Capital & Class. - 1978. - Issue 5

EDITORIAL This year's CSE Conference is a logical follow-on from that of 1976 on the Labour Process and that of 1977 on the State . Its theme is 'The Crisis of Capitalism and Working Class Strategies in the 1970s'. The conference will take up this theme in working sessions covering such areas as State Expenditure, Industry, Women, Education, Law, Nationalism, Science, Money, Community Politics, Housing, Social Policy, Ideology, Southern Africa as well as at a more general level . Many of the sessions will emerge from the work of CSE working and local groups that has been going on over the year and that is comprising a growing amount of CSE activity . Papers at the conference are only meant to be a basis for discussion in working sessions . A major emphasis of the conference will be the attempt to integrate the problems of specific strategies into a general framework . The experience of the various CSE Industry Groups is in this respect indicative . For most of them the need to move from the specific to the general level has been prompted by three issues : the use-value/value debate ; the forms of crisis and accumulation ; and the multidimensional forms taken by the reproduction of the capital-labour relation . We hope that discussion within the CSE can contribute to the debate on strategies through the development of coherent materialist analyses . In order to stimulate this debate we would like to encourage contributions for a new regular section in Capital and Class devoted to 'Working Class Strategies and Struggles' . We also hope that the development of these discussions within the CSE will provide a basis on which to achieve a closer integration of Marxist theory and the analysis of concrete situations . Some readers have felt that Capital and Class is too academic . If 'academic' is understood to mean 'politically irrelevant, we disagree with this . We feel that debates on such subjects as value theory (Armstrong et al and Rubin in this issue) and on class fractions (Clarke in this issue) are politically important because they are fundamental to an understanding of social relations in capitalism and of their historical limitations . We are aware, however, of the two related problems of making the analyses more accessible and of drawing out their political implications more explicitly . There are limits to what the Editorial Committee on its own can achieve in these respects . We feel that we and our contributors need to be more sensi-

tive to the need to write as clearly and simply as possible to

avoid confining discussion of crucial theoretical issues to the initiated . We are also particularly keen to see more contributions that offer concrete analysis . One barrier to accessibility is the (perhaps inevitable) extent to which authors refer in their articles to debates and positions which may not be familiar to their readers . To go some way towards meeting this particular problem we are introducing with this issue a new `Survey' section, with the aim of introducing readers to debates that have been taking place either in areas of specialised interest or more generally within the CSE but in the years before the current expansion of CSE membership and activity . We would particularly welcome suggestions for or contributions to this section . We are fully aware that our new sections can, at best, go only part of the way to solving these problems and the development of Capital and Class and its relationship to other CSE activity must be an important topic for discussion at the conference .

IN DEFENCE OF VALUE A REPLY TO IAN STEEDMAN An extended reivew of Marx after Sraffa, Ian Steed man London : New Left Books 1977

Philip Armstrong, Andrew Glyn, John Harrison Readers of the Bulletin will be aware of considerable past debate within the CSE on the definition and status of certain value categories . In Marx after Sraffa Ian Steedman argues, in part on the basis of points made in that discussion - in which he took a leading part - that all value categories constitute an obstacle to a scientific analysis of capitalism . This review article attempts a reply to his critique of value . INTRODUCTION The central argument of

Marx after Sraffa is

that :

"the project of providing a materialist account of capitalist societies is dependent on Marx's value magnitude analysis only in the negative sense that continued adherence to the latter is a major fetter on the development of the former" . (Marx after Sraffa p.207 henceforth simply 207) . The case for this 'Sraffa-based critique of Marx' (23) is argued on two levels . One level of argument is that certain propositions employing value magnitudes and advanced by Marx are wrong . Steedman discusses two groups of such propositions : those concerning the relations which hold at any point of time between the rate of exploitation and the value composition of capital on the one hand and the equilibrium rate of profit and prices of production on the other (the transformation problem), and those concerning tendential movements in certain value magnitudes and in the rate of profit and relationships between them (the tendency for the rate of profit to fall) . His objections at this level concern the logic of particular arguments put forward within a certain theoretical framework . The other level of argument concerns the status of the theoretical



2

CAPITAL AND CLASS

framework itself . Steedman's position here is that a central concept employed - value - is unsatisfactory ; that it is a bad abstraction, an unscientific concept . In presenting his case for rejecting the concept of value Steedman considers a number of (increasingly complex) models of an ' "abstract" capitalist economy' (16) . In each case he is concerned with a similar set of problems . These hinge around the rate of profit and set of relative prices which would exist if all capitals appropriated the same rate of profit . He approaches the determination of these magnitudes by means of a ' "physical quantities" framework of analysis' (27) . The latter is essentially a specification of the 'conditions of production' (14) of commodities in terms of physical inputs and labour performed per unit of output . Thus a specification of the conditions of production of commodity A would be that N workers using 0 units of commodity B, P units of -commodity C and so on produce Q units of commodity A in a specified period . Sometimes the actual conditions of production used form the starting point . Sometimes all known possible ways of producing all known possible commodities must be specified . In each case Steedman is centrally concerned to show three things : (1) That a full specification of the conditions of production and the real wage is sufficient to permit a unique determination of the rate of profit and relative prices which would exist if all capitals appropriated the same rate of profit . (2) That value magnitudes are at best superfluous to such an exercise (the simplest case) and at worst either indeterminate or, where determinate, sometimes negative (more complex cases) . (3) That it is possible to demonstrate that the performance of surplus labour is a necessary and sufficient condition for profits to be appropriated without reference to value magnitudes (the simplest case) and that difficulties can occur with the conventional concept of the rate of exploitation (more complex cases) . This review approaches Steedman's overall argument in three broad stages . Section one looks at the indeterminate or negative values Steedman obtains and the problems in defining the rate of exploitation which he raises in examining more complex cases . Section two considers the more general argument that, even in the simplest case, value categories have no useful role to play . Section three discusses the concept of exploitation, the transformation problem and the question of tendential movements in the rate of profit . Before discussing these questions it is worth briefly noting the importance of Marx after Sraffa . First, it is certain to be an influential book . It contains the fullest statement to date of a school of thought which has considerable currency within British Marxism . The argument is extended in certain new directions and its implications - at least as understood by one of its leading proponents - are spelled out more fully than previously . Since Steedman has succeeded in explaining the meaning of many of his mathematical results in a simple and intuitive way, the book should make the issues with which this school is concerned more accessible than hitherto .



IN DEFENCE OF VALUE

3

Second, Marx after Sraffa raises some important questions . While what follows will take issue with Steedman on many points it should nevertheless be clear that we share a certain amount of common ground with him . While a full discussion of the nature and status of value categories lies outside the scope of this review, it should be clear that we agree with Steedman that such categories possess a quantitative dimension and hence that difficulties in defining their magnitudes are legitimate problems to raise . We regard Steedman's demolition of the 'obscurantist response' (21) of many of his previous critics, whose work has been characterised by a more or less total evasion of such problems, as both justified and effective . Ritual denunciations of 'neo-Ricardianism' garnished with yet another banal reminder that capital is a social relation simply will not do . INDETERMINATE OR NEGATIVE VALUES AND PROBLEMS WITH THE RATE OF EXPLOITATION The value of a commodity is the quantity of socially necessary labour required to produce it . In the simple case - where all actual labour is socially necessary, only one unchanging technique for the production of each commodity exists and there are no joint products or fixed capital - the value of any commodity is clearly defined . It can be conceived of in three ways . (a) As a sequence of dated labour . Thus the value of a refrigerator, for example, can be thought of as the sum of the labour required to mine a certain quantity of iron ore in one production period, to convert the ore into still in the next period and to shape the steel into a refrigerator in the final period . (b) As the labour required to produce a net output consisiting of one unit of one commodity only . Thus the value of a refrigerator can be thought of as the total labour required in a 'sub-set' of industries which produce, simultaneously, iron ore, steel and refrigerators in proportions such (1) that the output of all industries except that producing refrigerators is equal to the input of those commodities into the other industries considered while (2) the output of refrigerators exceeds by one unit the number of refrigerators used as inputs into the sub-set of industries . (c) As the sum of the labour required to produce non-labour inputs and the direct labour performed transforming these inputs into the final commodity. So far as conceptualising the notion of the amount of labour required to produce a single commodity is concerned, this third way of thinking about values is only a less-developed version of method (a) above . It is listed separately here because, when considering the values of all commodities, the notion that the value of output of each industry must equal the value of non-labour inputs to and direct labour performed within it is in principle sufficient to calculate the value of each and every commodity produced . Indeed, the conventional method of determining values is to solve a set of simultaneous equations which expresses this necessary equality between the value of inputs plus direct labour performed and the value of output for each and every industry .

CAPITAL AND CLASS These three ways of conceiving of value are all equivalent in the simple case : they all correspond conceptually to the basic notion that the value of a commodity is the socially necessary labour required to produce it and they all yield the same numerical results . In the more complex cases Steedman considers, the notion of the amount of socially necessary labour required to produce a commodity is no longer unambiguous . Hence different ways of conceiving of value, which coincide in the simple case, may on occasions diverge conceptually and yield different numerical results . This type of situation is often encountered in science . It is a general feature of reality that phenomena which invariably accompany each other in certain restricted circumstances, and hence can often be subsumed in a single concept for purposes of analysis, diverge under less restricted conditions, and hence nesessitate some conceptual reworking if more complex conditions are to be understood . Approached constructively, the process of encountering conceptual ambiguity when analysing more complex cases and, as a result, refining categories and choosing between newly divergent definitions can play an important role in the development of knowledge . Unfortunately, Steedman approaches the matter in a purely negative way ; ambiguities are seen simply as further evidence for the need to ditch value categories altogether . If the problems he raises are to contribute to our understanding of capitalism, they must be rescued from this cursury and one-sided treatment . Steedman considers three more complex cases : ~1~ Where choice of technique exists . 2 Where joint production occurs . 3 Where fixed capital exists . Case (3) can be treated as a particular example of case (2), and case (1) can be considered with or without joint production . But it is initially convenient to examine cases (1) - (3) separately and in the above order . Choice of Technique If techniques chosen were always those which minimised the sum of direct and indirect labour input per unit of output then two methods of production would only be used simultaneously if they required the same total labour input per unit of output . So the value of the commodity would be the same whichever method was used, and hence choice of technique would present no problems for defining values . But techniques used in a capitalist economy are those which minimise costs per unit of output ; in other words, those which minimise indirect plus the paid part only of direct labour, both enumerated in price terms . Thus two or more techniques requiring different amounts of total labour per unit of output may be equally profitable and hence in use simultaneously . If this happens, the equilibrium rate of profit and relative prices will be the same whichever techniques are used but the values of commodities will not . Steedman concludes from this fact that the value of commodities may be indeterminate once choice of technique is allowed for . The scientific approach is to ask initially if there is a satisfactory way of attributing values to commodities in such circumstances . Consider two possible



IN DEFENCE OF VALUE

5

methods, each of which gives the normal answer in the simple case of no choice of technique . The socially necessary iabour required to produce a commodity could be thought of as : (1) that required by the technique which, of all those known to be possible, minimises total labour input . (2) that required on average given the actual combination of techniques used . We evaluate these approaches on the basis of their implications for the concept of the rate of exploitation . We adopt this criterion because we believe, in company with Steedman, that surplus value and the value of labour power are more useful categories for analysing capitalism than are the values of individual commodities. The rate of exploitation is normally thought of as the ratio of the quantity of social labour engaged in producing the surplus product to the quantity engaged in producing wage goods . Where one technique only is available for the production of each commodity, the amount of labour actually engaged in producing wage goods (or the surplus product) coincides with the amount which is required to produce those goods given technological possibilities . Once choice of technique is introduced, these two aspects of the simple notion of exploitation diverge . Hence only one can be maintained . Approach (1) above represents an interpretation of technical requirements while approach (2) reflects the actual situation . Approach (1) is the method proposed by Morishima for cases involving joint production (Morishima 1973 p616) . Steedman advocates this approach in such cases and, at least in one place, also appears to do so for the case of choice of technique without joint production (194 note 13, quoted in section three below) . But method (1) is seriously flawed . It diverges not only from the actual situation but also from any meaningful conception of technical possibilities . The techniques which would minimise labour input may involve a very high ratio of dead to living labour and thus be, in fact, impossible to use because insufficient past accumulation has occurred . We consider this objection to be sufficiently powerful to justify rejection of this approach . It is simply not helpful to define the value of labour power as the minimum work that would be required to produce the wage bundle if there were a hundred times more nuclear power stations, computer controlled machine tools and so on than in fact there are . Steedman's objection to method (2) is that, on this approach, an increase in the real wage may lead to an increase in the rate of exploitation, even though known possible methods of production do not change, because a change in the combination of methods used may lead to an increase in net output more than sufficient to offset the effect on the rate of exploitation of the rise in the real wage . Thus, he argues, 'it would seem that this measure does not adequately capture the full intuitive meaning of the concept of exploitation' (109) . We do not consider this to be a serious objection . It is counter-intuitive only if you think that the 'technical possibilities' and 'actual proportion of labour employed' aspects of the rate of exploitation must always coincide, as they do in the simple case . But once a range of techniques exists,



6

CAPITAL AND CLASS

between which capitalists choose on the basis of profit maximisation rather than of labour-input minimisation, the two aspects will in general diverge . Method (2) is an appropriate conceptualisation of the actual state of affairs at any point in time . This is the more fundamental aspect of the concept of the rate of exploitation, as Steedman implicitly admits when he says that the rate of exploitation must take into account the relation between capital and labour at a given time . Method (2) also yields determinate values for all commodities . The easier way to conceptualise this is to think of the combination of techniques used as a single hypothetical process, the value specifications of which consist of a weighted average of the value specifications of the different techniques acually used . The value of the commodity in question can then be thought of in any of the three ways outlined at the start of Section I ; they all coincide, as in the simple case (1) . joint Products It is obviously difficult to attribute values to individual commodities when joint production occurs . There is no simple way of conceptualising the amount of labour required to produce commodity x only when it is necessarily produced in conjunction with at least one other commodity . In the simple case of no joint products, the easiest way to determine individual values is to solve a set of simultaneous equations which expresses the fact that the value of non-labour inputs plus direct labour performed in each industry necessarily equals the value of output . If this method is used in the joint production case, the results may include some indeterminate and some negative values . (2) An implication of this possibility is that any aggregate value category may be negative, since commodities with negative values may predominate in the bundle under consideration . Hence Steedman's well-known example of positive profits with negative surplus value . In Marx after Sraffa he also provides an example with a negative value for constant capital (each employee being equipped with machinery which requires a negative amount of labour in its production) and, implicitly, one with a negative value of labour power (the working class needs to work a negative amount of time to produce its consumption requirements) . Had he chosen to emphasise these latter two examples rather than the first, however, more people might have suspected that something was wrong with his method of valuation, rather than worrying about whether surplus value is really the source of profit after all . To consider the issues involved it is convenient to reproduce and extend part of Steedman's figure II (160) (Readers unused to handling diagrams may find the following four paragraphs difficult . It is possible to skip them without losing the general thread of the argument .) Points A and B represent the net amounts of commodities 1 and 2 produced by unit labour when processes a and b respectively are used . All

points on the line AB between A and B represent combinations of commodities 1 and 2 which could be produced by having some labour operating process a and the rest operating process b . Points on the extension of the

IN DEFENCE OF LABOUR

7

3 commodity 2

,C

commodity 1

line AB (i .e . points which do not lie between A and B) do not represent possible combinations of output . In particular, point Q - with output of half a unit of commodity 2 and no commodity 1 - is not possible . However, it is possible (if somewhat counter-intuitive) to intepret point Q as representing the net output that would result from 1 1/2 units of labour operating process a and minus /1 2 a unit operating process b. If point Q is thought of in this way, OQ represents the amount of commodity 2 produced by unit labour and 1 /OQ is therefore the value of commodity 2 . Similarly, 1/OP (minus 1 in this example) is the value of commodity 1 . These are the results that would be obtained by solving the normal set of simultaneous equations to derive the values of commodities . Steedman refers to them simply as "the values" of commodities 1 and 2 . (3) If point W1 represents the real wage and the line Owl intersects the line AB at R1 then R1 represents the maximum amount of the specific combination of commodities constituting the wage bundle that could be produced . The fact that Wi is on the same side of R1 as the origin indicates that workers receive less than the maximum output possible and thus that the rate of exploitation - given by OW11 - is positive . ; If W and R are such that R does not lie between points A and B then, to produce exactly the bundle of commodities represented by point R, more than unit labour would be required with technique a and a negative amount of labour with technique b . In this situation, if W is on the other side of R from the origin, surplus value as defined by Steedman is negative . (4)



8

CAPITAL AND CLASS

This is obviously a meaningless interpretation of the mathematics . The real situation is that the exact combination of commodities represented by the points R or W cannot be produced : production of the required amount of commodity 2 involves the production of a surplus of commodity 1 . (5) The only sensible conclusion to be drawn is that values cannot be attributed to individual commodities in this case . This becomes obvious if you try and conceptualise value in any deeper way than as the solutions obtained by solving simultaneous equations . With joint production it is in principle impossible to calculate the stream of dated labour required to

produce a unit of the commodity in question by itself or to isolate a subset of industries the net output of which would consist of a unit of this commodity only . Steedman's results indicate not that values can in fact be negative but rather that a method of thinking about, and in principle calculating, values which works fine in the simple case just does not work with joint products . What is the implication of joint production for the rate of exploitation? For Steedman, it is obviously that the rate of exploitation as normally conceived need not be positive for profit to be appropriated (since the surplus product may have positive prices but embody a negative amount of value) . This position clearly falls if his negative values are rejected . But a problem remains . How can the wage bundle and surplus product be valued when the individual values of the commodities comprising them cannot be calculated? Steedman himself puts forward one suggestion : the value of labour power should be seen as the minimum labour required to

produces the wage bundle using any combination of known possible production methods. In the diagram, the labour required to produce wage goods is minimised by using process c, a process which is not actually used because it does not yield the going rate of profit . To produce the wage bundle W, it is necessary also to produce the bundle of commodities represented by point D . Total output is represented by point C . The value of labour power is thus proportional to the distance OD, surplus value to DC, and the rate of exploitation to DC/OD . This is equivalent to approach (1) in the choice of techniques case . The objection raised in that context applies equally here . A more satisfactory approach is one equivalent to method (2) in the

choice of techniques case . Production processes actually used are divided into groups . All processes producing a particular commodity are placed in the same group and hence all processes producing another product jointly with this commodity are also in the group . If a particular commodity is not produced jointly with any other, the group consists only of the techniques used to produce this commodity. Each group is then reduced in size to the minimum consistent with the production of at least the wage bundle . Thus a group producing only capitalist consumption goods is eliminated, one producing materials used equally in the production of wage goods and capitalist consumption goods is halved and so on . Within each group all processes are reduced in the same proportion . The total labour required after scaling down is the value of labour power .



IN DEFENCE OF VALUE

9

This method is best grasped intuitively by reference to the 'industry sub-set' approach to value in the simple case . Here, rather than having a sub-set for the production of each commodity - which is not possible with joint production - we have a sub-set for the production of the wage bundle together with as little else as is possible . In the diagram, if actual production is represented by point E then point F represents the minimum output consistent with the output of every commodity being at least sufficient to make up the wage bundle given the actual proportions in which processes a and b are used . So the rate of exploitation is EF/OF . The method is not perfect . Since some goods over and above the wage bundle may still be produced in the scaled-down system, it can be argued that this approach tends systematically to underestimate the rate of exploitation (though by nothing like the extent to which the Morishima/ Steedman approach tends to overestimate it) . Further, if joint production is sufficiently interlinked, it will not be possible to break the system down into groups in the way outlined above . The simplest example of a case in which this would be so is a system containing two commodities only one wage good and one capitalist's consumption good - which must be produced jointly . Here no scaling down whatsoever would be possible . Clearly if joint production were quantitatively important, the values of a significant number of individual commodities would be indeterminate . If joint production were not only quantitatively important but also significantly interlinked between the production of wage goods and commodities comprising the surplus product then the concept of the rate of exploitation would also lack precision (since, in this situation, necessary labour would include on our definition the production of a significant amount of the surplus) . This would be more worrying than would the indeterminacy of individual commodity values . But note that it is only the concept of the rate of exploitation that would lose precision, the concept of exploitation itself would not be damaged . Further, it is far from obvious that such interlinked joint production is in fact a significant feature of capitalist economies . Certainly Steedman presents no evidence on this score . The only real argument for the empirical importance of joint production offered by Steedman is the claim that any process involving fixed capital must be treated as one of joint production . But, as the next subsection shows, many of the problems encountered in analysing 'pure' joint production do not arise with fixed capital .

Fixed Capital The value of means and objects of labour is transferred to the product . Since means of labour last longer than one production period, only a part of their value is transferred in each period . This basic idea is independent of any particular way of attributing a magnitude to the value transferred in any given period . In order to put this fundamental point across simply, Marx usually assumed that machines retain equal efficiency throughout their lives, so that the same number of use value is produced in each period of their operation . In this situation, a constant proportion of a machine's value is transferred both to each use value and in each period - value transference is linear with respect, both to use value production and to time . This is



10

CAPITAL AND CLASS

a perfectly reasonable starting point ; most machines are in fact scrapped because of inefficiency relative to new types of machines rather than relative to their own performance when new. If the efficiency of machines varies with age, however, so that different amounts of use values are produced by their operation in different periods, linearity with respect to use-value production diverges from linearity with respect to time . In this situation, the obvious approach to adopt is to maintain linearity with respect to use value production, which is clearly the more basic notion, and abandon linearity with respect to time . But a problem remains . If the operation of a machine by a constant amount of direct labour creates less use-values in some periods than in others, it is not possible to maintain both the notion of linear value transference (with respect to use values) and that of all direct labour being socially necessary (and hence creating equal amounts of value) . For if the same amount of value were transferred to each use-value and the same amount of value added in each period, the total value of each unit of the commodity produced on some machines would exceed that of each unit produced on others . This would be at odds with the fundamental idea that all units of a commodity have equal values . Steedman's response to this problem is to drop the idea that the same amount of value is transferred to each unit of output in order to maintain the idea that all direct labour creates the same value . He then finds that peculiar things can happen . If the difference in productivity of labour operating machines of different ages is sufficiently large, both value transferred in certain periods and even the value of certain partly worn out machines may be negative . This follows from the fact that value transferred is conceived of simply as the difference between value added (which is the same on all ages of machine) and total value (which is proportional to the number of use values produced) . Consider the following example . Suppose that one unit of labour, working with a machine requiring one unit of labour for its production, creates 29 use-values in the first year of the machine's life, one use-value only in the second year and none thereafter. The value of one unit of output is 1/10 (30 units are produced in all, embodying a total of three units of labour - 2 direct and 1 indirect) . The value of output in the first year is 29/10 . Value added is 1 . Value transferred is therefore 19/10 . The value of the one year old machine is thus minus 9/10 . This is also the value transferred in year two (value added is 1 and the value of output 1/10) . How damning for value categories are these results of Steedman's? Not very . Firstly, the values of all newly produced commodities are determinate and positive . It is not obvious that the values of partly worn out machines are of any interest . (6) . Secondly, in this case, Steedman's negative values do express something real, even if in a somewhat counter-intuitive way . This can be seen most easily in the case in which machines are less efficient in the early stages of use (the phenomenon of 'running in') . Here it is intuitively reasonable that the machine should have a higher value in year two than in



IN DEFENCE OF VALUE

11

year one, reflecting the fact that its efficiency is initially improved by use . Steedman obtains this result . But rather than considering that some labour operating the machine in year one adds value to it and that the positive effect of this on the value of the machine exceeds the negative effect of value transference (intuitively reasonable), his method of treatment implies that a negative quantity of value is transferred during year one (intuitively unreasonable) . Where machines lose efficiency with age, the aspect of reality which is reflected by Steedman's negative values is the fact that the operation of old machines may be wasteful of labour . This is clear in the example given above : if machines were used for one year only, the value of the commodity would be 2/29 rather than 1/10 . Thus in an economy in which labour input were minimised, machines would be scrapped after one year . It is nevertheless profitable for capitalists to keep machines in operation for two years so long as the per capita real wage consists of less than one commodity . But understanding this in terms of negative value transference is difficult, to put it mildly . This leads on to the final point about Steedman's results for fixed capital - it is possible to devise an alternative approach which avoids negative values . This approach consists of dropping the idea that all direct labour creates the same amount of value, in order to maintain that of linear value transference (with respect to use values) . With this method both value transferred and value added are proportional to the number of use values produced . Thus both magnitudes are always positive, as are the values of all partly worn out machines . Thus, in the earlier example, value transferred in year one is 29/30, value created is 58/30 and the value of output is 29/10 . In year two value transferred is 1/30, value created is 2/30 and the value of output is 1 /10 . The objection that could be raised to this approach is that all direct labour should be considered to be socially necessary and hence should be regarded as creating equal amounts of value . In the example given above there is clearly a strong sense in which labour working one year old machines should not be regarded as socially necessary ; the use of such machines is wasteful of social labour . But the example is an extreme one . If the reduction in efficiency were small, the (direct plus indirect) labour time required to produce the commodity would be minimised by operating such machines rather than by scrapping them . Yet under this approach the labour operating one year old machines would be considered to add less value in a year than would labour equipped with new machines . The fundamental point is that, once again, aspects of the concept of socially necessary labour which coincide in the simple case may diverge in more complex cases . Here the divergence is between the 'non-wasteful' and 'equally efficient' aspects of the simple notion . Steedman maintains the first at the expense of the second and is thereby compelled to treat fixed capital in a counter-intuitive way . The alternative is to maintain the second at the expense of the first . This involves treating labour operating old machines, which it would be socially wasteful to scrap, as creating less value than that operating new machines, but it permits a straightforward and intuitive treatment of fixed capital .



