national governments, which collectively maintain a European budget. To the extent that national bureaucracies do not coordinate standards, there is need for reform. But European officials should not assume these tasks. Such an organizational model contains tensions and is likely to succumb to institutionalized inefficiency. In both cases the key question is the standards by which any institutionalized order should be judged. Given the multiplicity of such standards,


558

conflicts between them are common with respect both to their range and validity. The degree to which such standards become externalized in other social structures and living conditions is also open to debate, as are the chances of counteracting anticipated adjustments. To this extent, institutional orders are more than the organization of decision-making processes. They define the validity of standards as well as conflicts between them. The EC is a supranational political regime that became institutionalized for internal and external economic reasons. Since this task involves other political spheres, conflicts are possible. The dominant standard determining all others is the efficiency of trade relations within a large domestic market: the raising of international competition, the innovative power and general political viability of European states, and the like. The result has been economic integration and internal peace, the decline of chauvinism, and the creation of a European identity. But there are other standards, such as social justice, cultural independence, political participation, social solidarity and national identity. A greater allocation of jurisdiction to the EC would expand the range of acceptable standards, which in turn would increase possibilities for conflict within EC agencies and complicate the constitution of a collective will. A lower allocation would homogenize the standards internally and externalize conflict. There are good reasons for assuming that EC success depends on the relative homogenization of standards for collective decisionmaking. The expansion of jurisdictions on the basis of these standards creates more impetus for accommodation on the part of member states. There are also good reasons for letting the member states make these accommodations themselves. Not only do they control the relevant agencies and the structures of legitimation on which European integration as a whole is based, they also constitute a necessary buffer to shield the EC from national resentments, regional crises and employment fluctuations. To the extent that they remain functionally viable, they at least partially contribute to EC legitimacy. The problem of forming and reforming EC institutions is not that the EC is an “unfinished federal state,” as Walter Hallstein has claimed, but that it involves the consolidation of a confederation via a supranational order differentiated and organized to function as a state. In this sense, the EC should remain as it is and not develop into an EU. Thus the differentiation of jurisdictions within the EC should be as exact and explicit as possible. Neither should the EC follow a “logic of interdependence” in political matters, nor should it be the result of the implicit and increasing reduction of the nationstates’ activities. The more the latter is the case, the less possible will it be to allocate responsibility. The more member governments defer responsibility to the EC, the greater the pressure on the EC to acquire the means to fulfill these expectations. The reduction of regional disparities and the equalization of living standards cannot be the responsibility of the EC. This political responsibility should remain with member states because they have the socio-political means to implement programs. They should retain the means to create necessary incentives for capital investments. It is important to recognize the conflicts which arise in the attempt to coordinate competition in particular domestic markets. The maxim should be: when the jurisdiction is in doubt, it should remain with the member states; when in dispute, it should be decided by the European Supreme Court. The relation of parliaments and governments in a reconstructed Europe is far more complex than the model provided by the nation-state.15 The European Parliament is not

Beyond the Nation-State 559 the parliament of Europe but rather the parliamentary component of the EC. It exists in a relation of tension with the Council of Ministers and retains full decision-making power. It will not be reduced to a second chamber because the member states’ sovereignty is guaranteed. To this extent EC subjects are not EC citizens but “citizens of the [Common] Market.” 16 The European Parliament guarantees these citizens’ participatory rights. By the same token, the Commission is no European government but an independent agency for the administration and development of the economic community. It fulfills this task by initiating legislation. This is why it is necessary to broaden and strengthen the role of national parliaments in European politics. Only recently was it possible for the Bundestag to establish a long overdue committee to deal with EC questions. The “Europeanization” of national parliaments promises a greater democratization of European politics than a “denationalization” of the European Parliament. Given the number of new states in the international arena, the control of governments by national parliaments is necessary for the preservation of democratic participation and control. The development of various legal circles and ruling orders, of agencies technically entrusted to establish norms, of new states, of economic, financial and entrepreneurial markets no longer organized on the basis of territorially homogeneous entities, make the reform of [national] parliaments and governments necessary for a reconstructed Europe.

Notes Translated by G.L.Ulmen from “Der europäische Nationalstaat: Erbe und Zukunft,” in M. Rainer Lepsius, Interessen. Ideen und Institutionen (Opladen: Westdeutscher Verlag, 1990), pp. 264–68 and “Nationalstaat oder Nationalitätenstaat als Modell für die Weiterentwicklung der Europäischen Gemeinschaft” in Staatswerdung Europas? Optionenfür eine Europäische Union, edited by Rudolph Wildenmann (Baden-Baden: Nomos Verlagsgesellschaft, 1991), pp. 19–40. 1. Heinrich von Moltke, General Director of the 23rd Commission of the Bergdorf Discussion Circle, said: “At present a decision with respect to the domestic market must be made every week.” Cf. “Wie geht es weiter mit den Deutschen in Europa?” in Protokoll 90 (April 29, 1990), pp. 80f. 2. Cf.H.-J.Hoffman and Uwe Feist, “Die Europawahl 1989—Eine klassische Nebenwahl?” in Aus Politik und Zeitgeschichte, Vol. 8, No. 43 (1989). 3. On the concept of “regime” as a normative and regulative system of decision-making with factual validity, cf.: Beate Kohler-Koch, Regime in den internationalen Beziehungen (Baden-Baden: Nomos Verlagsgesellschaft, 1989). 4. For a critique of Imperial Germany, see Max Weber, “Parliament and Government in a Reconstructed Germany (A Contribution to the Political Critique of Officialdom and Party Politics),” in Weber, Economy and Society: An Outline of Interpretive Sociology, edited by Guenther Roth and Claus Wittich (Berkeley, Los Angeles and London: University of California Press, 1978), Vol. 2, pp. 1381–1469. 5. Cf.H.-P.Ipsen, “Europäische Verfassung—nationalle Verfassung,” in Bitburger Gespräsche (Munich: Beck Verlag, 1987), pp. 37–53. 6. Cf. the constitutional proposal of the European Parliament of February 14, 1984, reprinted in Eine Verfassung für Europa, edited by J.Schwarze and R.Bieber (Baden-Baden: Nomos


7.

8. 9. 10. 11.

12.

13.

14.

15. 16.

560

Verlagsgesellschaft, 1984). For a critical review, see H.-P.Ipsen, “Utopisches im ParlamentsEntwurf einer Europäischen Union,” in Einigkeit und Recht und Freiheit: Festschrift für Karl Carstens zum 70. Geburtstag am 14. Dezember 1984, edited by Bodo Börner, Hermann Jahrreiss and Klaus Stern (Cologne: Heymanns Verlag, 1984). On this development, see F.W.Scharpf, “Die Politikverflechtungs-Falle: Europäische Integration und deutscher Föderalismus im Vergleich,” in Politische Vierteljahresschrift, No. 26 (1985), pp. 323–56. Cf.David Easton. The Political System (New York: Knopf, 1965), pp. 171ff. Cf.Seymour Martin Lipset, The First New Nation: The United States in Historical and Comparative Perspective (New York: Basic Books, 1963). Cf.Fritz W.Scharpf in Die Zeit (December 12, 1986). Cf.M.Rainer Lepsius, “Die Prägung der politischen Kultur der Bundesrepublik durch institutionelle Ordnungen,” in Lepsius, Interessen, Ideen und Institutionen, op. cit., pp. 63– 84. Cf.D.Biehl. “Regionalle Entwicklungen und Regionalpolitik in der Gemeinschatt aus föderalistischer Perspektive,” in Plädoyer für Europa Beiträge zur europäischen Einigung, ed. by Astrid Schomaker, Daniel Gossel and Jens Lehnigk (Hamburg: Verlag Weltarchiv, 1989). Turkey, Malta, Cyprus, Austria and Greece have already applied for full membership, while negotiations are in progress to grant associate membership to the states of the former European Free Trade Association: Iceland, Norway, Finland, Sweden and Lichtenstein. Applications for associated membership are also expected from East European countries and even the newly independent parts of Yugoslavia. Cf.Emerich K.Francis, Ethnos und Demos Sociologische Beiträge zur Volkstheorie (Berlin: Duncker und Humblot, 1965); and M.Rainer Lepsius, “‘Ethnos oder Demos’: zur Anwendung zweier Kategorien von Emerich Francis auf das nationalle Selbstverständnis der Bundesrepublik auf die Europäische Einigung,” in Lepsius, Interessen, Ideen und Institutionen, op. cit. Cf.Weber, “Parliament and Government in a Reconstructed Germany,” in Weber, Economy and Society , op. cit. Cf.H.-P.Ipsen, “Europäische Verfassung—nationale Verfassung,” in Bitburger Gespräche, op. cit., pp. 37–53.