12

CAPITAL AND CLASS

Weaknesses in Steedman's Approach We have argued above that the difficulties for value categories which Steedman raises are less serious than he would have us believe . In this subsection we change tack somewhat and examine Steedman's alternative approach in the context of these more complex cases . We argue that, even considered purely within the terms of the problems which it explicitly sets out to solve, this approach is less theoretically robust than he suggests . A major difficulty in coming to grips with Steedman's alternative framework is that he switches rather too cavalierly between Sraffa's approach and Von Neumann's Take Sraffa first . In his approach there are in general n production processes and n commodities all of which have positive prices at the given real wage . If the wage were to change without either the commodities produced or the processes used changing, some prices might become negative . Sraffa interpreted this as follows : "This conclusion is not in itself very startling . All that it implies is that, although in actual fact all prices were positive, a change in the wage might create a situation the logic of which required some of the prices to turn negative ; and this being unacceptable, those among the methods of production that gave rise to such a result would be discarded to make room for others which in the new situation were consistent with positive prices ." (Sraffa 1972 p .59) Steedman has pointed out another possibility : the method used to produce each use-value might remain unchanged but the commodities produced (i .e . products with positive prices) might change . A combination of the two possibilities might also occur . (Steedman 1976 p .873-6) Thus, while Sraffa's approach is able to determine the equilibrium rate of profit and prices given a set of n processes and n commodities, it leaves open the question of how the processes and commodities in existence at any point of time are determined . This is where Von Neumann comes in . Beginning from all known possible methods of production and the real wage, Von Neumann seeks to determine not only the equilibrium rate of profit and prices but also the set of actual processes used and commodities produced . On the basis of Von Neumann's results, Steedman concludes that : "On the basis of the physically specified available methods of production and circulation and the real wage bundle, together with the assumption that capitalists accumulate all their profit, that 'overproduced' products are zero priced and that competitive conditions prevail, the Von Neumann analysis can be used to show that the rates of profit and growth are equal and uniquely determined and to 'determine' (though perhaps not uniquely) the social allocation of labour-time and the 'production prices' of all 'commodities' " (197) This is not very satisfactory . Firstly, a non-unique determination is a strange kind of determination . A more accurate description would be 'not entirely indeterminate' and, if there are a thousand or so possible sets of



IN DEFENCE OF VALUE

13

labour allocations and prices, indeterminate-in-inverted-commas would be a better shorthand than determinate-in-inverted-commas . Secondly, Von Neumann and Steedman's assumptions are extremely restrictive viz a fixed bundle of commodities comprising the real wage, no capitalist consumption and no workers' saving . It may be objected that these assumptions are standard in Marxist work . It is true that they form the normal starting point for an analysis of accumulation, but this is not what Steedman is concerned with . Marxist work on the determination of the equilibrium rate of profit and relative prices does not normally invoke these assumptions . This is because an absence of joint production is normally assumed, in which case both values and prices are independent of these restrictive assumptions . The point is that relative prices and the rate of profit are not independent of these assumptions once joint production is allowed for. Hence the assumptions constitute a far more violent abstraction when examining the problems with which Steedman is concerned in the joint production case than they do in such normal Marxist work as invokes them . A modified version of Steedman's example in Chapter 11 may serve to illustrate this non-independence . (Readers unused to handling numerical examples may find the following three paragraphs difficult . It is possible to skip them without losing the general thread of the argument) . Inputs

Commodity 1 Process 1 Process 2

5 9

Commodity 2

5

1 10

Labour

9 1

Gross Gross outSurplus put Prod . Cl C2 Cl C2

6 8'h 5 1 /6 13415 12 133/10 11 1 /3

The wage bundle consists of 3/6 commodity 1 and 5/6 commodity 2 . For convenience we assume, as does Steedman, that wages are paid at the end of the production period . In the absence of capitalist consumption, there would be more of commodity 1 to be accumulated than of commodity 2 whichever process was used . But the same amount or less of commodity 1 than of commodity 2 would be required as inputs . Thus there would be excess production of commodity 1, which would therefore have a zero price . With process 1, a

capitalist would advance 5 units of commodity 2 and receive 51/6 units, a rate of profit of 31/6 % . With process 2 10 units of commodity 2 advanced would yield 11 Y2 units, a rate of profit of 131/3 % . So only process 2 would be used and the rate of profit would be 131/3% . If, on the other hand, capitalists consumed 8 units of commodity 1, the output of commodity 2 would exceed input requirements whichever process was used . Thus commodity 2 would have a zero price and the use of process 1 alone would yield a rate of profit of 70% . Thus contrary to Steedman's implicit suggestion, in the general case the equilibrium rate of profit and prices cannot in principle be calculated from a full physical specification of all known possible processes and commodities and the components of the actual wage bundle . With joint production, the equilibrium rate of profit and prices depend not only on these factors but also on the use made of the surplus product .



14

CAPITAL AND CLASS

ARE VALUES REDUNDANT ANYWAY? This section is almost entirely concerned with the simplest case considered by Steedman : one in which all actual labour is socially necessary and in which no fixed capital, no joint products and no choice of technique exist . We first outline and briefly discuss the results he obtains for this case and then consider his claims as to their significance for the status of value categories . Steedman's results For this case, Steedman demonstrates four things which he considers important in his overall argument : (1) That a full specification of the conditions of production and the real wage is sufficient information from which to calculate the rate of profit and relative prices which would hold if all capital appropriated the same rate of profit. (2) That the values of individual commodities cannot in principle be calculated without a prior specification of these same conditions of prodauction (and, in the case of value of labour power, of the real wage) . (3) That it is possible to demonstrate without reference to value magnitudes that profits will be positive if and only if surplus labour is performed and that surplus labour will be performed if and only if profits are positive . (4) That 'in general profits and prices cannot be derived from the ordinary value scheme' (48 see also the accompanying diagram) . Point (4) is extremely misleadingly expressed . It suggests that a full specification of the values of commodities (including labour power) is insufficient to determine the equilibrium profit rate and prices . This is not so ; such a specification is quite sufficient in the simple case, as Steedman explicitly states elsewhere (202) . The correct point Steedman is making here is that the aggregate values of C, V and S for each industry are in general insufficient information from which to calculate the equilibrium rate of profit and relative prices . But this is in no sense an argument for the superiority of his alternative approach . In the general case a full specification of either the physical conditions of production or the values of all commodities is required to calculate the equilibrium profit rate and relative prices. Since the inputs and outputs of different production processes cannot be aggregated in physical terms (as they differ qualitatively), aggregate physical specifications of C, V and S can never be obtained for an industry . There is thus no possibility of ever calculating the equilibrium rate of profit and relative prices from (non-existent) aggregate physical specifications . In certain cases it is possible to do so from aggregate value specifications . (Cases in which it is possible are discussed in section three below) . Bearing this last point in mind, results (1) - ( 4) above are correct . Thus, in the simple case, the Marx-based critique of Steedman must take issue with the significance he attaches to them rather than with their validity . So what? Once the concept of value and the content of Steedman's propos-



IN DEFENCE OF VALUE

15

itions (1) - (3) above are grasped, there is nothing remotely surprising in the latter . Since the value of a commodity depends on the (socially necessary) labour expended both directly and indirectly in its production, values clearly depend, in general, on the conditions of production of all commodities . Since these conditions include both the direct labour time expended and the physical quantities of all non-labour inputs and outputs, there is nothing intuitively odd in the idea that it is in principle possible to calculate the equilibrium profit rate and relative prices directly from this data . Finally, since, in the simple case, there is no question of any commodity having anything but a determinate, positive value, both profit and surplus labour will clearly be positive if and only if a surplus product is produced . Hence it is hardly surprising that it is possible to show that profit implies surplus labour and vice versa without valuing individual commodities . The situation can be expressed most generally as follows . At any point of time there will be in existence various non-labour inputs to production and labour power . The labour power will be put to work with the non-labour inputs to produce other non-labour inputs and consumption goods. The particular ways in which the labour power and other inputs are combined constitute the methods of production in use . Neither values nor prices nor any other numerical magnitude dependent on either of them (rate of surplus value, rate of profit etc) are independent of the particular methods of production used . In general, they will change if different techniques are adopted . To think otherwise would be to believe that values and prices are physical attributes of use-values . The values of all commodities except labour power depend only on techniques used . The value of labour power depends on both techniques used and the real wage . All prices depend on both . Now, all this is really rather basic and obvious . What significance does Steedman then attach to his results? He makes the following claims for this simple case : (A) that values "play no essential role in the determination of the rate of profit (or of the prices of production)" (14) or, again, "it follows that value magnitudes are, at best, redundant in the determination of the rate of profit (and prices of production)" (202) (B) that "since the profit rate and all prices of production can be determined without reference to any value magnitude, the 'transformation problem' is a pseudo-problem, a chimera ; there is no problem of deriving profits from surplus value and production prices from values to be solved" (14-15) In addition, Steedman makes a claim for the case involving choice of technique which can conveniently be dealt with here because it is of a similar status to those made for the simple case - rather than resting on supposed indeterminacy or negativity of values it is an extension of his argument that, because values are secondary to, or purely derivative of, production conditions, they are redundant . This claim is that since the techniques actually used (out of the range of known possible ones) are determined by the criterion of profitability "Values can therefore be known only after the rate of profit has already been determined! The determination of the



16

CAPITAL AND CLASS

rate of profit is thus logically prior to any determination of value magnitudes" (65) . This last claim is quite simply wrong. All that are logically prior to a determination of values are knowledge of all possible methods of production and of the real wage and the application of the criterion of profit maximisation to determine which techniques will be used . Actual techniques used, values, prices and the actual rate of profit are all then determined simultaneously . Thay all have unique magnitudes in any system so specified . While any order of calculation is in principle possible none is logically prior to any other in Steedman's sense of the term . Considered in isolation, claim (B) is similarly just illogical . It does not follow from the fact that the equilibrium profit rate and prices can be obtained from physical specifications without reference to values that the relationship between values on the one hand and prices and the rate of profit on the other hand is of no interest . Thus claim (B) depends on claim (A) : if values are of no interest, there is no (interesting) problem about their relationship to price categories, if values are of interest, there is an (interesting and in the simple case easily soluble) problem about their relationship to price categories . This leaves claim (A) . In the second formulation quoted above the 'at best' is a rhetorical excess - Steedman nowhere argues that there are any problems with values in the simple case . The correct claim is therefore that values are simply redundant . This is clearly correct on one level . If all you are interested in is calculating the rate of profit and relative prices that would hold if all capitals appropriated the same rate of profit in a hypothetical, abstract economy at a point in time at which the full physical specifications are available to you, then there is no point whatsoever in calculating values ; you can economise on both concepts and algebraic manipulation by forgetting about value categories altogether . The other formulation - that values 'play no esssential role . . . ' - can clearly also be interpreted in this way . But, on this interpretation, Steedman's claim is no claim at all about the significance of his results . It is merely a verbal summary of them . To give the claim a substantive meaning Steedman must be interpreted as saying that the concept of value is not an aid to understanding the actual process of profit and price determination in the real world . We argue below that this substantive claim is unfounded . The argument is developed in three stages . First we discuss the process of profit and price determination and the role of transformation-problem exercises in an analysis of it . This is important because many of Steedman's results are obtained from such exercises. Secondly, we make some observations on the relationship between the results Steedman obtains from these exercises and his broader theoretical claims . Finally, in part three, we discuss two elements central to an historical-materialist analysis of capitalism : exploitation and tendential movements in the rate of profit . Here we consider both Steedman's explicit analyses and, more generally, the issue of whether a theoretical framework employing value categories is capable of illuminating these features of the system more effectively than is Steedman's approach . (7)



I N DEFENCE OF VALUE

17

Transformation-problem exercises The first point to make about the process of profit and price determination is that it is a highly complex one . It involves continual changes in the production conditions, values, and prices of individual commodities, in the rates of profit appropriated by individual capitals, in the real wage and so on . It is nothing less than the whole system in motion . It is therefore necessary to decide on which aspects of the overall process to focus . Marxism traditionally, and in our view correctly, concentrates on the reproduction of the system as a whole (as opposed to neoclassical economics, for example, which focuses on the way in which finite 'scarce' resources are allocated between individuals by the operation of market forces) . In particular, Marxism is concerned with two fundamental aspects of reproduction . The first is the social relations posited . The concept of exploitation is central here . The second is the ways in which the social relations within which reproduction occurs affect the development of the productive forces . An examination of the accumulation process including, of course, the contradictions generated within it - is central here . The second point to make about reproduction is that it is a continual process . There is no final outcome in terms of the rate of profit, prices or any other magnitudes . There is only the process itself . Finally, in the real world clearly neither values nor physical conditions determine profits and prices in any simple sense . Prices, for example, are not fixed by god or capitalists consulting a table of values or physical specifications and multiplying them by appropriate mathematical functions . The actual process whereby the rate of profits and prices are determined, and constantly redetermined, is one in which individual capitalists decide which

goods to produce on the sole basis of (existing and expected) prices . The carrying out of these decisions is the real material process tending to bring about a set of relative prices yielding an equal rate of profit in all sectors . How do transformation-problem exercises approach this process? What any such exercise - whether from physical specifications direct to the rate of profit and prices or one involving values - does in effect is to freeze certain aspects of the process (conditions of production and the real wage) and assume that other aspects (the movement of capitals in search of the maximum rate of profit and consequent price and profit movements) continue operating . It then calculates what the final outcome, in terms of the rate of profit and prices, would be if the unfrozen aspects of the process worked themselves out fully ; so that, if every capitalist were fully rational and equipped with an encyclopaedic knowledge of both the existing situation and the fact that it was not going to change, none would have any incentive to switch to the production of a different commodity .

Note that all the transformation exercise does it to calculate what the outcome would be in this purely hypothetical situation . It makes no attempt to analyse the processes involved . Note also that such transformations are hypothetical exercises in another sense : the production conditions and real wage specified are hypothetical . No one has ever assembled that data for a real capitalist economy . Being hypothetical, prod-

uction conditions and the real wage can equally well be specified in either



18

CAPITAL AND CLASS

physical or value terms . It is not a matter of being able to observe physical conditions but not values . A transformation of this sort is a wholly theoretical exercise . This is not to ridicule such transformations . They have a legitmate role to play in analysing capitalism . But it is important to be aware of the nature and the limitations of this role . Tranformation-problem exercises have one basic purpose : to demonstrate that conditions of production and the real wage together determine the average rate of profit appropriated . They do so in that they determine both total profit and total capital . What is the point of showing this? It is not to provide a theory of price and profit determination in the normal sense of the phrase . Since production conditions are frozen, it may at first sight appear that transformations of this nature are intended to provide a short run theory of price and profit determination . But they cannot fulfil this task satisfactorily because factors other than production conditions (e .g . demand) play a significant role in determining relative prices, and hence the profit appropriated by individual capitals, in the short run . Nor can such transformations in any simple sense explain long run movements in profits and prices (and a long run theory of profits and prices must seek to explain movements in these magnitudes) because of the assumption that production conditions do not change . The best transformation-problem exercises can provide here is a variety of comparative statics . The point of the demonstration is two-fold . First, it establishes that profit originates in production rather than circulation . The rate of profit appropriated by an individual capital may depend just as much on factors within circulation as on ones within production . But if production conditions and the real wage uniquely determine the average rateof profit, one capital's gain in circulation must necessarily be another's loss . Thus one purpose of transformation-problem exercises is to contribute to the analysis of exploitation . In section three below we argue that values are important from this point of view . The second purpose of the demonstration is to establish that changes in the average rate of profit can only result from changes in the production conditions and/or the real wage . Note that this is not a theory of movements in the (average) rate of profit . It is rather a precondition for such a theory : it shows which factors must be analysed if such a theory is to be constructed . In other words, it shows that an explanation of tendential movements in the profit rate must be based on an explanation of tendential movements in production conditions and the real wage (and systematic relationships between the two) . Having shown this, transformationproblem exercises, by their very nature, can contribute little to the construction of such a theory . Values are irrelevant to the fulfillment of this second function . But this is trivial . The important question is not whether values are necessary to the demonstration that changes in production conditions and/or the real wage determine changes in the overall rate of profit, but whether value categories are useful in analysing changes in production conditions and the real wage . We argue below that value categories have proved fruitful in this respect in the past and that there is every reason to believe that they will continue to do so in the future .



IN DEFENCE OF VALUE

19

Broader theoretical questions Steedman makes a leap from specific claims as to the occasional indeterminacy of values, and their redundancy in transformation-problem exercises e tc . to the general claim that values are not only redundant but also a positive obstacle to an overall analysis of capitalism . Moving from partial to total claims in this cavalier fashion is quite illegitimate . Steedman makes no attempt to argue that, taken together, the specific claims justify the general one . He simply slides from arguing specific claims to asserting the general one . Relatedly, Steedman again simply asserts that many elements of Marx's analysis, which he believes valuable and wishes to retain, are independent of value analysis . It is sometimes far from self-evident that this is so. Commodity fetishism, to take one example, would seem, for Marx, to include the notion that embodied labour is not recognised as being such . Thirdly, Steedman's view as to the significance of his results seems related to a general misconception about the nature of theoretical discourse . Value categories express certain common features of qualitatively different production processes in terms of an (at least in the simple case) single, quantifiable standard . To consider them redundant because they are logically derivative of and unable to provide more detailed information than the data they summarise is strange . Is the concept 'cat' redundant because it is derivative of and unable to provide as much detailed information as a full specification of all characteristics of all actual or possible cats? Steedman clearly cannot be against the use of general categories as such . But it does appear at times as if his scientific ideal is a kind of one to one mapping of concrete reality in thought - if only we knew the physical specifications of all production processes and the components of the real wage, we could both calculate and understand the process of formation of the average rate of profit! Replacing a three sector value model by a 3000 sector one of physical systems matrices clearly represents a more detailed mapping of reality . But it is not clear that it increases our ability to understand that reality . It is just as impossible to conceptualise a 3000 sector physical systems model without the use of simplifying categories as it is to think about 3000 cats without using concepts like 'cat' . A change in the real wage and/or the surplus product, for example, would be represented in Steedman's models by two different sets of matrices, the first representing the conditions of production of the wage bundle and the surplus product prior to the change and the second the conditions after the change . There would be more of some commodites in each matrix in the second set than in the first . There would be less of others. Some would have disappeared entirely . Others would have entered for the first time . To ask whether the rate of exploitation has risen - to take one potentially interesting question - it is necessary to have a means of aggregating qualitatively different use-values . The only serious candidates for the job are values and prices . We argue below that values are appropriate for this type of analysis whereas prices are not .



20

CAPITAL AND CLASS

This section has argued that, when understood properly, there is nothing remotely surprising about Steedman's demonstration that values depend on physical conditions of production and are redundant to the calculation (in principle) of the equilibrium rate of profit and relative prices. But the usefulness of value categories must be assessed on other criteria - whether they illuminate capitalist social relations and/or aid analysis of the accumulation process . The final section argues that they do both . EXPLOITATION AND TENDENTIAL MOVEMENTS IN THE RATE OF PROFIT This section discusses the concept of the rate of exploitation, the transformation problem and the question of tendential movements in the rate of profit . It is concerned to do two things : to discuss Steedman's objections to Marx's analysis of the transformation problem and of the tendency for the rate of profit to fall and to argue for the importance of value categories in examining all three questions . Since we agree with some of Steedman's criticisms of Marx, it is important to make clear that acceptance of these specific criticisms does not imply agreement that they constitute a criticism of value categories as such . Incorrect argument can be advanced within a correct theoretical framework - one does not lose confidence in the principles of arithmetic every time someone makes a computational error .

Exploitation "That r is positive if and only if S is positive is, in the present context (i .e . the simple case PA, AG, JH) . true but rather obvious . . . The very fact that the proposition in question 'runs both ways' . . . means at once that it does not constitute a theory of why r is positive . . . Neoclassical economists do not commonly invoke a concept of surplus labour but they could do so without causing the slightest inconsistency within their theory . . . Thus the only possible role, in the theory of profits, for the statement that 'r is positive if and only if S is positive' is as the final link in an argument, the earlier stages of which show why S is positive . . . such an explanation must necessarily run in terms of the determinants of the conditions of production and real wages . Marx's value magnitudes, however, far from being determinants of those factors are precisely derivative from them, consequently they can play no essential role in the theory of why profits are positive . The essential elements of such a theory may well include the classical 'double freedom' of the worker . . . and many other factors discussed by Marx . . . but they will not include any embodied labour-time magnitudes .' (58-59) . As Steedman is aware, few Marxists would say that values are determinants of profit in the same sense that the double freedom of the worker is a determinant . Value categories summarise conditions of production and the real wage . There is nevertheless a fundamental justification for summarising them in terms of embodied labour time . This justification is quite simply that what in the last analysis the worker is compelled by the status of 'double freedom' to do is to work .



IN DEFENCE OF VALUE

21

It is also misleading to say that a neo-classical economist could accept the concept of surplus labour without examining adequately the meaning and status that the concept could have within neo-classicism . For although a neo-classical could agree that : 'the rate of profit is thus positive only if some labour is done which is not directed towards the direct or indirect production of the wage bundle' (58) he or she would be bound to add : 'and there could be no wages if some capital services were not directed towards the direct or indirect production of the wage bundle . Furthermore, the value contributed by workers when producing goods for capitalists is precisely balanced by the value created by the capitalists via the provision of capital for the production of wage goods . Thus there is equal exchange between the classes and no exploitation .(8,) Any concept of surplus labour which is not derived from the position that labour is the source of all value is utterly trivial . For our neo-classical, the statement that : 'the capitalist class forces the working class to work longer than is required to reproduce the wage bundle' would be neither more nor less meaningful than the statement that : 'the working class forces the capitalist class to provide the capital services required to produce wage goods' . The absurdity of the equivalences which neo-classicals would be forced to argue for were they to refer to the concept of surplus labour, goes a long way towards explaining why they do not generally do so . Steedman's position on the rate of exploitation is hard to fathom . In one place (111) he suggests that an alternative to describing the surplus in terms of values (i .e . any set of commodities embodying total value x) is to specify the physical possiblities directly (i .e . all possible combinations of commodities which could be produced over and above the wage bundle) . In a multi-commodity economy, this procedure would be a trifle cumbersome, and its outcome impossible to conceptualise . In another place Steedman appears to suggest that the Morishima approach - which he advocates for more complex cases and which we have criticised above - is a preferable way of conceptualising the rate of expoitation even in the simple case where it is equivalent to the normal conception : 'in a single product, circulating capital system Morishima's approach will yield the same measure of surplus labour as will the additive value



22

CAPITAL AND CLASS accounts - though without any need to calculate individual commodity values - if the processes actually used by the capitalists are those adopted in the labour-minimising allocation (which in general they will not be (because of choice of technique PA,AG,JH))" (194 : note 13)

A clue to the supposed advantage of this conceptualisation may be the aside `without the need to calculate individual commodity values' . If so, this is just silly . Individual values are just as implicit in Morishima's approach when applied to the simple case as they are in the standard conceptualisation . In the absence of choice of technique and joint production, the two approaches are quite simply coincident . All this is necessarily a little speculative because Steedman nowhere makes explicit his basic position on the question . . The only thing that emerges clearly is the fact that he bends over backwards to avoid explicitly accepting the normal concept of the rate of exploitation even in cases which present no problems for valuing individual commodities . It is hard to interpret this as anything but a determination to avoid admitting the usefulness of value magnitudes in any circumstances . Fundamentally, there are only two possible ways of aggregating heterogeneous commodities so as to derive ratios between bundles of them : values and prices . Steedman's ambiguity on the question of the rate of exploitation reflects the fact that he rejects the first alternative whilst being unwilling to adopt the second . The latter fact at least is to his credit . Since some of his followers may be more foolhardy - reasoning that prices at least have the advantage of being always determinate and positive - it is worth spelling out the implications of embracing such a conceptual measure . Profits and wages measure the proportions of net output that each class receives . But the measure of net output, being based on prices, presumes that the value of this output is produced by both labour power and the advance of capital . If this is not obvious, consider the simple case of a commodity produced by labour operating fixed capital of infinite durability and using no materials . The price of production of such a commodity is the sum of the price of fixed capital and the quantity of labour power used in it prodduction, weighted by the rate of profit and the wage respectively . To use the price as the measure of the resources used to produce the commodity, and thus indirectly appropriated by its possessor, is therfore implicitly to regard labour and the advance of capital as the resources which produce the commodity's value (with their contributions - labour services and capital services in the neo-classical terminology - being measured by the wage and profit rate respectively) . So to aggregate the components of the wage bundle and those of the surplus product by their prices, which is to calculate wages and profits for they are the sum of the prices of production of these bundles is to calculate the quantity of resources used to produce the respective bundles in a way which includes the advance of capital as one of these resources . To calculate the ratio of profits to wages is thus to compare the bundles of resources appropriated indirectly by the capitalists (as inputs into the surplus product) with the bundle appropriated



IN DEFENCE OF VALUE

23

indirectly by the workers (as inputs into the wage bundle) in a manner which considers the advance of capital to be a factor contributing to the value of output . Thus, while to refer to this ratio implies nothing in and of itself (indeed, in the absence of alternative data, changes in the profit/wage ratio may be the best available indicator of movements in the rate of surplus value) to regard it as an adequate conceptualisation of the rate of exploitation is to accept implicitly that profit measures the contribution of capital advanced to the value of the product . The only notion of exploitation which this conception permits is a purely distributional one . A weak variant is the neo-classical conception . Here exploitation is seen as occuring if 'factor prices' do not accurately reflect 'contributions'-if profits are 'too high' relative to the 'contribution of capital'. A stronger variant would see exploitation existing so long as profit is appropriated by a capitalist class with monopoly ownership of the means of production . In either case the notion of workers being compelled to perform surplus labour is lost. At root there are only two possible ways of analysing production and appropriation . Labour can be seen as producing the entire net output in which case to be theoretically consistent the commodities received by each class must be reckoned (in a conceptual sense) in terms of their labour content. If this approach is adopted then exploitation can consistently be located in production . Alternatively, labour and fixed capital can be seen as both contributing to the production of net output in which case to be theoretically consistent commodities must be reckoned (in a conceptual sense) in terms of labour power and capital advanced (prices of production) . If this approach is adopted then exploitation must either be denied in the general case (the weak variant) or located entirely within circulation (the strong variant) . Any conception of capitalism which wishes to maintain a non-trivial notion of the compulsion to perform surplus labour must necessarily embrace implicit value categories . Whether or not these are explicitly calculated in all cases and, if so, in which way, are second-order issues .

Marx's solution to the transformation problem Steedman's chapter criticising Marx's solution to the transformation problem (Chapter 2) is the best introduction to the subject available . He explains the basic point simply and intuitively without resorting to algebra . Anyone who still has doubts about the correctness for the general case of Bortkiewitz-type solutions should read this chapter . As Steedman points out, output in any given period may be divided into three sets of goods : 'a bundle of commodities going to the capitalists (which constitutes net investment and capitalist consumption), a bundle of commodities which replace produced means of production (the physical aspect of constant capital) and a bundle of commodities going to workers as wages (the physical aspect of variable capital)' (29-30)



24

CAPITAL AND CLASS

To compare the relative sizes of these bundles, it is necessary to aggregate heterogeneous goods in terms of a common standard . If values are used as the standard, the 'value rate of profit' (S/C+V) can be calculated . If prices of production are used as the standard, the money rate of profit can be obtained . (9) Price and value rates of profit will only coincide if the average ratio of price to value of commodities comprising the surplus product is equal to that of commodities comprising constant and variable capital . Now, Marx assumed that the price and value rate of profit must always be the same for the economy as a whole and calculated relative prices accordingly . This is clearly illegitimate in the general case . Steedman puts the point neatly : '(Marx) assumed that S/(C+V) is the rate of profit but then derives the result that prices diverge from values, which means pricesely, in genral, that S/(C+V) is not the rate of profit (31) .' Thus far Steedman is quite correct . But it does not follow that Marx's approach does not constitute a legitimate starting point so long as it is made clear that it is only applicable to limited cases . Steedman fails to recognise that this is so because he does not see clearly the point that Marx was trying to make . Steedman sees the issue as follows : It is simple . . . to show how profits will be positive if and only if there is surplus value i .e . capitalist exploitation . That is all that is at issue ; any question of total profit and total surplus value being equal in magnitude is trivial' (34) He is quite right about numerical equality . This becomes obvious once the point made above about different ways of valuing the surplus product is grasped . Steedman is nevertheless correct to stress the point because much confusion has resulted from people ludicrously insisting that the two magnitudes must always be equal . (10) But Steedman is wrong to suggest that Marx was concerned only to show that, in aggregate, positive profits implies positive surplus value . He was also concerned to show that the process of profit appropriation by individual capitals is fundamentally one of sharing out total surplus value produced according to criteria and by means of mechanisms which have the effect of obscuring what is going on . This fundamental idea does not depend on numerical equality between profit and surplus value and is contained within Bortkiewitz-type solutions . But it can be shown more simply and intuitively in cases where numerical equality exists . It is therefore expositionally convenient to begin with such cases, so long as it is made clear that they are special cases .(/ 1) Following the last sub-section, it should not be necessary to argue that value categories are important in establishing this link between exploitation, surplus value production and profit appropriation . (12)



IN DEFENCE OF VALUE

25

Tendential movements in the rate of profit

In discussing Marx's formulation of 'the law of the tendency for the rate of profit to fall' (Chapter 9), Steedman makes four points : (1) Marx's 'expectations' of rising C/V and S/V combined with falling (S+V)/(C+V) and S/(C+V) are mutually consistent, but (2) A rising technical composition of capital need not lead to a rise in C/V . (3) A falling (S+V)/(C+V) - maximum rate of profit - need not lead, even 'eventually', to a falling S/(C+V) - actual rate of profit . (4) With a given real wage, technical progress can never lead to a fall in the rate of profit so long as capitalists act to maximise profits . Point (1) is correct, clearly presented and, given the confusion that surrounds the issue, well worth setting out . Point (2) is confusingly presented . Steedman says that 'it is clear that no inference can be drawn from the increasing 'mass' of means of production per worker (however defined) concerning changes in C/V' . (125) . This statement is strictly correct . But if the real wage is constant and productivity grows at the same rate in departments I and II, then movements in C/V are precisely correlated with movements in the technical composition of capital . (Changes in the ratio of means of production per worker to wage goods per worker exactly follow changes in the technical composition of capital if the wage is constant . Since C and V are the values of means of production and wage goods respectively, the ratio of one to the other will move in the same way as the ratio of the physical quantities if productivity grows equally in both departments) . Thus if, as Steedman says, it was 'plausible' for Marx to assume a rising rate of exploitation then it was equally plausible for him to assume that C/V would rise providing productivity growth were fairly similar across departments, the latter On point (3), all Steedman shows is that if the ratio of dead to living labour rises towards a finite limit, so that the maximum rate of profit falls towards a finite limit, then the actual rate of profit need not fall . But the traditional argument is ( or at least should be) that if the ratio of dead to living labour were to rise without limit, so that the maximum rate of profit fell towards zero, then the actual rate of profit would eventually have to fall . There are two substantive points to be made in this connection . The first is that it is impossible to establish a priori that a rising technical composition implies a falling maximum rate of profit (i .e . rising ratio of dead to living labour) . This is because the negative effect on the ratio of dead to living labour of the devaluation of constant capital resulting from productivity growth may exceed the positive effect of a rising technical composition . Hence, considered at this level of abstraction, the outcome is contingent. The second substantive point concerns the relationship between any tendency for the ratio of dead to living labour to rise and Steedman's fourth point above . Point (4) is correct and has been demonstrated on a number of occasions. Its significance is that techniques involving an increase in the ratio of dead to living labour cannot be introduced without limit because



26

CAPITAL AND CLASS

such techniques will only be introduced in conditions in which they will not have the effect of reducing the actual rate of profit and these conditions could not be met if the ratio of dead to living labour were to rise without limit . Now, Steedman's points (1) -(4) above embody substantive statements and point to interesting questions about the accumulation process . Taken in conjunction with much previous Marxist work, they show that it is possible to raise and at least begin to answer some important questions concerning the laws of motion of capitalism within a framework employing value categories . The fact that some people have advanced mistaken ideas within this framework in no way invalidates this judgement . Indeed, a theoretical framework in which this could not happen would be a strange basis for scientific discourse . To argue convincingly that value categories constitute an obstacle to understanding developmental tendencies within capitalism, Steedman would have to show either that the basic concepts are irrevocably flawed or that an alternative approach shows every indication of being more fruitful . Marx after Sraffa is hardly convincing on either score . The page of Steedman's own analysis - beginning with the promising question 'what can be said then about the movement of the rate of profit on the basis of relation (18) (the physical specification PA, AG, JH)? ends up : 'It can only be concluded that there are probably some forces at work tending to increase r and some tending to decrease r over time' (130-132) . This is formally correct but totally vacuous . Steedman can get little

further because he has no concepts with which to order proximately, describe and hence understand these forces . He has only snap shots of physical systems matrices . The only statements he can make about developments over time are that, between one frame and another, some elements have risen, some fallen, some disappeared and some made their debut . This is not much help in trying to understand the processes involved . Value magnitudes permit the construction of precisely the type of categories which are missing from Steedman's framework . The concepts of the ratio of dead to living labour and the rate of exploitation encapsulate in a conceptual ly-hand[able form sets of forces which are fundamental to the determination of the rate of profit . Moreover they do so by means of the self-same categories which most effectively illuminate the nature of capitalist social relations . It is noticeable that in the chapter on tendential movements in the rate of profit - the only one in the book in which laws of motion are discussed - Steedman's primary objection to value categories is not their supposed redundancy but rather their approximate nature . (14) . It is certainly true that the convential formula specifying the rate of profit (r) as a function of the ratio of dead to living labour (R) and the rate of exploitation (e) i .e. 1+r =(1+e) (I+R) is in general an approximation . But 1+e+R

this is not a very damning criticism if the only alternative is a totally unmanageable physical - or for that matter value specification of every single process and commodity .



IN DEFENCE OF VALUE

27

Unless and until Steedman can provide either a workable alternative to convential aggregate value categories or a more convincing argument that the latter are irrevocably flawed his critique of their usefulness in analysing laws of motion cannot be considered to have got off the ground .

NOTES

1

Philip Armstrong, Andrew Glyn and John Harrison are researching into the development of metropolitan capitalism since the war at The Institute of Economics and Statistics, Manor Road, Oxford . They would like to thank Ian Steedman and members of and readers for the Editorial Committee of Capital and Class for helpful comments on an earlier draft . A third possibility would be to define the value of a commodity (or any group of commodities, such as the wage bundle) as the labour

required by the technique(s) which of those in use, minimise(s) total

2.

3.

4

labour input . This would be a 'narrower' interpretation of 'technical possibilities' than is (1) . But it is open to both our objection to (1) and Steedman's objection to (2) . It has nothing to commend it and so is not considered further . To call these results 'Marx's additive definition of value', as Steedman does, is hardly fair to the man . Marx neither considered joint production in any detail nor calculated values by solving a set of simultaneous equations . Note that if Q and P are near to the origin, values are very large, approaching infinity as Q and P approach the origin . With pure joint production, very large or infinite values are likely to result from the relative proportions in which two commodities are produced being 'naturally' (at a certain level of development of the productive forces) determined and hence the same in all known possible techniques . A small point : Steedman is wrong to say that 'Since W lies above the extension N1 N2 (AB in our diagram PA, AG, J H), it follows that the real wage bundle embodies more than 6 units of labour and hence that additive surplus value is negative' . (159) The correct formation of the general point is 'since W lies on the other side of the extension N 1 N 2

from the origin . . . ' 5

If there is a third process which is equally profitable then again individual values cannot be calculated . This is similar to the case of choice of technique without joint production .

6

The second hand price of individual machines (whether market or internal accounting) may be of interest . But this will depend on (expected) future profit appropriation - being discounted future profit flows - rather than on the value remaining embodied in the

machine, however conceived . Without knowledge of the aggregate value of fixed capital in existence it would not be possible to calculate the rate of value expansion . But, with 'balanced growth', this



28

CAPITAL AND CLASS

magnitude does not depend on the scheme used to calculate value transferred . To avoid possible misunderstandings, it may be worth explicitly 7 stating a couple of things at this point . First, a fully adequate assessment of Steedman's claim would require the solution to a complex of problems which can conveniently be grouped under the heading of 'historical-materialist methodology' . Many of these problems are exceedingly difficult, many are interlinked and many are incapable of resolution on an a priori basis . This does not mean that they can be evaded . It means rather that no 'simple', definitive rebuttal of Steedman's claim - equivalent to, say, the isolation of an error in algebraic manipultion - is possible . (The obverse of this point is, of course, that Steedman cannot provide - or be expected to provide - arguments for his assessment of the implications of his results which have the same 'hard' status as those concerning their validity) . This point is perhaps obvious . It is certainly important . Second, in arguing for the usefulness of value categories we are not claiming that they are appropriate for analysing all aspects of a capitalist economy . Again this should be obvious . Unfortunately, some Marxists appear at times to regard prices, physical specifications and so on as purely ideological categories inappropriate to a scientific analysis of capitalism . Price is no less a Marxist category than is value . Both have a role to play in an analysis of capitalism and both can be, and frequently are, abused by being applied to problems for which they are inappropriate or ignored when examining processes whose analysis requires them . Nor is it the case, as is often asserted, that, while price categories may be relevant to the actions of individual capitals, all developments at the level of the system as a whole can be analysed purely in terms of values . A catalogue of errors of this sort would make an instructive if lengthy monograph . To anticipate an objection to this second point, it is not eclectic to use different concepts to analyse different aspects of a complex, overall process so long as those concepts (and the theoretical propositions derived from them) are related to each other in a coherent, overall theoretical structure . 8 If asked to spell out the argument, our friendly ideologue would oblige as follows . Total profits equals profits earned in producing wage and non-wage goods . Total profits also equals the sum of the prices of all non-wage goods (since profits constitute the demand for such goods) . The sum of the prices of non-wage goods equals the sum of wages and profits earned in their production . Therefore profits earned in producing wage goods (the value of capital services provided to workers) equals wages paid to workers producing non-wage goods (the value of labour services provided to capital) . Profit is the sum of the absolute prices of all commodities comprising 9 the surplus product . The rate of profit is the ratio of profit to the cost of capital advanced . Relative prices (i .e . prices of production) are sufficient information from which to calculate ratios . 10 Thus Mandel poses the following question : where, if the sum of profit



IN DEFENCE OF VALUE

29

expressed in price of production terms differs from the sum of surplus value, can the difference have come from? (Mandel 1973 p .63) . This question is analogous to the following : where, if the distance between Paris and London expressed in kilometers differs from the distance expressed in miles, can the difference have come from? 11 An objection might be that the basic case of numerical equality is one in which all prices are directly proportional to values . Here profit appropriation tends to be directly proportional to surplus value production and hence the extent to which mechanisms and criteria for profit appropriation obscure what is going on is limited . One answer to such an objection is that the case in which the processes are least obscured is precisely the best one to begin with . Another answer is that an alternative case of numerical equality is one in which the surplus product consists of the same commodities in the same proportions as those making up constant and variable capital . This case is less restrictive than may appear at first sight : It holds if all surplus is accumulated and accumulation proceeds at the same rate in all industries (the phenomenon of 'balanced growth') . Being the simplest case of expanded reproduction this is a legitimate and obvious starting point. 12 Two attempts have been made to use Sraffa's concept of the 'standard commodity' to clarify or make more rigorous Marx's analysis of the equilibrium rate of profit . Medio re-defines the value composition of capital (organic composition in his terminology) as the average for the standard commodity, as opposed to the actual average for the economy . (The difference is in the weights used to average the actual compositions in different industries) . He then demonstrates that the conventional formula expressing the equilibrium rate of profit as a function of the rate of exploitation and the value composition of capital does apply in the general case if his concept of value composi= tion is substituted for the conventional one . Eatwell seeks to achieve the same objective by means of re-defining the rate of exploitation . His concept of the latter is the proportion of output of the standard commodity which is appropriated by capitalists divided by the proportion which the working class could purchase given the existing wage . (The money wage purchases a particular bundle of commodities which, at prices of production, could be exchanged for a certain amount of the standard commodity) . If this concept of the rate of exploitation is substituted for the conventional one, a clear relation holds between the maximum rate of profit and the rate of exploitation on the one hand and the equilibrium rate of profit in price of production terms on the other . But Eatwell's concept of the rate of exploitation is bizarre . On his definition, necessary labour is time required to produce a bundle of commodities which the working class does not consume but which, at the existing rate of profit, could be exchanged for the bundle which it does consume . Since this concept of the rate of exploition depends on prices of production, it is closer to a profits/wages conception than to a conventional labour-time one (although it is expressed in value rather



30

CAPITAL AND CLASS

than price terms) . It serves as an excellent example of the way in which people, in the process of attempting to maintain precise functional relationships between concepts in more complex cases, can lose sight of the purpose of the concepts themselves . An apparently more general result may in fact be less interesting than a more restricted one if generalisation is achieved only by re-defining the categories involved such that they no longer express those aspects of reality which they are intended to encapsulate . These demonstrations - or at least Medio's, whose substitution of definitions does less violence to the categories involved than does Eatwell's- may be of some help in understanding the nature of the approximations involved in the standard formula relating the rate of profit to the rate of exploitation and value composition of capital . But the newly-defined concepts are in no sense operational ; neither Medio's value composition nor Eatwell's rate of exploitation describe even in principle actual situations . 13 In the appendix to this chapter, Steedman accepts the Fine and Harris definition of the organic composition of capital (o .c .c .) as C*/V* where C* and V* are means of production and wage goods respectively at base year values (Fine and Harris 1976 p .161) .On this definition, the o .c .c . and C/V move together if the values of means of production and wage goods do so . A more satisfactory interpretation of the o .c .c . as the value composition 'in so far as it is determined by its technical composition and mirrors the changes in the latter' is C*/(V+S) i .e . undevalued dead labour per worker . On this definition, the o .c .c. less devaluation of constant capital equal the ratio of dead to living labour . 14 This certainly appears to be the drift of the argument at the close of the chapter . At the end of the appendix, however, Steedman asserts that "neither the value composition nor the organic composition is a significant concept for the analysis of capitalist economies . Only the technical composition expressed as a vector of physically specified commodities is of significance" (136) . This looks like redundancy again .

BIBLIOGRAPHY Eatwell, J ., 1975, 'Mr . Sraffa and the rate of exploitation', Quarterly

Journal of Economics . Fine, B . and Harris, L ., 1976, 'Controversial Issues in Marxist Economic Theory', The Socialist Register . Mandel, E ., 1973, 'Value, Surplus Value, Profit, Prices of Production and Surplus Capital', International, Vol . 2, No .], Spring . Medio, A ., 1972, 'Profit and Surplus Value' in Hunt, E . and Schwartz, J . ed . A Critique of Economic Theory, London, Penguin .



IN DEFENCE OF VALUE

31

Morishima, M ., 1974, 'Marx in the Light of Modern Economic Theory', Econometrica.

von Neumann, J ., 1945-6, 'A model of general economic equilibrium,' Review of Economic Studies Sraffa, P ., 1972, Production of Commodities by Means of Commodities,

London : Cambridge University Press. Steedman, I . 1976, 'Positive Profit With Negative Surplus Value : A Reply to Wolfstelter', Economic Journal . Steedman, I . 1977 Marx after Sraffa, London ; New Left Books .

MERLIN PRESS E. R. E. R.

MANDEL MILIBAND/J . SAVILLE BOHM-BAWERK/ HILFERDING

Marxist Economic Theory Socialist Register 1977 Karl Marx and the close of his System

£3.00 £2 .50 £2 .00

These are just a selection of reduced price titles available through the Merlin Book Club . KELLEY REPRINTS OF ECONOMIC CLASSICS R .V . CLEMENCE P .H . DOUGLAS J . GRAY E . LOWENTHAL W . STARK

The Schumpeterian System The Theory of Wages The Social System The Ricardian Socialists Ideal Foundations of Economic Thought

For details of the Merlin Book Club and the Kelley Reprints, please write to :The Merlin Press, 3 Manchester Rd ., London E14

£ 6 .25 £12 .75 £11 .00 £ 5.25 £ 8 .50

CAPITAL, FRACTIONS OF CAPITAL AND THE STATE `NEO-MARXIST' ANALYSIS OF THE SOUTH AFRICAN STATE Simon Clarke This paper examines what has been called the 'neo-Marxist' or, more unkindly, the 'vulgar Poulantzian', approach to the state . The typical feature of this approach is a formulation of the relation between capital and the state in terms of the struggle between 'fractions of capital' to establish the position of 'hegemonic fraction' within the 'power bloc' . The state then represents the unity of capital in relation to the proletariat and the dominance of the hegemonic fraction in relation to the other fractions of capital . Research which is guided by this view of the state tends, therefore, to examine state activity in relation to the policies proposed by various political parties and pressure groups which are said to be the 'representatives' of particular fractions of capital . This examination is usually complemented by an account of the particular 'interests' of the fractions in play . The policy actually adopted by the state is then explained by reference to the supposed 'hegemony' of the fraction in whose interest the policy is said to be . This paper aims to develop a critique of this 'fractionalist' approach to the state through a critique of its conceptions of class, of class fractions and of the state . It is argued that the 'fractionalist' approach is 'overpoliticised' in seeing the state as condensation of the social, at the same time as being 'economistic' in directly identifying the state with capital . The critique is developed with reference to the best 'neo-Marxist' work on the state that has blossomed in recent years, that on the South African state . It is hoped, however, that the critique will have a wider relevance .(1)

1.

THE ORIGINS OF THE DEBATE

'Neo-Marxist' analyses of the South African state have emerged from the discussion which followed the important accounts of the nature of apartheid by Wolpe (1972) and Legassick (1974a, 1974b) . These accounts saw the development of apartheid as an expression of the class struggle in the specific context in which that struggle was conducted in South Africa : forms of control of the working class which had been developed in the



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

33

period of segregation were adapted to the higher state of the class struggle characteristic of the period of secondary industrialisation . Wolpe put the emphasis on the development of the apartheid system as a means of maintaining control of a working class whose real living standards had been eroded by the decline of the pre-capitalist forms of agricultural production in the Reserves . Legassick put less emphasis on the rather speculative and economistic argument of Wolpe's original paper, regarding apartheid as a response to developing militancy of the urban African working class without reducing this militancy to the decline in living standards or the response to the needs to maintain low wages . These papers, and especially Wolpe's, with its more daring attempt at explanation, have come under a lot of criticism . In the light of the criticism expounded by Williams (1975) in particular it has become difficult to sustain the argument about a direct link between the decline in the Reserves and the institution of apartheid . Williams raised the question of locating the development of apartheid in relation to the valorisation of capital rather than in relation to the stunted dynamic of the exhausted 'pre-capitalist mode of production' . Posing the question in this way made it immediately clear that Wolpe's emphasis on the cheapness of labour was erroneous . However William's own account rested on an erroneous analysis of the supposed peculiar features of gold mining (2) and the original contribution of his article, his insistence on the centrality of the accumulation of total social capital, was not developed in subsequent debate . Legassick's comments about the importance of secondary industrialisation and of increasing capital intensity indicated one direction in which the analysis might fruitfully be developed, but remained rather enigmatic in themselves . The main thrust of the debate, however, has not been to develop the accounts of Wolpe and Legassick in the light of Williams's criticism and Legassick's comments . Rather it has been to latch onto another weakness of the Wolpe-Legassick accounts which was their tendency to look at the needs of capital in a rather undifferentiated way, seeing apartheid as the response to the general need of capital for a cheap or a disciplined working class . Although they recognised the differentiation of capitalist interests, they did not analyse it . This weakened their attempt to locate satisfactorily the specificity of the apartheid system, other than by an implicit reference to the legacy of the period of segregation . The 'fractionalist' accounts that have developed in response to the original Wolpe-Legassick contributions have made the differentiation of capitalist interests their starting point, seeking to relate apartheid to the struggle between fractions of capital rather than directly to the struggle between capital and labour . The specificity of the apartheid state undoubtedly owes much to the markedly uneven development of capital in South Africa . What is at issue in this paper is not that unevenness, but the way in which it is conceptualised . I shall argue that the fractionalists base their account on erroneous conceptions of class, of class fractions and of the state that leads them to reinstate old liberal conceptions of the Apartheid state that see the state as the locus of struggle between 'progressive' and 'reactionary' capitalist interests, and not primarily as the locus of struggle between capital and labour .



34

CAPITAL AND CLASS

struggle between capital and labour . The importance of the contributions of Legassick, Wolpe and Williams was that they broke decisively with these liberal conceptions . Thus in questioning the fractionalist account I want to suggest that the debate needs to return to the priorities mapped out by the original contributions . I shall begin my critique by looking at the concept of a 'fraction' that is implicit in the fractionalist accounts . I shall then develop a critique of the underlying concept of class and the associated conception of the state . The unifying theme of the critique is the fractionalist conception of the economic and the political and the corresponding understanding of the relation between the two . This understanding I characterise as both overpoliticised and economistic : the two realms being defined independently, they can only be brought back together in a reductionist manner . The state is thus identified with capital and the class struggle is excluded from it . This critique might be called 'symptomatic' by some since the theory underlying the fractionalist analyses is never spelt out so as to be amenable to direct criticism . The paper proceeds from the direct critique of fractional ist approaches to consider how we might more adequately conceptualise the relations between particular capitals, on the one hand, and the relation between differentiated capital and the state, on the other . In conclusion an attempt is made to give the themes of the paper some direct political relevance by outlining an approach to the current crisis in South Africa that brings the class struggle once more to the forefront . 2.

WHAT IS A 'FRACTION' OF CAPITAL?

The theoretical basis of the 'fractionalist' accounts of the South African state is provided by the work of Poulantzas . (3) However this basis is given very little discussion, and remains largely implicit : the aim is to provide a concrete account rather than to develop Poulantzas's observations analytically . To introduce the discussion I want to raise the question of the identification and definition of 'fractions' of capital, and here neither Poulantzas nor the fractionalists are of much help . Poulantzas himself has a very abstract discussion of the question (Poulantzas 1973, pp .77-85) that makes it clear that fractions of classes can be based on political and ideological discriminations as well as on economic ones, but his account is too vague to make it possible to determine which social groupings that have or may have 'pertinent effects' at the political level constitute fractions and which do not . The fractionalist literature does not take us much further : in practice a fraction can be constituted on any basis that can define a distinct interest . Hence in the South African literature referred to (4) we find fractions differentiated according to the branch of production in which they are engaged : "mining capital", "manufacturing capital", "agricultural capital", "industrial capital" ; according to their 'nationality' : "imperial capital", "local capital", "metropolitan capital", "indigenous capital", "British capital", "South African capital", "foreign capital", "national capital", "Afrikaner capital", "English-dominated capital" ; according to the phase in the circuit of



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

35

capital : "finance capital", "industrial capital", "commercial capital", "landed capital" ; according to other social characteristics, especially size : "monopoly capital", "competitive capital", "big capital", "medium capital", "state capital", "progressive capital" ; or according to "political/ ideological" criteria in the form of the "English/Afrikaner traditions" . It might seem that this proliferation of features that can define an interest would lead to a limitless proliferation of fractions . This possibility is avoided by identifying different features with one another. A common example is the identification of the conflict, or 'contradiction', between mining capital, on the one hand, and agricultural and manufacturing capitals on the other, with that between imperial and local capital . On top of this is an identification of the conflict between imperial and local capital with that between big and medium or monopoly and non-monopoly capital . For some reason finance capital is often identified with national and industrial capital, despite its very close foreign and mining connections, and state capital is identified with Afrikaner capital despite the fact that state economic activity is focussed on the provision of infrastructural facilities for the extractive industries (cf . Morris 1975, appendix ; Kaplan 1976b n .3 ; Fransman and Davies 1977 pp .263-8) . The unfortunate result is that the problem of deciding at which level the fractions ought to be defined is avoided altogether (cf . Kaplan et al 1977, pp .5-6) . The proliferation of fractions is not only limited by collapsing distinguishing features into one another . While a common interest is necessary to the definition of a fraction, it is not sufficient . A number of individuals may have an interest in common without their constituting a fraction of a class . To become a fraction it is necessary that the group in question has 'pertinent effects' at the political level, which will normally, but not always, mean that this interest group is politically organised in some way : "A fraction . . . is defined in concrete class struggle and is recognised, inter alia, by a specific presence as a fraction at the ideological and political levels ." (Davies 1977, p 28) Thus fractions are not identified by locating particular interests in some specified way, but are rather identified by locating their 'pertinent effects' at the political level . If we concentrate on fractions of capital, and ignore the exceptional cases of fractions that have 'pertinent effects' without any kind of organisation (a rather dubious category), we can clarify what a fraction of capital actually is for this approach . A 'fraction' of capital is the product of the political organisation of a number of individual capitals who have an interest or interests in common, whether that organisation takes the form of a pressure group, or a political party (itself seen in traditional political sociological terms as another kind of pressure group), or some other part of the state apparatus (as when a particular government department 'represents' a particular fraction of capital) : fractions of capital only exist as such through their representation in the political process . Correspondingly they can only be identified by locating the interests 'represented' by the parties to various political conflicts . Concrete analysis is therefore concerned with attaching interests

36

CAPITAL AND CLASS

to the particular organisations or institutions involved in political conflict on the grounds either of the membership of the organisations (the organisation has many members with foreign connections/mining interests/ English speaking), or of the 'interests' it expresses (the organisation puts forward views that would seem to favour foreigners/miners/English speakers) . (5) We find, paradoxically, that although this approach seeks above all to establish the 'relative autonomy' of the state, the relation between the political representative and the interest represented is a simple and direct one . This inability to conceptualise the complexity of the relationship between economic and political expresses the adoption of that 'abstract' materialism that Marx sharply criticised in a very similar context for its substitution of the analytical for the dialectical method : "It is, in reality, much easier to discover by analysis the earthly kernel of the misty creations of religious than to do the opposite, i .e . to develop from the actual, given relations of life the forms in which these have been apotheosised . The latter method is the only materialist, and therefore the only scientific one . The weaknesses of the abstract materialism of natural science, a materialism which excludes the historical process, are immediately evident from the abstract and ideological conceptions expressed by its spokesmen whenever they venture beyond the bounds of their own speciality ." (Capital 1, p .494n) . Instead of showing how political conflicts arise as developed forms of class struggles constituted at the level of the relations of production by showing how various political and semi-political institutions develop on the basis of struggles centred on specific barriers to the valorisation of capital in South Africa, the procedure is reversed and simplified and the key to the class struggle is sought immediately in the relation between political organisations . The analyses therefore degenerate into an account of pressure group activities interpreted in terms of a theory of class . The central question that arises theoretically is that of the relation between the concept of class and that of a pressure group . Examination of this question will reveal the basis of the fractionalist concept of class and so illuminate the fractionalist conception of the state . 3.