The State, Volume 3 (Critical Concepts)

Buddhism: Critical Concepts in Religious - Volume 3

HISTORICAL JESUS, VOLUME 3 (Critical Concepts in Religious Studies)

Buddhism: Critical Concepts in Religious - Volume 4

Buddhism: Critical Concepts in Religious - Volume 1

Buddhism: Critical Concepts in Religious - Volume 5

Buddhism: Critical Concepts in Religious - Volume 6

Buddhism: Critical Concepts in Religious - Volume 7

Buddhism: Critical Concepts in Religious - Volume 8

Buddhism: Critical Concepts in Religious - Volume 2

Ulysses: A Critical and Synoptic Edition (volume 3 of 3)

Cybercultures: Critical Concepts in Media and Cultural Studies Vol. 3

HISTORICAL JESUS, VOLUME 1 (Critical Concepts in Religious Studies)

HISTORICAL JESUS, VOLUME 4 (Critical Concepts in Religious Studies)

HISTORICAL JESUS, VOLUME 2 (Critical Concepts in Religious Studies)

3 Games of State

3 - Games Of State

The Highwayman, Volume 3

William Shakespeare: The Critical Heritage Volume 3 1733-1752 (The Collected Critical Heritage : William Shakespeare)

Volume 3

Volume 3

Volume 3

The City: Critical Concepts in the Social Sciences. Volume 3: Economic Structure and Change in the Western City

Clinical Technologies: Concepts, Methodologies, Tools and Applications (3 Volume Set)

Ulysses: A Critical and Synoptic Edition (volume 2 of 3)

The City: Critical Concepts in the Social Sciences. Volume 5: Cities in the Third World

Concepts Of Theoretical Solid State Physics

Contemporary Political Concepts: A Critical Introduction

Myeloproliferative Neoplasms: Critical Concepts and Management

Critical Security Studies: Concepts and Cases (Borderlines)

The Early Asimov. Volume 3

The State, Volume 3 (Critical Concepts)

Buddhism: Critical Concepts in Religious - Volume 3

HISTORICAL JESUS, VOLUME 3 (Critical Concepts in Religious Studies)

Buddhism: Critical Concepts in Religious - Volume 4

Buddhism: Critical Concepts in Religious - Volume 1

Buddhism: Critical Concepts in Religious - Volume 5

Buddhism: Critical Concepts in Religious - Volume 6

Buddhism: Critical Concepts in Religious - Volume 7

Buddhism: Critical Concepts in Religious - Volume 8

Buddhism: Critical Concepts in Religious - Volume 2

Ulysses: A Critical and Synoptic Edition (volume 3 of 3)

Cybercultures: Critical Concepts in Media and Cultural Studies Vol. 3

HISTORICAL JESUS, VOLUME 1 (Critical Concepts in Religious Studies)

HISTORICAL JESUS, VOLUME 4 (Critical Concepts in Religious Studies)

HISTORICAL JESUS, VOLUME 2 (Critical Concepts in Religious Studies)

3 Games of State

3 - Games Of State

The Highwayman, Volume 3

William Shakespeare: The Critical Heritage Volume 3 1733-1752 (The Collected Critical Heritage : William Shakespeare)

Volume 3

Volume 3

Volume 3

The City: Critical Concepts in the Social Sciences. Volume 3: Economic Structure and Change in the Western City

Clinical Technologies: Concepts, Methodologies, Tools and Applications (3 Volume Set)

Ulysses: A Critical and Synoptic Edition (volume 2 of 3)

The City: Critical Concepts in the Social Sciences. Volume 5: Cities in the Third World

Concepts Of Theoretical Solid State Physics

Contemporary Political Concepts: A Critical Introduction

Myeloproliferative Neoplasms: Critical Concepts and Management

Critical Security Studies: Concepts and Cases (Borderlines)

The Early Asimov. Volume 3

Recommend Documents