THE THEORETICAL BASIS OF FRACTIONALISM

It must be stressed that the inability of this fractionalist approach to offer a properly historical materialist analysis of political struggles is not simply a methodological weakness, but derives from the theory on which it rests . The matter is best approached by identifying the concepts of the economic and the political, and of the relations between the two, which is implicit in this approach . The principle characteristic of these conceptions



CAP! TA L , FRACTIONS OF CA PI TA L A ND THE S TA TE

37

is the radical separation of the two 'levels' and a corresponding inability to integrate them satisfactorily in a single analysis . On the one hand the economic level is analysed as the sphere in which individual capitalists pursue their own particular interests, these interests being discrete and fundamentally independent of one another, determined by the specific circumstances in which they produce . Thus each capital is established on a more or less " 'independent' economic base" and seeks "the expanded reproduction of that base" (Kaplan 1975, p .14) . On the other hand the political level is analysed as the sphere in which individual (capitalist) interests come together to form social groups of various kinds - class fractions - with various common interests . Thus : "In identifying a concrete social formation it is necessary first to identify the different capitals according to the criterion of economic ownership . . . Having done this it is then necessary to establish the 'coalescence of interests' of the various capitals defined according to economic ownership . This 'coalescence of interests' will be determined by the objective interests of the different capitals and in turn, will determine the system of alliances that will be formed between capitals . Depending on the material conditions, different alliances may be established in response to different problem-situations ." (Fransman and Davies 1977, p .263) This conception of the relation between economic and political can be called overpoliticised because the economic level is not essentially social, but is a sphere in which specific individual material interests are constituted .

Thus it is only at the political level that these interests acquire their

social form in the representation of those interests through pressure groups : it is only through their 'pertinent effects' at the political level that class fractions effectively exist . The same is true a fortiori of social classes as is true of class fractions . Like a fraction, a class is seen as a coming together of a multiplicity of individual capitals, this coming together being the distinguishing feature of the political level . Thus Kaplan sees the class relation between capital and labour in terms of "a congruence of interests, in respect of labour, by all elements of capital" (Kaplan 1975, p .1) The unity of capital is not, therefore, something prior to the definition of particular capitals but is simply the sum of capitals each of which is established on a more or less "independent' economic base" (Kaplan 1975, p .14) . The unity and antagonism of classes is therefore defined in terms of a coalescence of individual interests rather than in terms of the fundamental relations of production . The question immediately raised by this conception of the relations between the economic and the political is that of .the identification of pressure groups (and similar organisations constituted at the political level) with social classes and class fractions : what is it that distinguishes a social class or class fraction from any other interest group in society? I shall argue that, given the conception of the economic and political that is the starting point of this approach, there is no way of establishing a privileged relationship between the two levels that is not an immediate identification of one with the other . that is not characterised by an



38

CAPITAL AND CLASS

economic reductionism . Thus, as we shall see, the other side of the overpoliticisation of this approach is an economism that assails it every time it tries to go beyond the bounds of the political . The question of the relation between classes and other interest groups is never explicitly raised . However, since it is the concept of interests that alone establishes the link between economic and political levels, they can only be distinguished from one another on the basis of the kind of interest represented . A class must, in the first instance, be distinguished from other interest groups on the grounds that it is constituted on the basis of an economic interest . (If economic interests are not privileged there is no way of distinguishing the representative of a class or class fraction from any other kind of pressure group, and so of distinguishing a Marxist from a political sociological analysis) . Of course the interests of a class or class fraction will also be determined by political and ideological factors, but the class or fraction must be constituted on an economic basis (6) . Since it is the particular kind of interest that defines a grouping as a class for the fractionalist approach, the definition of a class interest must be independent of the concept of class itself, if the account is not to be tautological . Thus a class interest must be a particular kind of individual economic interest, on the basis of which individuals coalesce into classes. Thus a class interest is characteristically defined by the specific features of the factor of production that appears to constitute the source of the income of members of that class : a class is defined by its ownership of that particular factor of production that defines its particular interest . Thus the proletariat is the class that owns labour power, the capitalist class that which owns means of production, and the landowning class is that which owns land . The definition of fractional interests then involves subsidiary determinations that may include ideological and political criteria . Thus for this approach classes and class fractions are defined in terms of economic interests but only appear at the political level . Hence the relationship between the economic and the political is necessarily treated as the simple and direct representation of economic interests by political groups . The only alternative to this economism is to abandon the privilege accorded to the economic altogether and adopt a pluralist approach that simply adds directly political and ideological interests to economic ones at the cost of losing the distinction between classes and other social groups (which is precisely the way in which Hindess and Hirst have now developed this type of analysis) . It is the idea of classes and class fractions as groups of individual interests whose social reality is only constituted at the political level that prevents this approach from achieving the understanding of scoiety as a "structured totality" that is its professed aim . At this point in the argument these charges may seem far-fetched : isn't it unreasonable to accuse an approach of adopting this general conception of society simply because it takes as its starting point the activities of individual capitals, and the organisation of those capitals into political groups? But these charges can be sustained, firstly by looking at Marx's own critique of the ideology of "revenue sources", and then by looking at the problems this approach faces when it tries to account for the activities of the state .



CAPTIAL, FRACTIONS OF CAPITAL AND THE STATE 4.

39

REVENUE SOURCES AND THE 'TRINITY FORMULA'

The problem with the fractionalist approach is that it remains at the level of appearances, it adopts the "conception of the agents of bourgeois production who are entrapped in bourgeois production relations" (Capital III, p .797), who come to see the 'factors of production' as their revenue sources and see their interests as distinct, independent of one another, and related not to their membership of social classes but to the ownership of factors of production with particular properties and with miraculous powers . In Chapter XLVIII of Capital valume III Marx offers a thorough critique of the "trinity formula" and the associated doctrine of revenue sources, a critique in the sense that he shows the origin of the illusion as well as revealing its irrationality . One cannot explain the concept of class on the basis of the concept of interest because interests are themselves dependent on social classes : revenues accrue to individuals not as owners of discrete factors of production, but as members of social classes who share as such in the distribution of the social product . Marx makes this point very clearly in criticising the illusion of the "trinity formula" : the "revenue sources" cannot themselves be conceptualised prior to class relations because they have social existence not as things, but as social relations . Thus : "capital is not a thing, but rather a definite social production relation, belonging to a definite historical formulation of society, which is manifested in a thing which lends this thing its social character . . . It is the means of production monopolised by a certain section of society, confronting living labour-power as products and working conditions rendered independent of this very labour-power, which are personified through this antithesis in capital . . . Wage-labour and landed property, like capital, are historically determined social forms ; one of labour, the other of monopolised terrestial globe, and indeed both forms corresponding to capital and belonging to the same economic formation of society ." (Capital III, pp .794-5) . It is not the factors of production that are the sources of revenue, for it is only because of the social forms within which these natural phenomena are inserted that they attract revenues to their owners in the sense that they become "sources by means of which one portion of value is transformed into the form of profit, another into the form of rent, and a third into the form of wages" (Capital III, pp .805-6) . However Marx also shows that in the series of mediations that transform social relations into the "enchanted and perverted world" of appearances (Capital III, p .806) . The social character of these relations is transformed and even capital comes to be seen as a thing : "capital, like land and labour, is simply considered as a material substance, that is, simply as a produced means of production, and thus is abstracted both as a relation to the labourer and as value ."

(Capital III p .803)



40

CAPITAL AND CLASS

This is precisely what the fractionalist account does in taking the individual at the economic level as its starting point, for it is only as a "material substance" that particular capitals can be seen independently of one another . From the social point of view, as capitals, particular capitals only exist as part of the total social capital, and it is only as such that they share in the distribution of surplus value that is produced in the exploitation of the working class by the capitalist class . Thus it is only by adopting a 'naturalistic ' conception of capital that it is possible to conceive of classes as associations of individuals, and to see capital as acquiring its social form, its class character, through political organisation . (7) Marx clearly brings out the connection between the naturalistic conception of capital that underlies the 'trinity formula' and the dissociation of production and distribution typical of the fractionalist approach . "The so-called distribution relations, then, correspond to and arise from historically determined specific social forms of the process of production and mutual relations entered into by men in the reproduction process of human life . The historical character of these distribution relations is the historical character of production relations, of which they express merely one aspect . . . The view which regards only distribution relations as historical, but not production relations, is, on the one hand, solely the view of the initial, but still handicapped, criticism of bourgeois economy . On the other hand, it rests on the confusion and identification of the process of social production with the simple labour-process, such as might even be performed by an abnormally isoolated human being without any social assistance ." (Capital, III,p .861) Marx spells out the implications of this concept of class : "Landed property, capital and wage-labour are thus transformed . . . into actual sources from which these value portions and respective portions of the product in which they exist, or for which they are exchangeable, arise themselves, and from which therefore, in the final analysis, the value of the product itself arises ." Capital I11, pp .805-6) Having noted that "it is the great merit of classical economy to have destroyed this false appearance and illusion", he goes on : "Nevertheless even the best spokesmen of classical economy remain more or less in the grip of the world of illusion which their criticism had dissolved, as cannot be otherwise from a bourgeois standpoint . . . On the other hand, it is just as natural for the actual agents of production to feel completely at home in these estranged and irrational forms of capital-interest, land-rent, labour-wages, since these are precisely the forms of illusion in which they move about and find their daily occupation . It is therefore just as natural that vulgar economy, which is no more than a didactic, more or less dogmatic, translation of everyday conceptions of the actual agents of production, and which arranges them in a certain rational order, should see precisely in this trinity, which is devoid of all inner connection,



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

41

the natural and indubitable lofty basis for its shallow pompousness . This formula simultaneously corresponds to the interests of the ruling classes by proclaiming the physical necessity and eternal justifiation of their sources of revenue and elevating them to a dogma ."

(Capital 111, pp .809-10) For Marx it is necessary to break with the world of appearances at the beginning . This is why we can argue that the errors of the fractionalist approach are already inscribed in the apparently innocent starting point they adopt . Marx offers a quite different conception of class from that espoused by the fractionalists . For Marx class relations are analytically prior to individual relations . From a Marxist point of view the individual capital cannot be the starting point, nor can the capitalist class be understood as the sum of individual capitals . A class is correspondingly not an interest group, defined as a coalescence of individuals with a common interest, but is an inseparable part of a relation of production and as such is analytically prior to the individuals who comprise it . Thus "every individual capital forms, however, but an individualised fraction, a fraction endowed with individual life, as it were, of the aggregate social capital, just as every individual capitalist is but an individual element of the capitalist class" .

(Capital III, p .351) It is only as part of the total social capital that individual capitals exist as capitals, for they cannot exist independently of social capital (capital-in-general), or of the class relation between capital and labour . It is this concept of capital-in-general that the fractionalist analyses lack, yet without such a concept it is impossible to conceptualise particular capitals as capitals . Marx is quite clear about the question of analytical priority : "To the extent that we are considering it here, as a relation distinct from that of value and money, capital is capital in general . . . we are still concerned neither with a particular form of capital, nor with an individual capital as distinct from other individual capitals etc . . . The latter relations are to be regarded as developments coming out of this germ . But it is necessary to establish the specific form in which it is posited at a certain point. Otherwise confusion arises ." (Grundrisse, p .310, Original emphasis) . (8) Thus the workers exploited by each particular capital are exploited as a part of the working class by a part of the capitalist class : the class relation between capital and labour is the precondition of the particular relation between a specific capital and the workers exploited by that capital . Correspondingly, class relations are prior to the specific economic, ideological and political forms taken by those relations, and are not the product of the association of individual members of the class . Classes, or more precisely class relations, are the necessary starting point for a Marxist



CAPITAL AND CLASS

42

analysis, without which relations of exploitation (as the economic form of those class relations) or the relations of domination (as the political form of those relations) cannot be conceptualised . It is the concept of class relations as being analytically prior to the economic, political or ideological forms taken by those relations (even though class relations have no existence independently of those forms) that makes it possible for a Marxist analysis to conceptualise the complexity of the relations between economic and political, their interconnections as complementary forms of the fundamental class relation, without abandoning the theory for a pragmatic pluralism . It is now time to return to the fractionalist accounts of the South African state in order to see how the fundamental theoretical distortion that I have identified affects the analyses offered, before indicating how the theoretical guidelines laid down in this section can inform a more adequate understanding . 5.

FRACTIONALISM AND THE STATE

I have so far concentrated my criticism of fractionalism on the conception of the economic and the political and of the relations between the two that underlies the concept of a 'class fraction' itself . I now want to show that the same criticisms can be extended to the fractionalist conception of the state, and in particular that the overpoliticisation of the analysis eventually leads to the dilemma of choosing between pluralism and economism . For the fractionalists the state is, in the first place, the arena in which particular interests are represented, in which they compete for state power . This struggle for access to state power is, for fractionalism, of fundamental importance to each and every capital because of the role of the state in relation to the valorisation process . Instead of the state being set in the context of the valorisation process, the latter is seen as determined in important ways by the state . In particular the separation of the economic and the political is paralleled by a separation of the production of surplus value from its distribution ; the production of surplus value takes place in the 'economy', while its distribution is necessarily mediated through the state . Thus Kaplan sees the "realisation of surplus" as necessarily passing through the state, so that the conflict between fractions of capital concerns "the question of surplus reallocation and investment" : "At the level of the capitalist mode, the metropolitan bourgeoisie needed to exercise hegemony to ensure that surpluses earned would be 'repatriated' and/or retained by the mining companies to finance their own further development ." (Kaplan 1975, pp .1,3,7) . Thus it is the overpoliticisation of the analysis, by subordinating the distribution of surplus value to the state, that gives the struggle between fractions of capital its importance and that makes the question of which fraction achieves dominance in that struggle so central to an explanation of the pattern of accumulation of capital itself . (9) This overpoliticisation



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

43

is only possible because the more fundamental link between the distribution of suplus value and its production is ignored . The outcome of the political struggle between fractions of capital is the dominance of a fraction or fractions within the unity of the 'power bloc' constituted by the state . The bulk of the work of the fractionalist school is concerned with identifying the 'hegemonic fraction' in any given period . The main problem faced by the attempt to establish which fraction is hegemonic is that there are no clearly agreed criteria by which one can identify the hegemony of that fraction . Thus while the turning points in South African history are generally agreed, coinciding with changes of government, there is no more agreement about which fraction is hegemonic than there is about what the fractions are in the first place . While

it is generally agreed that before

1924

foreign and mining interests dom-

inated there is considerable disagreement about subsequent developments . Some believe that this dominance continued, while others argue that national, agricultural and industrial interests became dominant for a period

after 1924, slipped back around the Second World War, and returned to dominance in 1948 . There is a surprising degree of agreement that since the 1960's fractions of capital have become increasingly difficult to identify, as national capital is internationalised, international capital 'naturalised', as mining capital enters farming and industry, as Afrikaner capital learns English, and as medium capital gets big . The implication of this kind of account of interpenetration is that the power bloc becomes monolithic and the fractionalist account becomes indistinguishable from the orthodox Communist Party theory of state monopoly capitalism . The question of the identification of the hegemonic fraction can first be simplified by distinguishing, as the consistent fractionalists do, between political and economic hegemony (cf. Kaplan, 1976c) . As Morris notes : "the fraction that is economically dominant may very well not be politically dominant . In the case of South Africa this is very clear from Kaplan's work on the earlier period of capitalist development, where he was essentially arguing that while gold mining was economically dominant it was the national bourgeoisie that was politically dominant ." (Morris 1975, appendix) . Hence there is no necessary connection between economic and political dominance . It is not clear what 'economic dominance' might mean, but it seems simply to be a question of the relative size or relative growth of

the 'fraction' concerned (cf . Kaplan et al .

1977, p .22)

Certainly there is no

attempt to analyse the relations between particular capitals as parts of the total social capital to arrive at some understanding of 'economic dominance' . The radical separation of economic and political dominance implies that political dominance is something to be secured only at the political level, through political mechanisms . It is not altogether clear what the mechanisms are supposed to be . On the one hand, a change of government is usually interpreted as the expression of a reorganisation of the 'power



44

CAPITAL AND CLASS

bloc' and its 'representation' which may involve a change in the dominant fraction . In other words dominance shifts from the fraction represented by the dominant party in the old government to that represented by the dominant party in the new one . In this case 'political dominance' means that the political party that supposedly 'represents' the dominant fraction has won an election and perhaps constructed a parliamentary alliance necessary to form a government . Thus : "When the results came in on the 18th of June 1924 they indicated that his party had lost and the Pact parties had won . It was an event which had more than mere party political significance . It was an index that hegemony within the power bloc had passed from imperial to national capital ." (Davies 1977, p .155) . On the other hand, we find discussion of the relations between fractions conducted not in terms of the relations between political parties in Parliament, but in terms of the policies proposed by various pressure groups, and above all employers' associations, in relation to the policies actually adopted by the state . The dominant fraction of capital is then the one whose espoused policies predominate in practice : "The consensus of state policies, with respect to local industrialisation, with those of the Mines and Commerce, provide a firm indication of which fractions of capital were hegemonic under the Botha-Smuts government ." (Kaplan 1976a, p .76) "The implementation of such policies therefore depended upon, and provide us with an index of, the hegemony during this period of national capital ." (Kaplan et al . 1977, p .10) " . . . the relative benefits to the various fractions of capital during the war, reflect this tenuous manufacturing hegemony ." (Kaplan et al, p .22) This co-existence of different criteria of dominance would be all very well if they coincided . However it is not in general the case that the policies pursued by the state turn out to favour the fraction whose 'representatives' dominate in the representative apparatus . This is especially clear in South Africa where, insofar as the dominant political parties have represented particular capitalist interests, it has been relatively weak and vulnerable capitals that have dominated such representation . Thus the 1924 Pact government, supposedly representing the dominance of 'national capital', did not in fact pursue policies especially favourable to 'national capital' at the expense of 'mining' or 'foreign' capital . It was not until the 1932 Fusion government, which brought the political representatives of 'mining capital' back into government, that really serious measures were taken to redirect surplus value from mining to industrial and agricultural capitals . Similarly the 1948 Nationalist government took measures from the start whose inevitable and clearly foreseen consequence was to be the liquidation of the 'national capital' the regime supposedly represented . The lack of coincidence between the interests apparently represented



CAPITAL, FRACTIONS OF CAPITAL AND THE STA TE

45

in the state apparatus with the policies pursued by the state leads to enormous confusion in the attempt to identify the dominant fraction in a way that will not render the analysis ridiculous . In most work the problem is evaded . Thus the most recent summary account seems to give up trying to identify hegemonic fractions : we find that the outbreak of World War I I produced a "disorganised power bloc", while with the end of the war "the structure of hegemony collapsed . . . the power bloc was (still) disorganised" . In the wake of the 1948 election we find the "acute and continuing disorganisation of the power bloc" and that "the power bloc remained disorganised for much of the fifties" (Kaplan et al . 1977, pp .21,25,27,28) . In another recent work the contradictions which emerge from the fractionalist analysis of the post-1948 situation are dissipated as follows : "despite the fact that industrial capital did not itself exercise direct political hegemony during the early Apartheid period, the state contined (in the name of national economic development) to guarantee the conditions necessary for the emergence of the not yet given relation of monopoly capitalism in industry" (Davies 1977, pp .291-2) These kinds of evasion are, however, unacceptable . If the aim of the analysis is to explain the form of the state and the policies adopted by that state in terms of the political dominance of a particular fraction in the power bloc that dominance cannot in turn be identified by specifying the interests served by state policies without reducing the explanation to a tautology . It follows that the explanation of political dominance can only be in terms of the political processes through which parties are elected and governments formed . The failure of governments to carry out policies that unambiguously express the interests they supposedly represent brings us back to the problem of overpoliticisation . The fractionalist approach condenses everything in the state so that it is unable adequately to explain the limits within which the state operates, the factors that restrict its ability to intervene in the accumulation process in favour of a particular interest . The problem that has to be confronted is that of explaining the limits on the ability of a particular capital or association of capitals to use the state to secure its own interests at the expense of those of other capitals, limits that are set by the subordination of the distribution of surplus value to its production . Davies has tackled this problem in attempting to explain why the 1924 Pact government, representing the hegemony of national capital, in fact did so little to press the interests of that fraction against mining capital . He offers three explanations . firstly a tautological explanation : "mining capital remained even after its removal from the position of hegemonic fraction a powerful member of the power bloc ." (Davies 1977, p .159) . He then argues that national capital's interests were in fact those of mining capital too :



46

CAPITAL AND CLASS "National capital's own development depended upon the appropriation through taxation, protection, etc . of surplus value produced in the goldmining industry and it had thus itself a direct economic interest in guaranteeing the conditions for surplus value in the industry to be produced and ensuring that the demands of other classes did not seriously interfere with the industry's surplus value producing process ." (Davies 1977, p .159) .

This argument completely undermines the fractionalist enterprise . If it is the case that the interests of particular capitals coincide with the interests of all other capitals in this neat way, then there is no reason for engaging in fractional struggles in the first place, and it does not matter which fraction of capital is hegemonic . Davies finally offers a straightforwardly economistic solution to the problem by proposing a reductionalist identification of the state with capital's needs : "The demands of capital accumulation continued, during the Pact period as before, to dominate over the demands of all other classes" Davies 1977, p .169) . Thus he explains why the participation of the working class (or rather the "white wage earning petit-bourgeoisie") in the Pact government had no effect on the ability of the regime to serve capital as a whole : the State "despite the alliance between white wage earners and the hegemonic fraction, continued to act in defence of the essential requirements of capitalist production against the demands of the white wage earning classes . Its fundamental role in the economic class struggle remained to defend the power bloc against the potential political threats posed by the struggle of other classes, including those of the white wage earning allies of the hegemonic fraction ." (Davies 1977, p .169) Thus we find the fractionalist couple of overpoliticisation/economism reappearing : the fractionalist approach is unable to explain the limits on the ability of the state to intervene in the interests of particular capitals (or of any other particular interests) with outfalling back into afunctionalist reductionism that simply identifies the state with the needs of capital . This reductionism is inherent in the understanding of capital and of the state characteristic of the functionalist approach : the approach is unable to conceptualise the limits on the intervention of the state because it has no concept of capital-in-general independent of the state itself . While fractions of capital exist at the political level through their political organisations, such organisations cannot constitute the unity of capital as a whole, since such organisations are based on the conflict between capitals . Thus, for the fractionalist approach, the unity of capital can only be created by an institution that stands above all these particular capitals . This is precisely the function of the state : the essential role of the state in its relation to capital is precisely to create this unity (while its role in relation to the working class is to create disunity) . Without the state capital as a



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

47

whole cannot exist, for there would simply be a number of feuding capitals . Hence capital as a whole, as a class, can only exist in and through the state . The state, for this approach, has therefore to replace Marx's concept of social capital or capital-in-general in explaining the way in which the relations between particular capitals are regulated . Thus the state becomes the central and determining institution of capitalist society, responsible in its essence for guaranteeing capitalist reproduction . Thus : "the task of politically organising the dominant classes falls to the state because, unlike the working class, the bourgeoisie is characteristically incapable of raising itself to the level of political domination through its own political parties . This is principally due to the division of the bourgeoisie into fractions with contradictory interests : a division which begins at the level of relations of production . . . Without the bourgeois state and all its apparatuses the bourgeoisie would be in capable of place its political interests . . . above the narrow fractional interests of its components ." (Davies 1977, p .30) .

For all the talk of the 'relative autonomy' of the state, we find that the state cannot but express the dominance of capital, that it is only through the state that capital is dominant (and it is significant that class relations for this approach are always described as relations of 'domination' rather than 'exploitation') . Thus the 'autonomy' of the state is only an autonomy in relation to particular capitals, whose attempts to press their particular interests, as member of fractions of capital, on the state are bounded by the attempts of other capitals, on the one hand, and by the state as representative of the unity of capital, on the other . This is why fractions of capital themselves only have existence and meaning in relation to the constituted state . Without the state there would be no capitalist class, merely particular capitals pursuing their private interests . This is also why the fractionalist accounts concentrate almost exclusively on the struggle between fractions of capital, and largely ignore the struggle between capital and labour . Since there is "a congruence of interest, in respect of labour, by all elements of capital" there is nothing to explain, the state is simply the agent of capital in its suppression of labour : in other words, the state is simply and immediately identified with capital-in-general . On the other hand, when we introduce the question of the "differentiation between the different classes of capitalists" we find that the interests of different capitals conflict, the conflict arising above all over "questions of surplus reallocation and investment" (Kaplan 1975, p .1) . It is here that the state has a relative autonomy, since it has to adjudicate between various interests . The working class is thus largely excluded from the class struggle at the level of the state : "The contradictions most directly and acutely reflected within the state are those among the dominant classes and fractions and between these and their supporting classes, far more than the contradictions



48

CAPITAL AND CLASS between the power bloc and the working class . The latter are basiccally expressed in the bourgeois state 'at a distance' . . ." (Poulantzas 1976, p .104 ; quoted approvingly by Davies 1977, p .74 n .90) .

Hence the 'class struggle' that plays such a central role in the rhetoric of fractionalism is actually the struggle between fractions of capital . The working class intrudes almost exclusively insofar as it affects the interests of a particular fraction of capital either as a 'supporting class' or as the object of state policy (10) . Hence the working class plays no direct role in the class struggle, but is rather treated as the object of the policy of the state . The 'labour question' is a matter for negotiation between the representatives of various fractions of capital . Once the labour policy of the state has been determined through this struggle, it is implemented by a state which appears in relation to the working class as the monolithic and undifferentiated power of capital . The end result is a view of the state in relation to the working class as omnipotent, monoplising the means of violence, commanding almost unlimited economic resources, controlling extensive ideological apparatuses, all of which can be directed to the suppression of unconstitutional action by the working class, and the emasculation of any constitutional offensive . (11) 6

THE FRACTIONALIST ACCOUNT OF THE APARTHEID STATE

The form of state, as much as particular state policies, is seen as the product of the struggle between fractions of capital and is determined essentially by the form in which the dominance of the dominant fraction is institutionalised . Hence the Apartheid state is not explained as a particular form of the struggle between labour and capital, but rather as a particular form of the struggle between the fractions of capital . The key to an understanding of the Apartheid state is to be found for the fractionalists, in an understanding of the 1924 election that established the dominance of 'national capital', a dominance subsequently eroded, but renewed in 1948 with the formation of the Nationalist government . A considerable fractionalist literature has now developed around the 1924 election, but the framework remains that originally defined by Kaplan . The attainment of hegemony by the 'national bourgeoisie' cannot easily be explained in terms of the relations between fractions of capital because national 'industrial capital' was economically insignificant as an independent force, while agricultural and commercial interests were divided, the economically most advanced tending, if anything, to support mining interests . Thus Kaplan starts off by asserting something that many others would doubt, that a national bourgeoisie did actually exist : "Both elements of the bourgeoisie in South Africa . . . established an independent economic base grounded in productive activity and as capitalist elements they sought the expanded reproduction of that base ." (Kaplan 1975, pp .] 3-14) The economically weak agrarian capital and emerging "urban-based" bourgeoisie were able to call on the support of the white working class to



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

49

challenge the foreign mining interests . This is because of the supposed role of the state in the reallocation of surplus : the national bourgeoisie could sustain economic growth without undermining the privileges of the white working class by taxing away the surplus of the mining capitals . Hence "the objective basis for an alliance between white workers and the indigenous bourgeoisie lay in the conflict over surplus . . . State power could now pass under the hegemony of an ascendent bourgeoisie who, although in alliance with a 'labour aristocracy' were not 'blocked' in their revolutionising of the means of production . . . Alone the 'indigenous' bourgeoisie could not have ended the domination of the 'metropolitan' bourgeoisie . The mobilisation of white workers was an essential prerequisite for ending such hegemony (Kaplan 1975,

pp . 16-17) I do not want to go into the question of the adequacy of this analysis

as an explanation of the 1924 election (12) . Suffice it to say that all the errors and ambiguities of the approach that I have already noted come to the surface in an unquestioned identification of political parties with class fractions, and a radical separation of the production of surplus value from its distribution that dissolves the limits on the ability of the state to intervene in the latter . While recent work (notably Davies 1977) has shown how mining capital had compromised itself in the eyes of the white working class by attempting to undermine the privileges of the white workers, there is no explanation of why it should have been the 'national bourgeoisie' that managed to pick up the pieces, rather than any other interest group (whether capitalist or not), or even the white working class itself . Nor, indeed, is there any explanation of why the mine-owners themselves had not used their earlier dominance to shift the burden of adjusting to the post-war crisis from themselves and 'their' white workers onto non-mining capitals . In the end, as always, the explanation falls back on an economistic reductionism : national capital had to take power because that was the only political solution, for capital, once mining capital had compromised itself . Thus national capital "put itself forward and achieved hegemony at least partly on the basis of its particular relationship with white wage earners and its unique ability to restore order and social stability ." (Davies 1977, p .170) "National capital had achieved hegemony at least partly because it had been able to set itself up within the power bloc as uniquely placed to organise a solution to the political problems posed for the power bloc by the white wage-earning classes ." (Davies 1977, p .1 79) The identification of capital with the state once again makes it impossible for this approach to theorise the relation of the state to capital in a nonreductionist way . The importance of this account is the general conclusion that is drawn



50

CAPITAL AND CLASS

from it, from which is developed the fractionalist account of the Apartheid state : "Support for 'the state' by the 'white working class' is absolutely vital to the ability of one or other fraction of the bourgeoisie to exercise hegemony ." (Kaplan 1975, p .19) The result of the formation of the Pact government was that a block was placed on the threatened erosion of the existing privileges of the white working class, and indeed the institutional apparatus of segregation was extended progressively to new areas in order to maintain working class support for the state - the 'civilised labour' policy that particularly benefitted poor whites (and that incidently had the backing of a// sections of capital) . In fact the period after 1924 did not see major advances of the white working class as a whole with the result that white working class support for 'national capital' was progressively eroded . Thus "From 1928 onwards, national capital was in alliance with only a fraction of the white new petit bourgeoisie . A considerable proportion of the rest of the new petit-bourgeoisie and white working class was however, being dominated by a racist and parliamentary cretinist ideology, still in the position of a supportative class - though to the form of state rather than to national capital exclusively ." (Kaplan et al . 1977, p .13) . The steady erosion of working class support for those in power prepared the way for the 1948 election, which is interpreted essentially as a re-run of 1924, with slightly different actors : manufacturing capital is no longer clearly 'national' and the new hegemonic fraction is variously identified as 'Afrikaner capital' and 'agricultural capital' . This time manufacturing capital compromised itself with the white wage-earners by threatening to undermine their position in the course of the transition from 'competitive' to 'monopoly' capitalism associated with the process of secondary industrialisation . The result was that it was Afrikaner/ agricultural capital that was able to promise the white workers the earth in order to achieve the position of hegemonic fraction, and that subsequently guaranteed and even extended the system of segregation with the development of the Apartheid state . Thus "The Apartheid state, in short, remained fundamentally a capitalist state but one in which certain fractions of the white traditional and intellectual petty bourgeoisie were able to 'advance' themselves in various ways and one in which the classes represented by the governing party were critically dependent on the support of their allies within the white wage earning classes ." (Davies 1977, p .295) The form of the Apartheid state is therefore determined by the need of a particular fraction of capital to secure working class support to raise itself to the position of hegemonic fraction . The development and pros-



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

51

pects of Apartheid are therefore determined by the course of the struggle between such capital fractions . The condition for this to be possible is located in the need for a racially organised division of labour at a certain stage in capital accumulation in the specific circumstances of South Africa (a need that has been perpetuated at least since 1924 by the policy of racial division itself) . Thus Davies concludes his thesis as follows : "The overriding conclusion which emerges from this study is that white wage earners came to occupy 'privileged' places in a racial division of labour primarily to meet the requirements of capital accumulation at the various phases and stages of capitalist development and secondarily through the actions of a capitalist state, acting under particular conditions of class struggle and within the limits of the capitalist mode of production, to defend or advance the political class interests of a power bloc of capitalist fractions at different conjunctures ." (Davies 1977, p .319) The result is that the debate about the nature of Apartheid has now come full circle . Because they base themselves at the level of appearances, because their analysis does not penetrate the political level to reveal it as the political form of more fundamental class relations, the fractionalist accounts quite unconsciously, but also quite inevitably, come to reproduce the standard liberal accounts of the development of Apartheid with which the contributions of Wolpe, Legassick and Williams originally broke . The most important feature of those contributions was that they broke decisively with the view that Apartheid could be explained in terms of the conflict between different capitalist interests, between advanced capitalists who sought to undermine the system and reactionary capitalists who sought to preserve it, the latter being maintained in power through the support of the white working class. Instead they argued that Apartheid had to be understood first and foremost in terms of the struggle between capital and labour and that the Apartheid state had to be understood not as the product of some kind of compromise with a section of the working class, but rather as the form taken by the domination of capital in the specific circumstances of accumulation in South Africa . (13) The fractionalist accounts represent an improvement on many of the liberal treatments firstly in that they amass a great deal of new empirical material and secondly in that they do not put the white workers' defence of their privileges down to a blind racism (although cf. Kaplan et al . 1977, p .13 quoted above) . The fractionalist accounts do seek to explain the position of the white working class in terms of their 'material interests', but the analysis of those interests is an abstract and mechanical analysis and not an historical materialist one . It is based on a radical separation of the conditions under which class struggle takes place and the class struggle itself. The former, the material foundation of the struggle, plays a purely passive role, it is external to and prior to the class struggle, acting as the inert material constraint imposed by the simple facticity of the world treated as a quasi-natural given . Thus Morris, for example, argues for the importance of "the class struggle and the decisive role of the state" and goes on to argue that



CAPITAL AND CLASS

52

"it is . . . always the class struggle which has primacy and which determines the actual outcome of the transition . . . The conditions under which the class struggle takes place set the limits to the conjuncture . They define the range of possible conceivable outcomes . They, however, neither specify the necessary outcome - nor even the particular path/road that the transition will take. It is only the concrete class struggle that is the motor of history ." (Morris 1976a, p .1 52) . We have therefore to analyse the "conjuncture" which is characterised by a "major structural contradiction" quite separately from the "rhythm of the class struggle", the former being the context "which set the contradictions of the class struggle" (Morris 1976a, p .153) . Thus the material foundation is seen as something external to the class struggle, and the interests of the white working class are analysed not as the interests of a part of the working c/ass, but as 'owners' of a particular revenue source whose interests are defined by the characteristics of that source . Thus the analysis remains within the abstract materialism of the better liberal accounts : abstract in the precise sense that it detaches the material foundation from its socio-historicial forms . It should not be surprising to find that the self-appointed scourges of "neo-Marxism" focus their attention on the accounts of Wolpe, Legassick and Williams, while scarcely diverging at all from the fractionalists in their own account of 1948 (Kantor and Kenny 1976, pp .31-3) . This leads Johnstone in his recent reply to go so far as to note that "virtually all the points they make, major and minor, are not only not criticisms and repudiations of this approach ; many of them are actually made by neo-Marxist work ." (Johnstone 1977, p .2) Unfortunately this is only too true . (14) 7.

THE LIMITS OF POLITICS : CAPITAL AND STATE

The fractionalists based their original critique of Wolpe and Legassick on the fact that both saw Apartheid basically in terms of an undifferentiated relation of capital to labour . Williams raised the question of seeing it in more differentiated terms, looking at the relation between departments within the reproduction of the total social capital : "In order to avoid a formal approach it is essential that we analyse the development of the capitalist mode of production from the standpoint of the total social capital, bearing in mind that capital only becomes a self-generating process when there exists a determinate and mutually dependent relationship both within and between Departments I and Ila, that is, the departments which are able to produce and replace the elements of constant and variable capital ." (Williams 1975, p .8) Williams's contribution set the analysis on the right lines, but suffered



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

53

from major weaknesses : in particular, it attempted to explain the relations between mining and industry that underlay the differentiated labour system in terms of a false argument about the character of gold mining instead of explaining it in terms of the relations between departments and between particular capitals in the reproduction of total social capital . Williams didn't theorise precisely how total social capital structures the relation between different departments and between particular capitals . Thus Williams's account left a gap in the analysis which made it easy for fractionalists to forget about total social capital altogether . As Wolpe and Legassick have themselves recently noted : "much of the subsequent discussion of the roots of state policy in this period has missed the point by seeing it in terms of the competing labour requirements of different sectors/fractions of capital, rather than situating such an examination in terms of a prior analysis of the conditions of accumulation of total social capital . Indeed, the ways in which state and Bantustan institutions operate to structure and control the flow of labour . . . is of central importance . But the possibility for the existence of such severe forms of extra-economic coercion is constituted by the conditions of accumulation of total social capital ." (Wolpe and Legassick 1977, p .102) The urgent task is to return to the path opened up by Williams's contribution by raising directly and explicitly the question how does capital- ingeneral structure the relations between particular capitals? Marx himself really only considered the relations between capitals in terms of economic relations, and even this he did not do in depth . Nevertheless his account is sufficient to provide us with a starting point . The first point to make clear is that capital-in-general is not to be understood as a form of capital alongside particular capitals, imposing itself on particular capitals with some ruthless logic . Capital-in-general only exists in the form of particular capitals and their interrelations . On the other hand, capital-in-general is not to be understood in terms of the relations between particular capitals as distinct from those capitals themselves . Hence capital-in-general is not external to the particular capitals, for, as I have already argued, it is only as forms of capital-in-general that particular capitals exist as capitals : they can only be seen as differentiated parts of a social whole . From the point of view of an existing particular capital, capital-ingeneral makes itself felt in the economic sphere through competition with other particular capitals . Again it must be stressed that this competition does not express the external determination of a particular capital, but rather it expresses the inner nature of that capital as a social phenomenon in the form of an external constraint . Thus competition is the fetished

form in which the already social character of the particular capital is impressed upon the capitalist . It is because capital appears in the fetishised form of a thing that its social character has to appear to it as something alien to it . This is why to remain at the level of appearances is to invert the real relationships so that

54

CAPITAL AND CLASS "competition has to shoulder the responsibility of explaining all the meaningless ideas of the economists, whereas it should rather be the economists who explain competition ." (Capital III, p .844) .

Hence : "conceptually, competition is nothing other than the inner nature of capital,

its essential character, appearing in and realised as the recri-

procal interaction of many capitals with one another, the inner tendency as external necessity .) (Capital exists and can only exist as many capitals, and its self-determination therefore appears as the reciprocal interaction with one another .)" (Grundrisse, p .414)

"Free competition is the relation of capital to itself as another capital, i .e . the real conduct of capital as capital . . . Free competition is the real development of capital . By its means what corresponds to the nature of capital is posited as external necessity for the individual capital ; what corresponds to the concept of capital, is posited as external necessity for the mode of production founded on capital ."

(Grundrisse, pp .650-1) The concept of capital-in-general does not dissolve the independence of particular capitals, but rather describes the limits within which the independence of particular capitals exists . For Marx the concept expresses the fact that these limits are specifically social limits . The limits within which

capital-in-general is bounded, and which press upon particular capitals, are limits on the ability of capital to extract and appropriate surplus value . It is only insofar as they are transformed in this way into social limits that natural barriers on capital are effective as barriers . Thus in his consideration of reproduction schemes in Capital volume II Marx shows how the requirement that "there exists a determinate and mutually dependent relationship both within and between Departments I and Ila" (Williams, 1975, p .8) that depends on the specific character of

the use-values produced in each department, is imposed on particular individual capitals in the form of competition . Marx thus shows how a constraint that expresses the material nature of the product is transformed into a social form, and he further shows that the form is adequate to the extent that it can, although it need not, establish the required relationship between departments . For bourgeois economics, by contrast, all limits to capital are treated as natural limits, and competition is treated as the eternal mechanism by which such natural limits affect the individual capital . This is why the materialism of bourgeois economics is a metaphysical, naturalistic materialism . For the fractionalists too, as we have seen, the limits are imposed on individual capitals not through competition, but through the state . The state is thus responsible for regulating the relations between capitals, and so for translating the limits, seen in naturalistic terms, on capital accumulation into constraints on particular capitals : the state plays the role of the Marxist concept of capital-in-general . This is why



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

55

the relations between capitals are treated in terms of a distributional struggle over the apportioning of the available surplus value, for the state intervenes between the production and the distribution of surplus value . Thus the fractionalist approach ignores the fundamental question of the different role of different capitals in the production of surplus value . Marx discussed the relations between particular capitals only in terms of competitive relations . He did not provide a rigorous analysis of the role of the state in regulating the relations between particular capitals . Nevertheless it is surely clear that the concept of capital-in-general is as fundamental to the Marxist analysis of the state as it is of the economy . The concept of capital-in-general is a concept that is analytically prior to the specification of particular capitals and of the state, it is the concept that makes it possible to theorise the social limits to the relations between particular capitals and to the activity of the state . By understanding the limits, the concept of capital-in-general makes it possible to understand the forms that capital imposes on the state . It is the concept of capital-ingeneral that enables us to theorise the objective limits imposed on the state that we refer to when we call it a capitalist state, the state of a society dominated by capitalist relations of production, and so that enables us to theorise the limits on the ability of particular capitals to press their particular interests on the state successfully . Normally the relations between capitals, and so between departments, are regulated by competition and the ability of the state to intervene in these relations without disrupting accumulation is limited . There are normally severe limits on the ability of the state to intervene to redirect surplus value . These limits are not given, as the fractionalists would have us believe, by the total available surplus value and the relative political strength of the various parties involved in political conflict. They are rather given by the role of surplus value in the expanded reproduction of the capitalist mode of production . Thus in South Africa in the 1920s and 1930s gold mining capitals could only, be super-taxed without undermining accumulation in gold mining (and so the accumulation of capital as a whole) to the extent that they were earning super-profits . This intervention was in fact possible because mining capitals did not see their superprofits eroded : they were protected by their monopoly position from the tendency for competition to impose the equalisation of the rate of profit . Thus the mining houses paid what were essentially royalties to the state in the form of an excess profits tax . In fact in the 1920s in South Africa such super-profits were limited . Hence the ability of the state to intervene to the disadvantage of gold-mining capitals was quite limited and so the state was limited in its ability to divert surplus value towards capitals active in industry and agriculture . On the other hand in the 1930s, following a devaluation whose effect was to divert surplus value towards gold mining capitals, state intervention on behalf of industry and agriculture did become possible on a large scale, despite the fact that the new government included the 'representatives' of mining capital, who were pressed to join the government precisely to oversee the interests of mining capitals in the new circumstances of the depression . Although the mining houses opposed this state intervention, they subsequently came to

56

CAPITAL AND CLASS

recognise that it had not been altogether to their disadvantage, as Williams noted (Williams 1975, p .14) : as a differentiated part of the total social capital mining capital may have gained more from the boost given to accumulation of social capital by the redistribution of surplus value than it lost in that redistribution itself. It can, and increasingly does, happen that competition fails to regulate the relations between capitals in accordance with the requirements of the accumulation of the total social capital : imbalances between departments that can prejudice sustained accumulation, for example, may not be adequately or sufficiently rapidly rectified through the operation of the market and pressure may develop for the state to intervene by one means or another . The kind of measures that have to be taken after a devaluation in order to neutralise undesirable redistributional effects are a clear example of this kind of intervention . Fiscal measures to favour capitals with a high organic composition, or capitals particularly vulnerable to foreign competition, or to counter some aspects of the centralisation and concentration of capitals offer other examples . Alternatively the state may intervene directly, through nationalisation, through direction of labour, through licensing systems or by any of a large number of other means to regulate the relations between particular capitals . In order to theorise state intervention on behalf of particular capitals it is necessary to go behind the level of appearances to consider the relation between capital-in-general and particular capitals, on the one hand, and capital-in-general and the state, on the other, and to consider these relations in the context of the fundamental class relation of capitalist society which is always that between capital and labour . Since particular capitals press interests that are the interests of a differentiated part of the total social capital, and since measures taken by the state in the immediate interests of particular capitals also have implications for total social capital, it is impossible to disentangle the particular from the general without a theory that can specify concretely the relation between the two in terms of socio-historical relations, a theory that can locate their interconnection in terms of the specific institutional forms through which it is mediated instead of in terms of an abstract discussion of particular and general capitalist interests . In the closing sections of this paper I want to indicate the ways in which a Marxist theory can do this . 8

CAPITAL-IN-GENERAL AND FRACTIONS OF CAPITAL

The 'fractions' that play such a prominant part in the fractionalist accounts are not in fact fractions of capital at all, in the sense of being differentiated parts of a whole, for there is no concept of the whole prior to its parts . On the other hand, for Marx all particular capitals must be seen as 'fractions' in the sense that all are parts of a whole . However the parts of the whole can be differentiated in various ways, not all of which have the same status . In each case, however, the 'fractions' have to be defined in terms of the various objective criteria of social differentiation, and not on the basis of the more or less contingent coincidence of independently constituted particular interests .



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

57

When Marx discusses the differentiation of capital-in-general he makes a clear distinction between specifying the particularity of individual capitals as one individual capital "as distinct from other individual capitals" and specifying it in terms of the "particular form of capital" (Grundrisse, p .310) . The "particular form" of capital refers to the specialised "functions" in the circuit of capital which at a certain period of capitalist development give rise to specialised capitals : productive capital, bank capital, commercial capital, landed capital : "in commercial and money-dealing capital the differences between industrial capital as productive capital and the same capital in the sphere of circulation are individualised through the fact that the definite forms and functions which capital assumes for the moment appear as independent forms and functions of a separate portion of the capital and are exclusively bound up with it ." (Capital, 11, p .318) These different fractions of capital cannot be treated on a par with one another, they do not simply represent a number of distinct capitalist interests, because the different fractions relate differently to the production of surplus value, and so to their existence as parts of capital-in-general . Thus only productive capital participates in the production of surplus value : "Industrial capital is the only mode of existence of capital in which not only the appropriation of surplus value, or surplus-product, but simultaneously its creation is a function of capital . Therefore with it the capitalist character of production is a necessity . Its existence implies the class antagonism between capitalists and wage-labourers The other kinds of capital, which appeared before industrial capital amid conditions of social production that have receded into the past or are now succumbing, are not only subordinated to it and the mechanisms of their functions altered in conformity with it, but move solely with it as their basis, hence live and die, stand and fall with this basis . Money-capital and commodity-capital, so far as they function as vehicles of particular branches of business, side by side with industrial capital, are nothing but modes of existence of the different functional forms now assumed, now discarded by industrial capital in the sphere of circulation - modes which, due to social division of labour, have attained independent existence and been developed one-sidedly ." (Capital, 11, p .55 . Note that the term 'industrial capital' is here used to refer both to 'industrial capital' as social capital in the capitalist mode of production and to 'productive capital' as a moment in the circuit of industrial capital) . Money- and commercial-capital can only be understood as capitals in relation to the circuit within which they function : "It is only in connection with the entire process . . . that capital in the process of circulation appears as capital (whereas in the process of



58

CAPITAL AND CLASS production it appears as capital through the subordination of the labourer to the capitalist and the production of surplus-value) ."

(Capital I11, pp .336-7) . Landed capital has no connection with the production of surplus value or with its realisation, but participates only in its distribution . Marx in fact argues, explicitly in the Grundrisse and more ambiguously in Capital, that bank capital occupies a special position because it is capital existing "in its elemental form as capital" and as such is capital in its "general form, although belonging to individual capitalists" . (Grundrisse, p .449) ; Thus Marx argues that capital-in-general exists not only,as we have considered it so far, as an abstraction "which grasps the specific characteristics which distinguish capital from all other forms of wealth - or modes in which (social) production develops" but also that "capital in general, as distinct from the particular real capitals, is itself a real existence . This is recognised by ordinary economics, even if it is not understood, and forms a very important moment of its doctrine of equilibrations etc . For example, capital in this general form, although belonging to individual capitalists, in its elemental form as capital, forms the capital which accumulates in banks or is distributed through them, and, as Ricardo says, so admirably distributes itself in accordance with the needs of production . Likewise through loans, etc, it forms a level between the different countries . . . While the general is therefore on the one hand only a mental (gedachte) mark of distinction (differentia specifica), it is at the same time a particular real form alongside the form of the particular and the individual ." (Grundrisse, P .449) "In the money-market only lenders and borrowers face one another . The commodity has the same form - money . All specific forms of capital in accordance with its investment in particular spheres of production or circulation are here obliterated . It exists in the undifferentiated homogeneous form of independent value - money . . . It obtains most emphatically in the supply and demand of capital as essentially the common capital of a class - something industrial capital does only in the movement and competition of capital between the various individual spheres . . . Moreover, with the development of large-scale industry money capital, so far as it appears on the market, is not represented by some individual capitalist, not the owner of one or another fraction of the capital in the market, but assumes the nature of a concentrated, organised mass, which, quite different from actual production, is subject to the control of bankers, i .e . the representatives of social capital . . ." (Capital, III, p .361) . This analysis raises a number of problems that cannot be treated in depth here . Firstly, Marx is elsewhere notoriously scathing about moneycapitalists', and especially bankers', understanding of the capitalist mode of production . We should not therefore interpret him as arguing that the



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

59

interests or the policies of bankers can be identified with the needs of capital-in-general . Rather we should interpret him as arguing that it is through the form of the fluidity of money capital that capital-in-general imposes itself most directly and immediately on individual capitals, and that it is in their relation to their bankers that individual capitals experience this . Secondly, the development of imperialism, as the era in which the differentiated functional forms of capital cease to have independent existence and are reintegrated in finance capital, has important implications for the analysis . On the one hand, it imposes barriers to the fluidity of money capital which have a whole series of important implications . On the other hand, it creates a new nexus of relationships to capital-in-general . In particular, productive capital ceases to have independent existence as a form of capital, and so particular capitals for which "the capitalist character of production is a necessity" cease to exist as significant social forces . The implication of the 'separation of ownership from control' is that specific interests that derive from the particular form of productive capital cease to be represented by differentiated independent capitals, and come to be represented by functionaries of capital . This is important when we consider the representation of particular capital interests because the interests of functionaries cannot simply be identified with those of capital in the appropriate form, as many Marxists attempt to do, because those functionaries are 'immobilised' in a way that capital is not, and so may well have an interest in the productive enterprise to which they are attached that extends beyond the commitment of capital to it as a source of surplus value . (15) . The conclusion of our examination of the "particular forms" of capital is that the fractions that develop on the basis of those forms relate differently to the reproduction of social capital . Hence no particular capital interest can be identified with capital-in-general, nor would we expect any particular capital to have an adequate understanding of its own interest as a part of the social capital since it will have only a partial view of a circuit in which its own fate depends on that of others . Finally, we can conclude that capital-in-general presents itself not only in the differentiated forms just discussed, but also makes itself felt in the fluidity of money capital that is the form in which, in the developed capitalist mode of production, competition makes itself effective for the individual capital . Before developing the implications of these observations for the analysis of the relations between capital-in-general and the state we must consider another kind of differentiation that constitutes particular capitals as 'fractions' of capital . This is the differentiation of individual capitals from one another in accordance with the branch or department of production in which they are active in their productive phase . The first, and by now familiar, point to note is that we cannot define the particularity of individual capitals by identifying them with the characteristics of the means of production or product or with the circumstances of their particular labour process, as fractionalism attempts to do . Capital cannot be identified with such physical characteristics because it is not immobile, tied to a particular material form, but is mobile :



60

CAPITAL AND CLASS "It is a necessary prerequisite of the aggregate process of production, especially for the social capital, that it is at the same time a process of reproduction and hence a circuit of each one of its elements . . . Therefore it can be understood only as motion, not as a thing at rest . Those who regard the gaining by value of independent existence as a mere abstraction forget that the movement of industrial capital is this abstraction in actu. "(Capital, 1 1, pp .104-5)

It is only in one phase of its circuit that a particular capital is confined to a given physical form . This is a point of fundamental importance, for it completely undermines the fractionalist account of the pattern of accumulation in South Africa . This is because one cannot in general identify state policies which favour the application of capital to particular branches or departments of production with policies which favour particular capitals . The identification is only possible if there are specific barriers that suppress the competition between capitals so as to prevent other capitals from entering that branch of production . Otherwise other capitals well seek to apply themselves to the favoured branches and the normal processes of competition will eliminate the specific advantages . Any gains made by particular capitals will then be fleeting, short-term gains . Thus, where there are not specific barriers to entry, we would in general expect attempts by particular capitals to secure an advantage by means of measures that favour a particular branch of production to be self-defeating . We would then expect it to be more likely that such measures were introduced by the state in order to achieve the effect they do in fact achieve : to attract more capital to the favoured branch in order to alter the material relationship between branches, perhaps because of the failure of the competitive process to achieve this . This is not to argue that measures to favour particular branches of production cannot favour capitals . I have already noted that restrictions to entry to the branch in question will permit this to happen . In the South African case there are barriers to entry to the mining sector both as a result of the physical limitations of the availability of gold and because, at least until recently, it has not been possible to amass sufficient quantities of capital to inaugurate mining operations within South Africa itself . Hence state policies that favoured mining at the expense of other branches would indeed also favour particular capitals, and specifically 'foreign' capitals, and the advantage would not be liquidated by competition . For this reason the temptation for mining houses to use the state to secure particular advantages was genuine, while the attempt to do so generated the political opposition to mining capital that has limited their access to state power for over fifty years . There are other circumstances in which favouring particular branches also favours particular capitals . I have already noted that some short term gains can be made before competition makes itself effective . More importantly, a permanent advantage will accrue to capitals with a certain portion of capital immobilised in a particular branch if conditions are such that their rate of profit falls below the average . This is very likely, for example, in a recession which has adifferential effect on different branches of production .



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

61

This was the case, for example, in South Africa in the 1920s, when the recession hit both industry and agriculture particularly hard, threatening the existence of many capitals committed to those branches, and the jobs of many white workers employed in them . Thus an alliance was formed between the displaced white workers and some capitals engaged in agriculture and industry to secure government support . The election of the Pact government in 1924, far from expressing the strength of the 'national bourgeoisie', actually expressed their economic weakness . Their political victory led to measures that favoured accumulation in industry and agriculture to a limited extent in the '20s, but on a larger scale in the '30s . The effect of these measures, however, was to attract other capitals (including an increasing amount of 'foreign' capital) to these branches . Since the 'foreign' capitals had considerable competitive advantages over purely 'national' capitals the latter soon found themselves once more in difficulty, beginning to demand measures, particularly in relation to the provision of financial resources, that would enable them to survive against the competition . The election of the National government in 1948 expressed, amongst other things, these demands . But once again the achievement of these particular 'national' capitals was a sign of weakness and not of strength . Halving discussed the conditions that serve to define fractions of capital, we can now discuss the conditions that are most likely to lead particular capitals to organise themselves politically, and this can lead us back to the question of capital and the state . In the first place it goes without saying that particular capitals have an interest in common with all other capitals in preserving the rule of capital and in ensuring that the state serves to express the domination of capital over labour accordingly . Different capitals will have slightly different emphases in relation to this common interest : productive capitals, for example, will be most concerned with specific aspects of the subordination of the working class while other capitals will perhaps be more concerned with the generalised defence of property . We would expect particular capitals to take their "social responsibilities" in this respect most seriously to the extent that the rule of capital is most seriously threatened . In the second place we can surmise that, insofar as the particular forms of capital in the circuit of capital have independent existence as particular capitals, the interests represented by these different forms will differ . Moreover there are possibilities for state intervention in the relations between these forms of capital that permanently favour one against another . For example, restrictive legislation surrounding the disposal of land makes it possible for the landed class to extract an absolute rent, and associated with measures that favour agriculture can also increase the differential rent that accrues to the landed class . The struggle against such privileges is an important feature of the early development of capitalism, the success of which marks the elimination of landed capital as a particular form . Likewise the state, through its monetary activity, can affect the division of surplus value between interest and the profit of enterprise, a division that is, however, largely reduced to insignificance with the development of finance capital, so far as the relation between capitals is concerned . Thus the fractioning of capitals into particular forms is politically important in the



62

CAPITAL AND CLASS

period of competitive capitalism, but with the development of the imperialist stage its significance is considerably reduced . In the third place we can surmise that capitals defined in terms of the branch of production to which they are committed will be most likely to organise in pursuit of their particular interest in two very different circumstances . On the one hand, if they are engaged in a branch of production that they are able to monopolise for one reason or another so that particular advantages gained are likely to be retained . This is likely to be the case in those extractive industries whose raw materials are particularly scarce : gold and to some extent platinum, uranium, diamonds etc (all of which are found in South Africa) . With the increasing development of monopoly in other branches of production we might expect an increase in the political representation of capitals engaged in such branches . However, we would not expect the blatant exploitation of monopolistic advantages through the state to be something that would be politically feasible on a long-term basis . On the other hand, we would expect that particular capitals who are economically weak would organise themselves politically in order to restore their position . Finally, we would expect particular capitals to organise themselves in various ways in accordance with their particular interests in relation to this or that policy . This kind of organisation cannot properly be called a 'fraction' of capital, but is rather a 'capitalist interest group' . Its basis will depend on the policies of the state in relation to which it is formed (cf. Fransman and Davies 1977, p .263 quoted above) . However, we would expect such a pressure group to have an importance to the extent to which it can express a constellation of interests and to the extent to which it can reasonably expect to secure favour for the interests represented . Both conditions favour the organisation around the issue of 'nationality' : the form of the nation state makes it relatively easy to establish nationality as a basis on which to discriminate against particular capitals, while the associated ideology makes it relatively easy to secure political support for such discrimination . At the same time 'nationality' can encompass a series of interests that can be served by a number of different policies . On the other hand, the organisation of capitals on such a basis is likely to be predominant only in one phase of capitalist development, that in which in the nation in question the process by which 'national capital' comes to be integrated into the circuits of world capital is in its early stages . 'Nationalist' policies with regard to particular capitals tend to improve the terms on which 'national' capitals are integrated into the circuit of world capital rather than to resist that integration . We can now see why South Africa provides such fertile ground for the hypotheses of the fractionalists, for in South Africa practically all the conditions that might be thought to favour the political organisation of particular capitals have prevailed for at least a part of this century . Thus South Africa has in fact seen an unusual degree of coincidence between particular capital interests and political organisations . The problem with the fractionalist account is that it is unable to explain adequately the emergence or the significance of such organisations and of the conflicts in which they engage .



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE 9.

63

CAPITAL-IN-GENERAL AND THE STATE

We must now return to the fundamental question of how capitalin-general imposes itself on the state . For the fractionalist approach the subordination of the state to capital is effected through the formation of the power bloc within which one capital fraction is hegemonic . Thus it is through the system of representation that capitalist interests are translated into the domination of the state by capital . There are two important problems that immediately arise . Firstly, there is the problem of explaining how the representation of particular capitals can effect the subordination of the state to capital-in-general . We have seen that no particular fractional capitalist interest can be identified with capital-in-general since different capitals relate in different ways to the process of accumulation as a whole . Although every capital is but a fractional part of the total social capital we could not expect the representative of that capital to be aware of the implications of this interdependence, nor to put the `needs' of capital-ingeneral above his own interests . Thus we find as often as not that the representatives of particular capitals have a very short-sighted and often quite inaccurate understanding even of their own interests as capitals . It might be argued that the subordination of the state to capital-ingeneral is ensured as a result of the interaction between the various capitalist interests represented . This argument would be based on an analogy between economic competition and the political process . Just as competition is a process through which capital-in-general imposes itself on particular capitals, so the competition between particular capitals in the political sphere ensures that the needs of capital-in-general emerge as a kind of resultant . This argument continues to rely on the fractionalist identification of capital-in-general with the state to ensure that the interests of the former will prevail, for it lacks any explanation of why it should prevail . Competition does not necessarily impose the needs of capital-in-general on particular capitals (which is why the concept of the `needs' of capital-in-general is itself a very ambiguous one) . Indeed it is to the extent that capital-in-general fails to impose itself on particular capitals that the state is called on to intervene . Thus this approach would want capital-in-general to prevail politically just when it is unable to prevail economically . Finally, this approach continues to assume that the subordination of the state to capital is effected through the system of representation . This assumption raises a second important problem . It is important to distinguish between the system of representation and the legislature . Although it is quite common for particular capitals to organise politically, their organisation normally takes the form of the pressure group that seeks direct access to the executive rather than the form of the political party (although particular capitals will often contribute financially to one or more political parties - not necessarily capitalist parties for capital is less prone than Marxism to the parliamentarist illusion) . There are two main reasons for this . Firstly, it is very rare for the representatives of particular capitals to be able effectively to put their own particular interests as the the general interest . Secondly, because for capitals to represent their



64

CAPITAL AND CLASS

interests effectively it is necessary for them to be associated, if at all, with the party in power and not to prejudice their position by being too closely identified with the party in opposition . Thus the parliamentary form is not in general a means of representation adequate for capital . This does not mean that capitalists do not finance parties, nor that some parties are associated more closely with some interests than with others, but to offer a reductionist analysis of the system of representation by identifying parties with capital interests is to neglect the complexity of the system, on the one hand, and to overemphasise the parliamentary form of representation of the other . To argue that the system of representation is the means by which capital subordinates the state to itself seems to raise insuperable difficulties. The difficulties arise because of the radical separation of the economic and the political, that expresses the separation of distribution from production . This leads to a view of the state as some kind of metaphysical entity whose relation to society can only be re-established in a reductionist way . However the state is simply a specific institution that has abrogated to itself certain powers and that is inserted in a particular society . As such the state is subordinate to the domination of capital that characterises that society, a domination that does not have to be produced anew at the level of the state since it already exists . Thus the state is subordinate to capital because it is inserted in social relations that are themselves defined by the dominance of capital . Thus political relations do not reproduce economic relations, rather they complement them as different forms of the same fundamental social relations . Thus, for example, competition between capitals, as the economic form, and the political regulation of the relations between capitals, as the political form, are clearly complementary forms through which the interdependence of capitals as parts of the total social capital can be regulated (cf . Holloway and Picciotto 1977, pp .77-9) . When we ask how capital imposes itself on the state, we are not asking what are the instruments chosen by capital-in-general in order to make itself effective, for capital-in-general exists only through the action and interaction of particular capitals, and correspondingly only secures its domination through those particular capitals . Thus the domination of capital-in-general over the state is not direct but is mediated through capital's domination of all social relations . When we ask how does capital impose itself on the state we are therefore asking how does capital impose its domination over society, on the one hand, and what are the specific institutions through which the state is integrated into that society, on the other . It is only in this mediated way that we can say that the effectivity of the state is subordinate to and limited by the capital relation . Moreover it is only by regarding the relation between capital and the state in this mediated way that we can understand how the contradictory foundation of the capital relation is expressed at the level of the state . The dominance of capital is itself established on a contradictory foundation, for in consolidating its dominance it is also preparing the conditions for its overthrow . The contradiction is expressed in mediated form within the state, which has both to foster the intensification of capitalist exploitation and to



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

65

maintain the political conditions for the continuation of this exploitation . It is only in the concrete continuation of the specification of this contradiction that the policies of the state are determined . The system of representation certainly plays a part in structuring the subordination of the state to capital . This indeed is not simply through the mobilisation of particular capitals, for the forms of representation themselves express the subordination of the state to capital as a form of the domination of capital over society . Historically the forms of popular representation have developed out of institutions which were forms of representation of property whose main powers concerned the defence of the rights of property, above all against encroachment by the state . Popular representation thus developed as the representation of another form of individual property, (16) labour power, and popular political parties developed on this basis, seeking to improve, among other things, the terms on which labour power was sold . Hence even the forms of popular representation that have developed as a part of the capitalist state have served to impose on the state, in a mediated and contradictory form, the dominance of capital ; and this is not because the state is itself a form of capital but because the domination of capital over labour already exists in the economic sphere, and is compounded by the ideological forms within which popular representation developed . At the same time the forms of popular representation express the domination of capital in a contradictory form : while they provide a mechanism that can create the political and ideological conditions for the execution of particular policies, they can only do so at the expense of providing access to the state apparatus, of a limited kind, for groups who may carry their resistance to the state and to the rule of capital into that apparatus . The system of representation provides one of the ways in which the state is integrated into society, and so in which capital imposes itself on the state . It is not, however, the only way, and various Marxist theorists have analysed other mechanisms that serve to express, in a mediated form, the domination of capital at the level of the state (e .g . Offe's work on the administrative structure of the state, of Hickel's on the "tax state") . To conclude this section, however, I want to discuss the mechanism by which capital-in-general makes itself felt at the level of the state that has been unduly neglected, despite the fact that it is the form in which the capital relation impinges most directly on the state . This is in the fetishised form of the money relation . In the last section I noted Marx's argument that it is in the form of money-capital that capital-in-general achieves real existence because it is the form in which their social character is manifested to particular capitals . Money-capital not only confronts particular capitals, however, but it also confronts the state . The relation between money-capital and the state is realised through the state's responsibility for the regulation of the monetary system, and is institutionalised in the relation between the Central Bank, that peculiar institution that is a part of the state, but is normally beyond the jurisdiction of legislature and the executive, and the state . This relation is not essentially that between a particular fraction of capital, 'bank', 'money' or 'finance' capital, and the state, but rather involves



66

CAPITAL AND CLASS

capital "in its elemental form" . The separation of the particular fractional interests from those associated with capital in its elemental form is a striking characteristic of the historical development of Central Banks . This relation between money-capital and the state is a relation in which capital-in-general only makes itself felt by the state in a mediated form . On the one hand, the money form is an intensely mediated form of appearance of the capital relation . On the other hand, it is only insofar as disturbances in the monetary system are further mediated politically that they serve to structure the activity of the state . The stability of the money form is, for various reasons that need to be more adequately theorised, an essential underpinning of the stability of the state form itself . It is most particularly in the form of a monetary crisis that an economic crisis makes itself felt at the level of the state . But this is also the most fetishised, and so least adequately analysed form of the crisis : the form in which the crisis appears most detached from the mode of production whose contradictions it nevertheless expresses . This is particularly important in the case of South Africa, for it is in the form of a monetary crisis, and specifically a crisis on the foreign exchanges, that the basic structural weakness of capital in South Africa is currently expressing itself economically . The task of a Marxist analysis of that crisis must be to demystify that money form and thereby to establish the connection between the economic and political expression of the crisis in order to reveal both as the forms of appearance of a crisis in the reproduction of the capital relation . 10 . CONCLUSION : CAPITAL AND THE CRISIS IN SOUTH AFRICA In the last two sections I have outlined a conception of the relation between capital in general, fractions of capital and the state quite different from that of the fractionalists . This discussion should only be seen as an outline, for the essence of the argument is that the connections between capital and the state cannot be seen as abstract connections of functional necessity but must be seen as concrete connections mediated by specific historically evolved institutions . Thus the connections I outlined in the last two sections are not determinate connections, they merely indicate the relationships within which particular historical determinations are worked out . When I refer to 'limits' imposed by capital-in-general these limits should not be seen as absolute : it is often only through their violation that their character as limits becomes clear . For this reason the real significance of the difference between the approach I have just outlined and that of the fractionalists can only really be brought out by contrasting historical accounts (cf . Innes and Plaut, 1977) . Within the framework of this paper I cannot offer an alternative account of the development of the South African state . However I can indicate the differences by contrasting the approaches to the current crisis that seem implicit in the two positions . The fractionalists do not offer us any developed analysis of the current crisis in South Africa . This is probably because for them there is no crisis, at least at the level of the state . The separation of economic and political means a separation of economic from political crisis . Despite its economic



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

67

difficulties the South African state is not in crisis, for the cohesion of the power bloc is not threatened, nor is the hegemony of the hegemonic fraction . Indeed the integration and interpenetration of capitals has largely suppressed fractional struggles, so it is difficult to see how, for the fractionalists, a crisis at the level of the state could be possible . Thus the implicit picture that we are offered is one of a state which is the direct representative of an omniscient and omnipotent class of monopoly capitalists, supported by a racist white working class, striving to cope with an economic crisis, the crisis being seen as the context in which the class struggle

takes place and not as one form of the class struggle itself . This indeed is very much the approach taken by Johnson in his recent book (Johnson 1977), a work that is much influenced by what he calls the "new school for South African historiography" . Since the economic crisis is divorced from the class struggle, resolution of the crisis is regarded as being unproblematic, a matter of economic policy, without significant repercussions at the level of the state . Since there is an unmediated conception of the relation between capital and the state, the implementation of the necessary policy is also regarded as being unproblematic . The economic crisis of South African capitalism, the increasingly militant struggle against capital in South Africa, cannot themselves generate a political crisis, for that depends on the rupturing of the internal cohesion of the capitalist class . The only alternative offered to the liberal faith in the progressive character of enlightened capital is a complete absence of hope . An alternative conception of the relation between the economic and the political, however, can lead us to an altogether different understanding of the crisis and of the revolutionary prospects in South Africa . The current economic crisis in South Africa is a manifestation of the failure to solve one of the perennial problems of South African capitalism, that of the relations between the different departments of social production . We have already seen that Williams posed this problem correctly in arguing that we have to look at the relations between departments not primarily in terms of the political struggle between fractions of capital but in terms of the reproduction of total social capital . However William's own analysis was deficient in two respects . Firstly, as we have seen, because of his erroneous understanding of gold mining . Secondly, because he only considered the reproduction of total social capital within South Africa itself. This led him to ignore the possibility of South African capital exchanging products on the world market in order to establish the material basis for the expanded reproduction of capital within South Africa . This is a fundamental omission, since South Africa has one of the most tradedependent economies in the capitalist world . The imbalance between departments in South Africa is not expressed directly, but in the form of an endemic balance of payments problem . The pattern of accumulation in South Africa has been such that, of the three major sectors of the economy, while agriculture and mining have both had a consistently low propensity to import and high propensity to export, manufacturing has had a high propensity to import and a low propensity to export . This in turn reflects a number of factors, but in particular the fact that department I, the production of means of production, and



68

CAPITAL AND CLASS

the production of many intermediate products has been and remains undeveloped in South Africa . The low organic composition in mining and agriculture, and the low domestic demand for the products of agriculture consequent on depressed black living standards are also factors that should not be neglected in considering the present crisis . The result is that, contra the fractionalists, the growth of manufacturing has not been achieved at the expense of mining and agriculture, but has depended crucially on the rapid growth of the latter, the relation between the sectors being regulated through the balance of payments, either through direct market adjustments or through state intervention . The balance of payments crisis, which for the state takes the form of a monetary crisis, is thus the form in which a crisis in the relations between departments manifests itself at the level of the state . It is thus through the state's responsibility for the monetary system that the state is constrained to regulate the relations between departments within the reproduction of total social capital . While mining and agriculture set the limits on the rate of accumulation that can be sustained in South Africa, manufacturing is the leading sector in the sense that, for a number of reasons it is necessary to maximise the rate of growth of manufacturing . This is not because "the role of the Apartheid state was to create the conditions for monopoly capital to assume hegemony some time in the 60s (albeit a hegemony exercised through a governing class drawn from the political representatives of capitalist agriculture and its allies .)" (Davies 1977, p .292) Nor is it because the state is identified with monopoly capital in some still more direct way . Rather there has been a generalised political commitment to the expansion of manufacturing industry not only on the part of capital, but also on the part of the petty bourgeoisie and white working class whose subordination to capitalist relations of production is expressed politically in their identification of their own interests with those of accumulation in South Africa, on which the stability of employment and enhanced standards of living depends . This identification is all the closer to the extent that these groups are especially privileged and to the extent that their privilege appears to depend on the subordination of the black working class . There have been two fundamental reasons for concentrating on the growth of manufacturing . Firstly, it was necessary to develop a diversified economic base in order to limit the vulnerability of the South African economy to fluctuations in world markets, in the short term, and to lessen dependence on the extractive sector, which has always been thought to have a short life, in the long term . Secondly, and most important since the late 1950s, the rapid growth of manufacturing has been clearly seen to be a fundamental condition for the maintenance of control of the working class . Up to the late 1940s, the growth of manufacturing was seen as providing jobs for whites displaced from the land and providing more productive employment for whites in protected jobs in the public sector . More recently the rapid growth of manufacturing employment has been closely



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

69

connected with the social control of the increasingly proletarianised black population, both because of the discipline imposed on the worker by employment, and because the repressive system of labour control tended to encounter less serious resistance from the working class when employment was rising than when it was stagnant . This is the fundamental context of accumulation within which the system of labour control that is at the core of Apartheid has developed since 1948 : not as a monolithic system of repression but as a diversified system of control designed to allocate labour to the various sectors on terms that were consistent with the level of development of the productive forces, which was very uneven between sectors, and as a system that could clearly differentiate between those workers on whom the discipline of labour would be imposed and those who would be more directly and immediately subject to state control . During the 1960s a sustained boom, fuelled by an unprecendented inflow of foreign capital and buoyant primary exports, made it appear that South Africa had broken through the bottleneck, as foreign investors brought in new intermediate industries and the most advanced techniques . The make or break character of this boom is best indicated by the readiness of the state to resort to inflationary financing, something it had refused to do in the 1950s because of the impact on the export sectors : inflationary solutions to the problem of financing rapid accumulation effectively transfer surplus value from mining and agriculture, which face rising costs and stagnant export prices, to manufacturing, sustaining the growth of the latter while impeding that of the former . Thus the growth of manufacturing in the 1960s was tending to knock away the props on which previous manufacturing growth had been based . Thus, in volume terms, gold exports have been more or less stagnanat since 1965, while the rapid volume growth in food and livestock exports from 1965-8 has been sharply reversed subsequently . This meant that the sustained boom of the late 1960s depended on the ability of manufacturing to develop from a substantial liability in balance of payments terms into a considerable exporter . In fact this did not happen, and there are few signs that it will happen . In volume terms manufacturing exports have been stagnant since 1968, although this conceals considerable variation between sectors . If anything South African manufacturing, far from breaking through into export markets, is becoming more vulnerable to foreign competition on the home market . Even where South Africa might be thought to have a considerable comparative advantage it is exposed : the Iron and Steel Corporation has had to resort to large scale dumping . The balance of payments crisis that has developed over the past three years has been affected by a number of factors other than the imbalance between sectors . Similar factors had served to postpone the crisis over the previous three years . If South Africa is not to rely on a continuing run of unlikely accidents to sustain accumulation it is necessary to solve the basic structural problem of the South African economy : the problem of the uncompetitive nature of South African manufacturing industry on world markets .(17) By the early 1960s it had become apparent that South Africa could not compete on world markets on the basis of her cheap



70

CAPITAL AND CLASS

labour . That is not to say that South African labour is not cheap, but simply that labour elsewhere was cheaper and could be worked more intensively . Hence it was believed that the answer was a programme of capital intensification which would allow South Africa to compete on the basis of advanced technology and a relatively sophisticated labour force . However the weakness of South African manufacturing was never a technological weakness but always has been a weakness in the social relations of production . Thus South Africa has been technologically backward because capital has not installed modern technolgoy . This is certainly in part because labour in the past was cheap, industry was protected, and the labour force was not highly skilled . However modern technology also requires a 'disciplined', 'responsible' labour force, and this South Africa did not have. Following Sharpeville, the savage repression and the rationalisation of the system of labour regulation made it appear as though the conditions under which this discipline could be imposed and modern technology could be profitably employed had been created . However the very process of the introduction of modern technology created the conditions under which the 'indiscipline' of the working class would begin to manifest itself in an organised form, for the introduction of the new technology meant changes in the imposition of the rule of capital at the point of production, on the one hand, and the creation of large scale redundancy, on the other . During the boom of the 1960s economic expansion made it possible to keep the lid on the situation as employment opportunities expanded and some limited concessions could be made as the condition for the introduction of new work practices . The steadily worsening recession of the 1970s, greatly exaggerated by the measures taken to correct the balance of payments deficit in late 1976, has progressively exacerbated the conflict both at the point of production between labour and capital directly, and in the urban ghetto, between the state and the working class . It should be clear that the current political crisis in South Africa cannot be separated from the crisis of accumulation, but rather that it is the political form of that crisis . The contradiction that besets the South African state is that it has to resolve the economic crisis without intensifying the political crisis to such an extent that the very persistence of capitalism in South Africa is threatened . The economic crisis is the expression of the problems confronted by the production of surplus value, above all in the manufacturing sector . In order to resolve this crisis it is necessary to restructure social relations in the immediate process of manufacturing production in such a way as to increase the rate of exploitation . This restructuring involves primarily the intensification of labour and the tightening of work discipline in the manufacturing sector . However, this restructuring is only possible at the cost of an intensification of class struggle that threatens to take an increasingly political form . This intensification of class struggle will manifest itself in two forms . Firstly, the struggle at the point of production between capital and labour can be expected to become increasingly acute . Moreover, to the extent that the restructuring of the relations of immediate production involves the erosion of the position of sections of the white working



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

71

class, we would expect a tendency to increasing militancy on the part of the white trade unions . Over the past three years the real wages of white workers have been eroded even more than those of black workers . Whether this increasing militancy on the part of the white workers also takes a political form, and what political form that militancy takes, is a very important question that has not been confronted . It is certainly to be expected that capital and the state will attempt to turn this militancy against the black working class . However it is not certain that such an attempt will be as successful as it has been in the past, for the scope for concession to the white workers is limited . The second form in which the class struggle is to be expected to intensify is in the struggle between the state and the reserve army, especially that section concentrated in the urban ghettoes, a reserve army that is likely to be increased more rapidly by the growing relative surplus

population than it is reduced by the absorption of labour by continued accumulation . It is this moment of the class struggle that has been most dramatically manifested over the last year, both in the struggles in the townships and the attempts by the state to intensify influx control . The economic crisis in South Africa appears most acutely in the form of a monetary crisis, a crisis of South Africa's international financial position . This form of appearance is not an illusory form, for it is the acute monetary crisis that compels the state to intervene ever more actively to resolve the underlying crisis . The fundamental economic crisis is not, however, a monetary crisis but is a crisis of surplus value production, a crisis in the social relations of production . In their attempts to resolve this crisis, capital and the state in South Africa have to take the initiative in intensifying the class struggle, tightening labour control at the point of production and increasing the reserve army by the creation of relative surplus population . At this level there is no possibility of compromise, no basis on which concessions can be made to defuse the growing political crisis . The specifically political problem the South African state confronts domestically is that of containing the political struggles generated by the intensification of class conflict in the economic sphere . On the one hand increasingly open and pervasive repression is being used to prevent the class struggle from taking a political form . On the other hand, to the extent that the development of political struggles cannot be prevented, the state is seeking to deflect and divide the political initiatives that emerge . It is here that the South Africa state is attempting to make limited "concessions" to the black working class, creating sterile forms of black political representation in the Bantustans and in the urban ghettoes, and attempting to make selective concessions, especially to those with permanent urban residence, in order to intensify the divisions in the working class that the recent struggles have begun to overcome . The crucial political question that confronts the South African revolution at the moment is whether the black working class can build on its new found unity and forge a cohesive revolutionary movement in the face of these not unsubtle attempts to divide and deflect it . A subsidiary question is that of whether that movement can in turn open up and intensify divisions both between white

72

CAPITAL AND CLASS

workers and the state, and between the different capitalist interests at work in South Africa. In this concluding section I have given a brief account of the basis of the current crisis in South Africa that should be sufficient to indicate the distance that separates the fractionalist approach from that which I have attempted to delineate earlier in the paper . The most important difference is that the account I have offered does not rigidly separate the economic from the political, and distribution from production, but rather seeks to lay bare the intimate connections between the economic and political form of the crisis . Revealing these connections also makes it clear that the state cannot deal separately with the economic and political crises that it confronts, for these are two forms of the same crisis . Finally, by relating the two forms of the crisis this analysis enables us to grasp the contradictory foundation on which the actions of the South African state is based . It is only by grasping in a very concrete way the forms in which the fundamental contradiction of capitalism in South Africa is expressed that we can begin to analyse the ways in which the revolutionary movement can exploit the current crisis to open up the contradiction and to drive the struggle on to the point at which the proletariat can take its place as the ruling class .

FOOTNOTES 1

2

Earlier versions of this paper have been discussed in the CSE Neocolonial State Group, at the CSE conference and in the Coventry CSE Group . I have benefitted from all these discussions . I would particularly like to thank Martin Legassick, Duncan Innes, Sol Picciotto, Bob Fine, Simon Frith, Veronica Beechey for their help and encouragement . Williams offers two arguments about the specificity of gold that supposedly explain the peculiarity of South African capitalism . On the one hand, he argues that because the production of gold cannot contribute to the production of relative surplus value there is no incentive for gold capitals to increase the organic composition . They will rather try to depress wages to increase their surplus value . This arguent is false because there is an incentive for the particular capitalist to reduce his cost of production, and so increase 'his' surplus value, irrespective of whether this contributes to the creation of relative surplus value . Thus the tendency to replace living by dead labour applies just as much to luxury production as to other branches, although its implications are slightly different . On the other hand, Williams argues that gold mining does not enter into the equalisation of the rate of profit because gold does not have a price and so cannot be overproduced . From this he concludes that gold producers will attempt to pack as much living labour into their product as possible . The absurdity of the conclusion (which would imply that capitalist gold mining



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

73

should be carried out by unassisted labour) should have alerted Williams to the fallaciousness of the premise . While it is true that, as the money material, gold does not have a price (an exchange value expressed in gold), it does nevertheless have a 'price' in the sense that the setting of a 'price' of gold in terms of particular currencies sets up an exchange relation between gold and all other commodities taken together . If this 'price' is not in principle subject to variation in accordance with the production of gold, then it will bear no necessary relation to the value of gold, and gold will to that extent be disqualified from functioning effectively as the money material . Thus, far from being immune from the normal laws of capitalist production, gold production's conformity to those laws is essential if it is to perform satisfactorily its function as the money material . (Cf. on these points the slightly different argument of Bradby 1976) . 3 The most self-consciously Poulantzian works are those of Kaplan and Morris, and the more recent work of Fransman and Davies . However the approach is implicit in many other recent treatments of South Africa to a greater or lesser extent . These examples are drawn from papers listed in the bibliography . 4 Because they derive from an ongoing debate and because many are unpublished I have omitted specific attributions . It must be stressed that a proliferation of views in an ongoing debate is to be expected . But one would also expect such a proliferation to lead to a more fundamental examination of the issues . In case no such identification is possible the analyst need not despair . 5 If the 'representative' of a fraction does not represent the fraction's interests this is construed as "indicating the relative autonomy of the economic and political class interests of the capitalist fraction" (Davies 1977, p .152) 6 The only general reference to the foundation of a fraction that I have found is that of Kaplan et al . 1977, p .5 which refers to "differing role in the expanded reproduction of capital" and "interests corresponding to their particular place in the relations of exploitation", although these observations are not developed . 7 I have argued at some length elsewhere that this is in fact the basis of Poulantzas's theory of the state (Clarke 1917) For further discussion of the relation between capital-in-general and 8 particular capitals see Rosdolsky 1974, 1977 and Banaji 1976 . It incidently also leads the fractionalists quite consistently to refer to 9 the "contradiction within the dominant classes" as a "debate" based on "charge" and "counter charge" and on varying "interpretations" of policy (Morris 1975, appendix ; 1976, pp .173-5) 10 Thus the black working class makes only fleeting and entirely peripheral appearances in the recent article entitled : "Class struggle and the periodisation of the State in South Africa" (Kaplan et al . 1977) . If the South African state is set in the context of the class struggle between capital and labour, rather than that of the secondary struggles between different capitalist interests, we can get an altogether different, and much more satisfactory, view of South African history .



74

CAPITAL AND CLASS

Innes and Plaut, in a very important paper, show that the measures taken by the 1924 Pact government, far from expressing the alliance of national capital with the white working class, continued and even intensified the systematic policy of dividing the working class in order to intensify its exploitation . They also argue that this restructuring of social relations in the 1920s and 1930s, with its consequent intensification of exploitation, was the condition for the explosive growth of manufacturing from the 1930s . Thus the pattern of accumulation in South Africa is not explained, as it is by liberals and fractionalists alike, in terms of the struggles over industrial protection that concerned the distribution of surplus value among capital fractions, but rather is explained in terms of the production of surplus value and of the class struggle between capital and labour (Innes and Plaut, 1977) 11 Cf. Kaplan et al 1977, pp .29-30 ; Johnson 1977, conclusion . 12 Fransman and Davies 1977, passim, make some telling criticisms . Innes and Plaut 1977, is a devastating critique . 13 Davies 1977, consistently recognises that policies of segregation and Apartheid actually had the support of all capitals and yet retains alongside that recognition a consistently fractionalist explanation . 14 Yudelman, in a review of Johnson's seminal book, has noted the recurrence of liberal themes : "On the historical level one is struck how frequently Johnson's view of the mining indsutry, workers and the state is exactly the same as those of the liberal historians he claims to be supplanting, although his terminology is different . He even repeats liberal errors where a Marxist framework would have justifiably led him elsewhere, e .g . he espouses the hoary of cliche that the mining industry won the battle of 1922 and lost the war of 1924, when the Pact government was elected . In this he has been followed by other neo-Marxists ." (Yudelman 1977,p .205) . Morris, in a very acute critique of liberal theories, identifies precisely the basis of the complementarity between economism and overpoliticisation in the radical separation of economic and political, without realising that the same is true of his own fractionalist work : "What we have here is the arbitrary introduction of a political explanation for an economic event - industrialisation - simply because it happens to suit the argument . In other words, the initial absolute separation of the 'economic' from the 'political' paradoxically allows the latter to be reintroduced into the analysis at any point in time and to any degree of effectiveness . Since the economy is abstracted from its social and political base, and since the two are not theoretically integrated, the whole system can be stood on its head when ever it is so desired ." (Morris 1976b, p .285) 15 Lebowitz 1973, offers a provocative analysis of the debate between neo-Ricardian and neo-classical economics in terms of the radicalisation that takes social democratic forms and that is expressed in the counterposition of the rationality of the relations of capitalist production to the irrationality of bouregois distribution relations, and so seeks to overthrow the latter while preserving the former .



CAPITAL, FRACTIONS OF CAPITAL AND THESTATE

75

16 The participation of people in the state as citizens, which for Poulantzians tends to be the defining feature of the capitalist state, is in fact only a special case of their participation as property owners. 17 Thus the Reynders Commission, set up in 1971 to investigate South Africa's export performance aruged that "an improved export performance by gold over the short to medium term may well obscure the necessity of making timeous (sic) steps to effect such structural changes" (p .157) . The Commission was under no illusions about the basis of the lack of competitiveness of South African industry : "the Commission . . . wishes to emphasise that efforts to increase the growth potential of the country should not, as would appear to have been the case in the past, focus on a greater supply of physical factors of production (labour and capital) only, to the almost complete exclusion of the improved utilisation of such factors ." (p .37) It also clearly understood the importance of an improved export potential : "exports are important for various political reasons . With regard to the domestic situation, exports will create new opportunities for production and employment and contribute to the economic stability and strength of the South African economy which, in turn, is essenpolitical stability and strength . As far as the tial for domestic external situation is concerned, exports affect this country's ability to play its proper role in the development of neighbouring territories and to make a positive contribution to the country's outwards policies ." (p .19)

BIBLIOGRAPHY Adler, T . ed (1977) : Perspectives on South Africa. African Studies Institute . University of Witwatersrand . Banaji, J . (1976) : Marx, Ricardo and the Theory of the Value Form,

Marx1stisk Antropologi, 2, 2-3 . Bienefeld, M . and Innes, D . (1977) Capital Accumulation and South Africa, Rape, 7 . Bradby, B . (1976) Note on Williams, 1975, CSEB Clarke, S . (1977) Marxism, Sociology and Poulantzas's Theory of the State, Capital and Class, 2 . Davies, R. (1977) Capital, The State and White Wage-Earners : An Hist-

orical-Materialist Analysis of Class Formation South Africa 1900-60. Ph .D Thesis, University Fransman, M . (n .d) The Role of the Non-Capitalist 11 (1975) The Political Economy of South

and Class Relations in of Sussex . August 1977 Mode . Mimeo . Africa . Mimeo .



76

CAPITAL AND CLASS

Fransman, M . (1976a) The Roots of South African 'Expansionism'. M imeo . ,1 (1976b) Theoretical Questions in the Understanding of South Africa, Marxistisk Antropologi, 2, 2-3 . Fransman, M . and Davies, R . (1977) The South African Social Formation in the Early Capitalist Period, circa 1870-1939 . Some Views on the Question of Hegemony, in Adler, 1977 . Holloway, J . and Picciotto, S . (1977) Capital, Crisis and the State, Capital and Class, 2 . Innes, D . (1976) The Mining Industry in the Context of South African Economic Development, 1910-1940 . Mimeo . Innes, D . and Plaut, M . (1977) Class Struggle and Economic Development in South Africa : The inter-war Years . Institute of Commonwealth Studies Paper . Johnson, R . (1977) How Long will South Africa Survive? MacMillan, London . Johnson, F . (1977) Comments on Kantor and Kenny, 1976 . Mimeo . Kantor, B . and Kenny, H . (1976) The Poverty of Neo-Marxism, journal of Southern African Studies . Kaplan, D . (1975) The State and Economic Development in South Africa, Mimeo . (1976a) The Politics of Industrial Protection in South Africa, 1910-1939, journal of Southern African Studies . (1 976b) An Analysis of the South African State in the 'Fusion' Period, 1932-9 . Mimeo . "

(1976c) Comments on Innes, 1976 . Mimeo . (1977) Capitalist Development in South Africa; Class Conflict and the State, in Adler, 1977 . Kaplan, D . and Fransman, M . (1975) Some Thoughts on the Value of Labour Power etc . Mimeo . Kaplan, D . and Morris, M . (1976) Labour Policy in a State Corporation : A case study of the South African Iron and Steel Corporation, South African Labour Bulletin, Jan . and April . Kaplan, D . ; Morris, M . ; Davies, R . and O'Meara, D . ("et al .") (1977) Class Struggle and the Periodisation of the State in South Africa, Rape, 7. Lebowitz, M . (1973) The Current Crisis of Economic Theory, Science and Society, 37 . Legassick, M . (1974a) South Africa : Capital Accumulation and Violence, Economy and Society. I (1974b) Legislation, Ideology and Economy in post-1948 South Africa, Journal of Southern African Studies . Lewis, D . (n .d .) African Trade Unions and the South African State 1947-53 Mimeo . Marx, K . (1957) Capital, Volume II, FLPH Moscow . Marx, K . (1962) Capital, Volume III, FLPH Moscow Marx, K . (1973) Grundrisse, Penguin, Harmondsworth . Marx, K . (1976) Capital, Volume I, Penguin, Harmondsworth . Moorson, R, (n .d) . The Foundations of Modern South African White Politics, 1947-53, Mimeo .



CAPITAL, FRACTIONS OF CAPITAL AND THE STATE

77

Morris, M (1975) Periodisation, Class Struggle and the State in South Africa, Mimeo . (1976a) The Development of Capitalism in South African Agriculture, Marxistisk Antropologi, 2, 2-3 . (also in Economy and

Society) "

(1976b) The Development of Capitalism in South Africa,

journal of Development Studies . Poulantzas, N . (1973) Political Power and Social Classes, Sheed and Ward and NLB, London . 11 (1976) The Crisis of the Dictatorships, NLB, London .

Reynders Commission of Inquiry into the Export Trade of the Republic of South Africa (1972-3) . Rosdolsky, R . (1974) Comments on the Method of Marx's Capital, New German Critique, 3, 62-72 . 11 (1977) The Making of Marx's Capital, Pluto, London . Simpson, H . (1975) Structural Change and the Labour Crisis in South Africa in the 1940s . Mimeo . 11 (1976) Review of the 'New School' Mimeo . Williams, M . (1975) An analysis of South African Capitalism : Neo-Ricardianism or Marxism? CSEB, IV 1 . Wolpe, H . (1972) Capitalism and Cheap Labour Power, Economy and

Society . (1975) Draft Notes, Mimeo . Wolpe, H . and Legassick, M . (1977) Bantustans and Capital Accumulation in South Africa, Rape, 7. Yudelman, D . (1977) The Quest for a neo-Marxist approach to contemporary South Africa, South African journal of Economics, 45, 2 .

History Workshop a journal of socialist historians Highlights of issue 5 (May 1978) : Anna Davin : Imperialism and Motherhood Mike Newman : Labour's Struggle against British Fascism, 1933-6 Tony Wailey : The Seamen's Strike, Liverpool 1')66 Hannah Mitchell : Art and the French Revolution William Rosenberg : Workers' Control in the Russian Revolution Martha Maclntyre : Women's History in Australia Nicholas Jacobs : 1 he German SPD School in Berlin, 1906-14 plus Workers' Theatre Movement, Local History, Archives and Sources, Libraries, Music . . . Subscription C5 a year (2 issues), S12 overseas from HISTORY WORKSHOP, P 0 Box 69, Oxford OX2 7XA

SURVEY THE SOVIET ECONOMY I N THE 1920s AND 1930s Mark Harrison

INTRODUCTION In the history of Soviet studies in Britain, Marxist ideas and influences have always been an undercurrent . This undercurrent has sometimes acted as a positive force ; sometimes it has been merely argumentative and factitious . But it has never escaped its subordinacy . Individual Marxists have worked in isolation, reliant upon the research, the formulations and the good will of non-Marxist colleagues . For the most part the influence of Marxism has been deflected, absorbed or contained . Thus, in examining the British field today, among the most important questions are the health and strength of bourgeois scholarship, and the prospects for developing a creative Marxist tendency capable of transcending its past 'moral and intellectual' encirclement . Here one clarification is necessary . I have referred to the position in Britain, but the links between English and American scholarship are many and obvious . Is there, in fact, a specifically 'British' Soviet studies? The internationalisation of Soviet studies has proceeded a long way, but it is far from complete . While language barriers, which separate British from Italian, French or German work, do not inhibit the emergence of an 'Anglo-American' entreprise, the very different political conjunctures continue to distinguish the balance of tendencies in Britain and North America . And I must make it clear that for me Britain is 'here' and North America is 'over there'. I can only write about this relationship from my own experience and knowledge, and inevitably my perception of the influences emanating from 'over there' will be incomplete . Finally, I shall seek to examine a number of positive developments in the study of the inter-war Soviet economy and its socialist transformation, and I shall cover the decade which began in 1966 . This decade has a special significance . It began with the publication of the last, sixth edition



SURVEY: THE SOVIET ECONOMY

79

of Maurice Dobb's Soviet Economic Development since 1917, and finished with his death . If, as I shall argue, the last decade has been one of transition, Dobb was a central figure in that process . THE SOCIALIST TRANSFORMATION OF THE ECONOMY

1

The question of primary accumulation'

For many years it was considered that the October revolution, in destroying the economy of the great landlords and kulaks in 1917-1918, had destroyed the ability of agriculture to produce a surplus . The redistribution of land to the rural poor, and the consequent re-emergence of a massive 'middle' peasantry, meant that the rural population consumed more of its own production, and sold less to the towns . The authority for this view was no less than Stalin . (1) In the conditions of the 1 920s, Soviet industry could not grow beyond the limits of its pre-revolutionary development without re-establishing an enlarged flow of foodstuffs (especially grain) and industrial crops from agriculture to feed expanding industrial production and employment . According to the conventional view, this enlarged flow, or 'tribute', was re-established through the collectivisation of agriculture in 1929-1930, and the enforced delivery of collective-farm 'surpluses' to the state at very low prices . (2) In this way, it was held, Stalin had carried out the strategy of primary socialist accumulation advocated by the Trotskyist economist Preobrazhensky (although by means of a coercive force not envisaged by Preobrazhensky himself) . (3) . Thus the primary accumulation of the socialist sector in the 1930s had been at the expense of the peasantry . The crude - but effective method of forced collectivisation had financed Soviet industrial growth . Recent work has largely discredited this system of views . In terms of agricultural production it is now accepted that the 1920s had seen a considerable improvement in crop yields, and in diversification away from a grain monoculture . Among Soviet historians both Moshkov and Barsov have argued that the revolutionary land distribution, and the revival of the middle peasantry, may have destroyed the semi-feudal estates and the more developed small-capitalist farms - but they also made possible a more productive peasant agriculture . (Moshkov 1966, pp .19-24 ; Barsov 1969, pp .22-3) (4) . This view is reflected in the most recent English work (Cooper, Davies and Wheatcroft, n .d .) . In terms of agriculture marketings to the state and the urban-industrial sector, there has been some controversy . Did the peasant agriculture of the 1920s really market much less to the towns than the pre-revolutionary agrarian structure? Or was this a Stalinist falsification? Would it have been so difficult to raise the level of marketings through a system of price incentives of the type advocated by Bukharin (Karcz 1967 ; Davies 1970 ; Karcz 1970)? From today's standpoint it is possible to suggest a number of tentative conclusions . Firstly, the Soviet peasantry before collectivisation probably was marketing less grain (Carr and Davies 1974) . But it was marketing more of other agricultural products, because it was producing more . The decline in



80

CAPITAL AND CLASS

grain marketing, although serious, was probably not caused by the diminished role of kulak and large-scale farming ; for example, less favourable price incentives may have had something to do with it (5) . However the real bombshell was the publication in the Soviet Union of a work by Barsov which examined the question of the agricultural surplus on the basis of previously neglected statistical and archival materials . If we define this surplus, roughly, as agriculture's sales to industry minus agriculture's purchases from industry, and measure this sum in real or constant-price terms, then the agricultural surplus did not increase as a result of collectivisation . Over the period of the first Five Year Plan (1928/29-1932) the agricultural surplus was no greater than in 1928, nor did it apparently increase subsequently . (6) There has been controversy over the correct value measurement of the surplus . But nearly everyone now accepts that Barsov has proved that collectivisation did not enable the extraction of an increased surplus from agriculture, and that agriculture did not finance the increase in industrial accumulation in the 1930s . (7) The reason for this was that the livestock sector declined dramatically after collectivisation (some animals died for lack of fodder, others were slaughtered in the collectivisation drive itself) . This necessitated massive industrial investment in agriculture to replace lost animal draught-power and fertiliser . In addition, however unfavourable to agriculture was trade at state prices, peasants were still able to sell food and buy manufactures at highly favourable (if often semi-legal) terms of trade in town bazaars . In Ellman's view, therefore, the undoubted sufferings experienced by the peasantry in the early years of collective farming did not reflect any contribution they were making to feed and fuel industrial development, but resulted from a series of'pure losses' (in both human and animal terms) . The costs of industrial growth were born by the industrial working class, in particular by the millions which it recruited from the countryside as industrial employment expanded (Ellman 1975) . Why was collectivisation so costly? In the 1920s the agrarian strategy of the Bolsheviks was to rely on the poor peasants, to win (or at least neutralise) the middle peasants, and to isolate and squeeze out the resurgent class of kulaks . In the upshot class lines became inextricably confused . There even arose the concept of the 'ideological' kulak : we are against the kulak, and he who is not a kulak but is not 'for us' is just as bad as a kulak . The campaign to liquidate the kulaks as a class became one of human liquidation, and collectivisation became an arbitrary, even terroristic process imposed on the whole of the peasantry . Some scholars have concluded that the Bolshevik agrarian strategy was misconceived because it misunderstood the class structure of the countryside . (8) They could not isolate the kulaks because the kulaks did not possess all the objective attributes of a social class, for example stability of membership and position . Some misgivings about the latter view have focussed upon the relationship between social mobility and the formation of a new class (Harrison 1977) . A separate area for further work is the formation of a Stalinist conception and practice in the field of class alliances and their politics - not only as if affected in the process of collectivisation, but also as it has affected and dogmatised our perception of it .



SURVEY: THE SOVIET ECONOMY

81

Perhaps we have also moved towards a new assessment of the strategy of 'primary socialist accumulation' . Firstly, the performance of peasant agriculture in the twenties was better than we had previously assumed . Secondly, the performance of collective agriculture was rather worse . The acceleration of socialist accumulation in the thirties was not financed by agriculture . This raises two fundamental questions . Firstly, given the need for extremely rapid industrial growth, could a small peasant agriculture have met the implied needs? Probably the dominant view would still answer this question in the negative, and is correct to do so (Cooper, Davies and Wheatcroft, n .d .) . Secondly, was there an alternative strategy for the whole economy, based on Bukharinist lines - a longer but less costly route to the goal of a developed socialist economy? This latter question is considered separately below . We are also led to another question . As Szamuely (of Hungary) and Millar have separately pointed out, the Marxian notion of primary accumulation concerns the transformation of relations of production, not just the accumulation by one mode of production of the surplus product of another . (9) The collectivisation of agriculture was a transformation of social relations, as indeed was the history of the whole economy between the wars . The question of alternative transformations cannot be answered without a political economy of these social relations . To this question, also, we return below .

//

Planning the industrial transformation

Despite many difficulties and setbacks, the USSR achieved in one generation a massive industrial transformation, on a scale which had required three or more generations for the advanced capitalist countries . This was achieved by a centralised, hierarchical planning system which piled resources into the heavy industrial sectors . This was the world's first planned economy (and it later became a model for more than one other socialist contry) . The conventional post-war wisdom produced a curiously dualistic assessment of this record . On one side were those who emphasised above all the chaos, arbitrariness and inefficiency of the process, often by comparison with a mythically functional equilibrium model of capitalist growth ; often to a point where virtually all Soviet statistical claims had to be disbelieved because they were inconsistent with the underlying analysisone example of this approach being that of Jasny 1961 . On the other side were those few such as Dobb who, partly out of a scrupulous and later vindicated defence of some (though not all) Soviet statistical claims, partly out of a fundamental position, emphasised the regularities of Soviet industrialisation and the unprecendented advances (Dobb 1966 ; on the question of Soviet studies see especially pp .261-8) . The fundamental problem which remained, however, was how to understand the gains in relation to the setbacks, and to understand the Soviet economy, in its world context, as a system which produced particular types and modes of advance, but which failed to produce others . In the early 1960s both Hunter (in the States, and Kaser developed a



82

CAPITAL AND CLASS

concept of planning priority . In the economy there are priority sectors (heavy industry) and non-priority sectors . The operational plan assigns higher growth rates to priority sectors . However the plan is also 'taut', that is, taken as a whole its goals exceed the productive possibilities of the economy . The reason for this is in order to stimulate the mobilisation of all possible resources, and to ensure that the economy grows at the maximum rate - that is, there is no 'slack' . But in carrying out the plan, difficulties will arise because the plan as a whole cannot be achieved : for example, there will be a shortage of steel . In this case available steel is allocated to the priority sector, which fulfils or overfulfils its targets, while non-priority sectors fail to fulfil their targets (Hunter 1961, Kaser 1962 . Kaser's article is a review of )asny 1961) . Thus plan underfulfilment does occur, but in a 'planned' way . As Hunter pointed out, it was also possible for excessively 'taut' plans to result in a complete breakdown of the economy (Hunter 1961) . To what extent could this explain the successes and failures of the early Five Year Plans? More recently Hunter returned to this topic with a fresh analysis of the first Five Year Plan (1928/9-32) (Hunter 1973) (10) . Employing the techniques of linear programming, he tried to establish whether, on the basis of the known economic situation in 1928, and the assumptions concerning technical and external conditions embodied in the plan adopted, it was possible to achieve the consumption and capitalstock targets envisaged for 1932 . He found that to achieve these targets, even with the unrealisitically favourable assumptions of the plan, would have required a 26% decline in personal consumption in 1929 over 1928 ; alternatively, with a 9% cut in consumption, the plan targets could have been met by 1936. Without a reduction in consumption the economy could not have grown at all . In spite of some significant criticism of Hunter's own assumptions (Davies and Wheatcroft 1975), there is little dispute over his broad conclusion, that even on the grossly optimistic assumptions of the planners, the first Five Year Plan targets were infeasible . Was this simply a case of a 'taut' plan, designed to mobilise resources and to incite the Soviet population to heroic efforts? Simultaneously with Hunter's article appeared a varied and interesting discussion (Campbell 1973 ; Cohen 1973 ; Lewin 1973 ; Hunter 1973) . Most of the contributors concerned themselves with the historical record of 1928-1932 (in disregard of Hunter's perfectly correct assertion that he had examined the first Five Year Plan, not the Plan period) . From this some important points were clarified . Firstly, the targets of the plan document were infeasible not because the planners were calmly organising a 'plan as propaganda' campaign to mobilise resources, but because some thought that absurdly high targets were really feasible, while others feared arrest if they did not agree . Secondly, such over-ambitious planning had several consequences . One was a tendency towards disintegration of the planning system itself. The planning offices and Ministries concerned with priority sectors hived themselves off from the others, who were left to sink or swim as best they could (Lewin 1973) . (11) This was far from any processes of rational and



SURVEY: THE SOVIET ECONOMY

83

consecutive 'planned underfulfilment' of the type previously described by Hunter . On the contrary it was a bitter struggle for survival in a political arena where the costs of failure were extremely severe . In addition, the material achievements of this planning process must

be seen in the proper light . By

the heavy industrial targets had by

1932

and large been met . But the plan was curtailed at the end of this year not as a reflection of its total success, but because the economy was in serious difficulty (12) . Planning is a social relationship which involves not only planners and politicians but also technical and managerial specialists, workers (both 'free' and 'forced') and peasants . There has been fresh work on the uneasy position of the technical intelligentsia, caught between the class power which had created it and the class which it had to organise in production (Lampert, n .d) . But little has been added to our understanding of the development of the working class or of, say, the Stakhanov movement of the middle thirties which one writer has described as a first 'economic and

technical' popular movement of the new working class (Spours

1976) (13),

and which in Dobb's view had contributed to both the success and disruption of the second Five Year Plan . A further area to receive fresh illumination is that of the labour camps within the Soviet economy . Writing in the middle sixties

Swianiewicz had interpreted the rise of the camp sector in the

1930s

in

terms of two factors. The first was the need to mobilise the rural underemployed into an active resource, a need met partly through the deport-

ation of the kulaks at the end of the

1920s .

The second was the fact that

in conditions of acute industrial labour scarcity and a disorganised labour market, the NKVD constituted a privileged recruiter of labour for its own industrial and extractive projects - that is, it was the only employer with

the power of arrest (Swianiewicz

1965) .

But he refrained from comment

on the efficacy of the camp economy itself . However, recent disclosures of both Medvedev and Solzhenitsyn, which have appeared in translation in English, tend to confirm that the labour camps also experienced problems of infeasible planning and 'plan breakdown', problems of economic control rather similar to those experienced in other sectors of the planned economy . (14)

In summary this economy of the

1930s,

which has so often been

described as a 'war economy', with a highly centralised system of 'command planning' and the adminstrative allocation of physical resources, was characterised by a series of contradictions . It is not just another matter of the contradiction between plan and reality - of course unforeseen circumstances and mistakes are inevitable . There was also the way in which the command economy itself generated the autarchic operation of particular industrial sectors. This command economy also induced chaotic disequilibria in labour markets and the markets for consumer goods (especially food), and coincided with widespread development of black markets . Moreover the operation of the economy diverted much human ingenuity from the task of economic construction to the needs of administrative survival . All this is not to deny the major difficulties, international, demo-



84

CAPITAL AND CLASS

graphic and technical, amid which the Soviet economy made striking gains . (15) But it does raise further 'fundamental' questions. Firstly, was the route pursued the most effective one under the circumstances? Barsov, for example, has suggested that the rate of planned accumulation in the early 1930s was too high ; a less ambitious plan would have enabled a higher rate of growth to have been achieved (Barsov 1969, p .96) . Well, let us pose it as a question . Firstly, could a different set of targets have achieved better results? Secondly, what was the influence upon this of the system of planning which emerged? A planning system is a set of social relationships . The rise of the Soviet planning system is also a transformation of social relations both at the economic, and at the political and ideological instances of society . We cannot begin to ask why the route which was followed was chosen, why a particular planning system emerged, and why it operated as it did, without raising, once again, the questions of political economy . THE QUESTION OF HISTORICAL ALTERNATIVES Through every upsurge and faltering of the Soviet economy, and through every crisis within the historical sciences, elements of triumphalism within Soviet historiography have persistently re-emerged . By contrast the dominant Western assessment has experienced several metamorphoses since the war . The shared experience of the war-time alliance against military fascism lent its own triumphalism to historical work (especially of a popular character) in Britain . Stalin was seen as the leader who was clear-sighted and ruthless enough to have done 'what had to be done' to preserve socialism in one country . To his personal domination was ascribed the successes of collectivisation in agriculture, and the rapidity of 'forced' industrial isation . The 'forcing of the pace' was seen as an integral and functional element in socialist economic development . The inauguration of the Cold War translated this view into its mirror image - the preservation of socialism requires a Stalinist dictatorship, and a permanent 'war economy'. If Stalin had done 'what had to be done', it was in order to preserve a totalitarian political system akin to Nazi fascism . There simply did not exist any problematic which could have enabled us to examine the possibility of historical alternatives . To what extent was the Stalinist economic and political system integral to socialism, in generalor in one country? In what sense could one speak of different political and historical options open to the Soviet state and its people in the 1920s and 1930s? Since then many hew doors have opened . In the USSR the 20th and 22nd Congresses of the CPSU, and the discussions of the 1960s on the reform of the economic mechanism, have greatly influenced Marxist thought in the socialist countries and in the West . So have the different advances and false starts in the other socialist countries . The emergence of powerful tendencies towards a military detente between the superpowers has also served to weaken many bourgeois ideological dogmas. All this has made possible new historical work, and new assessments .



SURVEY: THE SOVIET ECONOMY

85

Today, relatively few writers sustain the inevitability thesis . A much more common position is the standpoint of pragmatism : one may draw up a list of the costs and benefits accruing to the Stalinist strategy of economic development, and leave assessment to the reader (16) . In characterising this pragmatism, I do not dispute the very great contribution to historical understanding made by many who adopt it . Nevertheless this pragmatism reflects a failure of bourgeois economics in the post war period . Firstly, such pragmatism is inherently relativistic . To define the costs and benefits of the Stalinist economic strategy, it is necessary to define an alternative 'scenario' of development . But which alternative? Suppose the October revolution had never happened : would Russia today be another peripheral agrarian-fascist regime - or would it be Europe's economic miracle? Suppose not Stalin but Bukharin, or Trotsky, had succeeded to the Bolshevik leadership : would they have been compelled to follow the same road - perhaps with a more or less human face? Were there, in fact, real political options? One cannot deny the endless demagogic possibilities of this or that alternative . As Carr reminded us, those who ask what might have happened in the absence of the October revolution usually do so because they regret its failure not to have happened (Carr 1964, pp .96-7) . Secondly, however, such questions cannot be denied meaning . What is important is to situate them politically . The failure of pragmatism, in this sense is to see the evolution of the state and of planning in terms of 'economic policy', rather than of political economy . The victory of one economic strategy over another is not just the outcome of a process of policy choice . It also means the political defeat of one bloc of forces, the defeat of its programme, its ideology and consciousness, at the hands of the victors . The concept of historical alternatives must therefore be founded in analysis of the material forces which were in fact contending for power. At the same time this must always be a logical exercise rather than an attempt to rewrite an alternative history . From this point of view, one of the most significant developments recently has been the attempt to recover historically the programmatic alternatives associated with Bukharin . As well as Cohen's political biography, there is also Lewin's attempt to trace the continuities from Bukharin, defeated in the 1920s, to the revival of alternative economic concepts in the 1960s (Cohen 1974 ; Lewin 1974) . The significance of this work (as well as of his own) is best summed up by Szamuely in the conclusion to his monograph : "The familiar course of history has not proceeded along this (Bukharin's : MH) lengthier way of socialist construction but has implemented a shorter though more expensive solution . Later generations are no longer in a position to revise or invalidate the once realised choice of historical alternatives : these are unalterable facts . But, simply because they happened thus, because of their being facts, these decisions are fixed in the consciousness of later generations as necessary, even as the only possible solutions ; and when, in new situations, some



86

CAPITAL AND CLASS similar alternatives have to be decided between, ready-made solutions may have a powerful retrograde impact - if we do not exactly know the substance of the original alternatives and the decisive circumstances of the historical choice ." (Szamuely 1974, p .109)

POLITICAL ECONOMY AND THE FETISHISM OF FORMS One of the most famous Marxist conceptions is that the fetishism of commodities, arising within capitalist production, makes necessary the political economy of capitalism . In the capitalist mode of production social relations between people appear as exchange relations between things . To penetrate this surface layer of social reality requires a scientific, materialist approach, the approach of Marxist political economy . A corollary, generally accepted by Soviet Marxists in the 1920s, was that under socialism social relations would for the first time be nakedly visible to human experience . Socialism would bring the abolition of money forms, of commodity production and trade ; the law of value would be displaced by planning . The need for a scientific political economy would disappear, its place taken by politics and political choice . (17) Looking back, this view can now be seen as an idealisation of what Engels termed 'barrack socialism', of a naturalised, militarised and autarchic economy, which willingly found its first model in the Soviet economy of War Communism - the battle for survival in the civil war of 1918-1921 . (18) . This view was also part of the ideological formation of the Stalin period, of the first Five Year Plan and of Stalinist 'political economy' . (19) A further attribute of this 'political' economy, also noted by Szamuely, was the reactionary notion that the consciousness of the masses, dulled by their experience of fetishised capitalist relations, must be replaced by the organising consciousness of the socialist state (Szamuely 1974, p .32) . At the roots of this ideology can be identified a fetishism of the bureaucratic state and of command planning . Seeing the masses as the passive instruments of a dominant state, it could not comprehend the phenomena of spontaneous activity of the sections and classes of Soviet society - of mass resistance to collectivisation ; of mass demoralisation (reflected in high rates of labour turnover, absenteeism and industrial accidents) in the early thirties ; of widespread participation in semi-legal and illegal trade ; of mass involvement both in the euphoric utopianism of voluntaristic planning, and in all kinds of administrative strategems to distance each individual or group from the immediate consequences of voluntarism . Or at least, these phenomena could only be comprehended as the survival of petit-bourgeois tendencies, the work of class enemies and of foreign agents . All this stood for the failure of a society to understand its own experience of social transformation from which, instead, were created new fetishes . At the same time a Marxist sub-current has persisted, based upon the persistence of commodity forms under socialism, which recognised the possibility of different forms of socialist property and of state planning ;



SURVEY: THESOVIETECONOMY

87

which possessed a critique of the bureaucratic state, but not of commodity forms . In the pre-war period this sub-current was incorporated politically in the tradition of Bukharin, and academically in the Lange-Lerner school of 'market socialism' . This latter tradition is much stronger today than ever before. It has played an important and progressive role in the USSR in the post-Stalin debates on economic reform (Lewin 1974 ; Ellman 1973, especially pp .] 34-

17S) . Because of its lack of a critique of markets, however, in the West this tradition has been partly absorbed into the conventional pragmatism . Thus for most Anglo-American students of the Soviet economy, the question of economic reform is still an essay on pragmatism . Given the 'social parameters' for and against change, given the strengths and weaknesses of 'commandism' and 'market' solutions, which option will improve the growth of the Soviet national product? The partial failure of the Bukharinist political economy is reflected in another type of reductionism : the persistent reduction of questions of the development of socialist social

relations to a matter of the physical increase in production . (20) Sixty years after the October revolution it is clear that socialist societies generate their own ideological formations ; that socialist relations of production, and the relationship between state and society are problematic rather than automatically constituted and nakedly visible, and that is why we need a political economy of socialism . The possibility of progress today exists at two levels . Firstly the historical work described above strongly challenged the idea of the functionality and inevitability of the bureaucratically organised 'war economy' of socialism . Secondly, in trying to tackle the modern Soviet and East European reality, a handful of writers has sought to establish a critique of

both the bureaucratic state and commodity forms under socialism (21) . This integration - and critique - of the two traditions can be seen as a first stage of renewal of the political economy of socialism . Probably the most coherent contribution has been that of the Polish economist Brus (now in emigration in Britain) . At one level he has restated the necessity of a correct integration of plan and market (rather than of

opposition between them) . (22) But this has been said by others and is not new . More importantly, he has re-established the notion of socialism as a process of development, in which the relations between state and society are constantly being transformed . (23) On the one hand it is necessary to establish the proper scope and limits, in each historical situation, of local, enterprise and trade union autonomy within the planning framework . On the other hand it is equally necessary to ensure the proper mass-democratic control over the state and planning itself . Thus the transformation of socialist social relations consists not only of one moment - the moment of nationalisation and the creation of state property - but of a process of socialisation of the means of produc-

tion. (24) It can be argued that the form of 'command' planning and centralised state ownership is only one crude, primitive form of social ownership . The fetishism of this form, born historically in the conditions of 'socialism in one country', retards the development of mass creativity,



88

CAPITAL AND CLASS

or diverts it into dysfunctional channels . It also retards our historical understanding of the evolution of the Soviet economy . But the relationship between central planning and autonomous creative forces cannot be solved independently of control over the plan itself . No matter how great is the national product, sectional autonomy or consumer choice, they will not suffice to develop socialist society to new and higher levels without democratic control over the centre . Finally, behind these two political economies stand two opposing programmes for working class power . According to the version shared by Stalinists and Trotskyists, the power of the working class is posed primarily at the level of the coercive, political state and law, of 'commanding heights' in the 'war economy' . In the version belonging to the Bukharin tradition, class power is posed in terms of a democratic alliance of popular forces which is based first and foremost in civil society but which has, of course, expression at the level of the state . CONCLUSION In the last decade a significant transition has occurred . In the fields both of theory and of knowledge (and both are essential to learning from history) we have seen many important advances . Still they are only a beginning . A Cold War mentality persists in many other areas of Soviet studies in Britain and the United States, from literature to strategic studies . There are also problems of finance, as well as the usual question of access to primary achival materials, although there are now many established research links with Soviet institutions . Another problem is that to develop Marxist ideas requires Marxists . In Britain, for example, there are only two Communists engaged in research on the Soviet economy, past or present . But in a wider sense Marxist scholars in Britain (I cannot speak for the USA) have new organisations and journals with which to confront our new possibilities. (25) So the real work still lies ahead . One of its most vital components must be a great awareness of new work in other countries . Both American and Soviet scholarship are having an important impact in Britain today . For American publications the time-lag is, of course, rather brief . But it takes up to five years for major Soviet works to have their full impact in Britain, even at the highest academic levels (I will not speculate on the time-lag which operates in the opposite direction) . Most of us barely know the names, let alone the writings of the major scholars in France, Holland, the FRG or Italy . Therefore I hope that we in Britain will also play our part in remedying this situation .

FOOTNOTES Lecturer in Economics, University of Warwick, Coventry, England . The author is a member of the Committee for Study of the European Socialist Countries of the Communist Party in Great Britain . He is grateful to a



SURVEY: THE SOVIET ECONOMY

89

number of scholars who have kindly allowed him to refer to their unpublished work . The views expressed by the author are his personal responsibility, as are any errors of fact, interpretation or omission . An earlier version of this paper was printed in Italian translation in Studi Storici, Rome . 1 His speech 'On the Grain Front' of 1928 was translated, for example, in Stalin, 1940, one of a number of selections through which his writings became known to the British public . More recently this view has been supported, with varying reservations, by Dobb, 1966, pp .216217, Lewin, 1968, pp .176-177, Straus, 1969, p .84. Carr and Davies 1974, p .971 . 2 Among Communists a somewhat critical acceptance of this thesis has become the norm . See for example Pollit, 1972, Dunman, 1975,p .123 3 Millar, n .d . suggests that the popularisation of this view originated with Erlich, 1950. See also Erlich, 1960, Spulber 1964, Dobb 1965 . Preobrazhensky's 1926 was published at this time in English translation (by Brian Pearce) as The New Economics . One of the most critical analyses of the traditional type was that of Mandel 1968, pp .553-554 . In Britain further work is being continue on the dynamics and 4 efficiency of agricultural production by S .G . Wheatcroft, (n .d . (a) (b), (c)) . 5 This question is approached in Harrison, n .d . but work in this area is still continuing . A diametrically opposite viewpoint -that excessively favourable terms of trade were at work - is suggested by Millar, 1970 . 6 Barsov 1969, considers the years 1928-1932 . In a later article, Barsov 1974, he develops this theme, introducing data for 1937 and 1938 . 7 This discussion followed on the heels of that involving Karcz and Davies . See Millar, 1970, Nove 1971 a, Millar 1971, Nove 1971b . The first full presentation of the Barsov data awaited Millar 1974, and Ellman 1975 . What was at stake? Nove disputed that the agricultural surplus could be measured at all ; Millar disputed the use by Ellman (following Barsov) of rouble-values transformed into Marxian labourvalues as a measure of the transfer of surplus-product . 8 This view tends to underlie the interesting and important work of Lewin 1968 . See also, in this connection, Lewin 1966 . This view was stated more strongly and explicitly by Shanin 1971 and 1972 . A considerable impetus has been given to this line of work by the appearance of Chayanov 1966, a translation into English of Chayanov's 1925 . For discussion of the Chayanov phenomenon see footnote 5 above, and also Harrison 1975 . 9 Szamuely 1974, p .41 . Here in fact Szamuely refers to the views of Bukharin in 1920, when the latter was still closely associated with Preobrazhensky and the left . To one not familiar with the state of Hungarian sciences Szamuely's work contains a surprising originality and richness ; its translation from the Hungarian deserves a big welcome . See also Millar, n .d . 10 This article has been reprinted with some revision (Hunter 1976) . Hunter's work is continuing in the field of computer simulation of Soviet economic growth in the 1930s .



90

CAPITAL AND CLASS

11

This development of sectoral autarchy may explain part of the contrast between impressions of overall chaos, and of the rational organisation of rapid expansion and technical advance in particular sectors . See for example Cooper n .d .(b), particularly interesting for its assessment of the serious economic burdens imposed by the threat of the coming European war ; Cooper 1976, Lewis n .d ., a record of the creation of a synthetic rubber industry . This is discussed in relation to Hunter's work by Lewin 1973 ; a somewhat less catastrophic picture is presented by Davies n .d . (a) . Spour's article marks the first attempt in Britain to apply Gramscian concepts to the evolution of the Soviet state . Medvedev 1972, and more specifically Solzhenitsyn 1975, both contian many points of relevance . However the latter source is hardly unprejudiced . Some problems of interpretation are considered by Harrison 1976 . There are interesting developments in the attempt to analyse exactly which difficulties were exogenous, and beyond the control of the Soviet state ; see once again Cooper, Davies and Wheatcroft, n .d . The ability of the Soviet economy to benefit from the import of Western technology in the twenties has been documented by Sutton 1968 . On the reality of external constraints in the subsequent collapse of trade relations and technical agreements see Dohan 1976 . For example Wilber 1969, pp .109-133, Gregory and Stuart 1974, pp .442-446, Nove 1975, pp .109-113 . In other respects all these are excellent and interesting works, though Wilber's is now very dated . Gregory and Stuart's is one of the best modern text books . Curiously enough, Marxists were aided and abetted in this theoretical abstraction by some bourgeois economists who chose to continue work in the Soviet Union . See for example A .V . Chayanov 1921 . This is the main theme of Szamuely 1974 . Some stages and levels in these developments are considered by Davies n .d . (b) . But there are still primitive survivals of the old political economy, even in Britain : see for example Ticktin 1974, a reply to Selucky 1974 . The relationship between the development of the productive forces and the qualitative changes in social relations has received considerable attention in the Soviet Union since the war . This field is thoroughly surveyed in Cooper n .d . (a) . See Ellman 1973 . A number of important essays by Maurice Dobb provided an essential background to this development, for example his 1970, 1975 and his last essay, published after his death, 1976 . See also Brus, 1972, 1973, 1975 . See for example Brus, 'Some General Problems of Decentralization in a Socialist Planned Economy' and 'Economic Calculus and Political Decision' (the latter was first published in Warsaw) in Brus, 1973 . Brus, 'Commodity Fetishism and Socialism' in Brus 1973, is an essay to which the present author would like to acknowledge a special debt . There seems to be more than formal similarity between Brus's idea of the 'socialisation' of state property, and the Gramscian notion of the 'civilisation' of the state in socialist society .

12 13 14

15

16

17

18 19

20

21

22 23 24



SURVEY: THE SOVIET ECONOMY

91

25 The traditional organisation of Soviet scholarship in Britain is the National Association of Soviet and East European Studies, with its own journal Soviet Studies published in Glasgow . At the end of the 1960s the Conference of Radical Scholars of Soviet and East European Studies was established (also in Glasgow), and publishes a journal called Critique which has contained much interesting material (although practically nothing on economic history) . Most recently the Communist Party of Great Britain, through its new Committee for Study of the European Socialist Countries, has started to publish a slim journal, Socialist Europe, of which two issues have appeared at time of writing .

BIBLIOGRAPHY A .A . Barsov, 1969, Balans stoimostnykh obmenov mezhdu gorodom i dereveni, Moscow . A .A . Barsov, 1974, 'Nep i vyravnivanie edknomicheskikh otnoshenii mezhdu gorodom i derevnei', in Novaya ekonomicheskaya politika : voprosy teorii i istorii, Moscow . Wlodzimierz Brus, 1972, The Market in a Socialist Economy, London . Wlodzimierz Brus, 1973, The Economics and Politics of Socialism, London . Wlodzimierz Brus, 1975, Socialist Ownership and Political Systems, London . Robert Campbell, 1973, 'What Makes a Five-Year Plan Feasible?', in Slavic Review, June . E .H . Carr, 1964, What is History?, Harmondsworth . E .H . Carr and R .W . Davies, 1974, Foundations of a Planned Economy, Voi I, Harmondsworth . A .V . Chayanov, 1921, 'Ponyatie vygodnosti sotsialisticheskogo khozyaistva' in V .A . Chayanov and A .L . Vainshtein, eds ., Metody bezdenezhnogo ucheta khozyaistvennykh predpriyatii, Moscow . . A .V . Chayanov, 1966, The Theory of Peasant Economy, Homewood, Illinois . Stephen F . Cohen, 1973, 'Stalin's Revolution Reconsidered', in Slavic Review, June . Stephen F . Cohen, 1974, Bukharin and the Bolshevik Revolution, New York . J .M . Cooper, n .d . (a) 'The Concept of the Scientific and Technical Revolution in Soviet Theory', RC/C No . 9, University of Birmingham . J .M . Cooper, n .d . (b), 'Defence Production and the Soviet Economy 1929-41', SIPS No .3, University of Birmingham . J .M . Cooper, 1976, 'The Development of the Soviet Machine Too] Industry 1971-1941', Ph .D thesis, University of Birmingham .



92

CAPITAL AND CLASS

J .M . Cooper, R .W . Davies and S .G . Wheatcroft, n .d ., 'Contradictions in Soviet industrialisation, University of Birmingham, unpublished . R .W . Davies, n .d . (a), 'The Soviet Economic Crisis of 1931-33', SIPS No .4, University of Birmingham . R .W . Davies, n .d . (b), 'The Emergence of the Soviet Economic System 1927-1934', SIPS No .9, University of Birmingham . R .W . Davies, 1970, 'A Note on Grain Statistics' in Soviet Studies, January . R .W . Davies and S .G . Wheatcroft, 1975, 'Further thoughts on the First Soviet Five-Year Plan', in Slavic Review, December . Maurice Dobb, 1965, 'The Discussions of the Twenties on Planning and Economic Growth', in Soviet Studies, October . Maurice Dobb, 1966, Soviet Economic Development since 1917, 6th ed ., London . Maurice Dobb, 1970, Socialist Planning : Some Problems, London . Maurice Dobb, 1975, 'Some Historical Reflections on Planning and the Market', in C . Abramsky, ed ., Essays in Honour of E.H. Carr, London . Maurice Dobb, 1976, 'Commodity Production and Socialism', in Socialist Europe, No .] .

Michael R . Dohan, 1976, 'The Economic Origins of Soviet Autarky 1927/28-34', in Slavic Review, December . Jack Dunman, 1975, Agriculture : Capitalist and Socialist, London . Michael Ellman, 1973 Planning Problems in the USSR, Cambridge . Michael Ellman, 1975, 'Did the Agricultural Surplus Provide the Resources for the Increase in Investment in the USSR during the First Five Year Plan?', in Economic Journal, December . Alexander Erlich, 1950, 'Preobrazhensky and the Economics of Soviet Industrial isation', in Quarterly Journal of Economics, February . Alexander Erlich, 1960, The Soviet Industrialisation Debate, Harvard . Paul R . Gregory and Robert C . Stuart, 1974, Soviet Economic Structure and Performance, New York, London . Mark Harrison, n .d ., 'Soviet Peasants and Soviet Price Policy in the 1920s', SIPS NO . 10, University of Birmingham . Mark Harrison, 1975, 'A .V . Chayanov and the Economics of the Russian Peasantry', in Journal of Peasant Studies, July . Mark Harrison, 1976, 'The Gulag Archipelago : Solzhenitsyn as Historian', in Socialist Europe No .1 . Mark Harrison, 1977, 'Resource Allocation and Agrarian Class Formation : The Problem of Social Mobility Among Russian Peasant Households 1880-1930', in Journal of Peasant Studies, January . Holland Hunter, 1961, 'Optimal Tautness in Development Planning', in Economic Development and Cultural Change, July . Holland Hunter, 1973, 'The Over-Ambitious First Soviet Five-Year Plan', in Slavic Review, June . Holland Hunter, 1976, 'A Test of Five-Year Plan Feasibility', in Judith Thornton, ed ., Economic Analysis of the Soviet-Type System Cambridge . Naum J asny, 1961, Soviet Industrialisation 1928-1952, Chicago . Jerzy F . Karcz, 1967, 'Thoughts on the Grain Problem', in Soviet Studies, April .



SURVEY. THE SOVIET ECONOMY

93

Jerzy F . Karcz, 1967, 'Back on the Grain Front', in Soviet Studies, October Michael Kasper, 1962, 'The Nature of Soviet Planning', in Soviet Studies, October . N . Lampert, n .d ., 'The technical intelligentsia in the 1920s and 1930s', University of Birmingham, unpublished . Moshe Lewin, 1966, 'Who was the Soviet kulak?', in Soviet Studies, October . Moshe Lewin, 1968, Russian Peasants and Soviet Power, London . Moshe Lewin, 1973, 'The Disappearance of Planning in the Plan', in Slavic Review, June . Moshe Lewin, 1974, Political Undercurrents in Soviet Economic Debates, London . Robert A . Lewis, n .d ., 'Priorities in the First Five-Year Plan : The Building of SK-1', SIPS No .6, University of Birmingham . Ernest Mandel, 1968, Marxist Economic Theory, Vol .l I, London . Roy Medvedev, 1972, Let History judge, London . James R . Millar, n .d ., 'What Does Primitive Capitalist (Socialist) Accumulation Really Mean?', University of Illinois, unpublished . James R . Millar, 1970 (a), 'A Reformulation of A .V . Chayanov's Theory of The Peasant Economy', in Economic Development and Cultural Change, January . James R . Millar, 1970 (b), 'Soviet Rapid Development and the Agricultural Surplus Hypothesis', in Soviet Studies, July . James R . Millar, 1971, 'The Agricultural Surplus Hypothesis : A Reply to Alec Nove', in Soviet Studies, October . James R . Millar, 1974, 'Mass Collectivization and the Contribution of Soviet Agriculture to the First Five-Year Plan', in Slavic Review, December . Yu . A . Moshkov, 1966, Zernovaya problema v gody sploshnoi kollektivizatsii sel'skogo khozyaistva SSSR, Moscow . Alec Nove, 1971 (a),'The Agricultural Surplus Hypothesis : A Comment on James R . Millar's Article', in Soviet Studies, January . Alec Nove, 1971 (b), 'A Reply to the Reply', in Soviet Studies, October. Alec Nove, 1975, Stalinism and After, London . Brian Pollit, 1972, 'The Soviet Economic Debate in the Twenties', in Marxism Today, April . E . Preobrazhensky, 1926, Novaya ekonomika, Moscow . E . Preobrazhensky, 1965, The New Economics, Oxford . R . Selucky, 1974, 'Marxism and Self-Management', in Critique No .3 . Teodor Shanin, 1971, 'Socio-Economic Mobility and the Rural History of Russia 1905-1930', in Soviet Studies, October . Teodor Shanin, 1972, The Awkward Class, Oxford . Alexander Solzhenitsyn, 1975, The Gulag Archipelago, Vol .[ I, London . Ken Spours, 1976, 'Marxist Theory and the Soviet Superstructure', in

Socialist Europe No .] . Nicolas Spulber, 1964, Soviet Strategy for Economic Growth, Bloomington, Indiana . Joseph Stalin, 1940, Leninism, London . Erich Strauss, 1969, Soviet Agriculture in Perspective, London .



94

CAPITAL AND CLASS

Anthony C . Sutton, 1968, Western Technology and Soviet Development, 1917 to 1930, Stanford . S . Swianiewicz, 1965, Forced Labour and Economic Development, Oxford . Laszlo Szamuely, 1974, First Models of the Socialist Economic Systems, Budapest . H .H . Ticktin, 1974, 'Socialism, the Market and the State', in Critique No .3 S .G . Wheatcroft, n .d . (a), 'Views of grain output, the grain forage balances and grain surpluses in the 1920s', University of Birmingham, unpublished . S .G . Wheatcroft, n .d . (b), 'Soviet grain production and consumption in the 1920s and1930s', University of Birmingham, unpublished . S .G . Wheatcroft, n .d . (c), 'Work in progress on the reappraisal of the efficiency of Soviet agricultural production in the 1920s and 1930s', University of Birmingham, unpublished . Charles K . Wilber, 1969, The Soviet Model and Underdeveloped Countries, Chapel Hill, North Carolina .

DEBATE TRADE UNION INTERNATIONALISM AND THE SUPRA-NATIONAL STATE John S . Baker

This note is a comment on an article in a previous issue . It argues that international trade unionism is at bottom a product of workers and their aspirations ; its development clearly involves responses to the internationalisation of capital, among many other features of capitalist development, but it is not governed solely by this . In actual fact, however, international trade unionism has been penetrated and moulded by U .S . corporations and the CIA, through U .S . union leaders, in order to make it a tool for breaking down barriers to the expansion of U .S . multinationals .

The contribution by Werner Olle and Wolfgang Schbller, World Market Competition and Restrictions upon International Trade Union Policies (Capital & Class No.2), is an excellent "opener" to a subject rarely touched upon by Marxists . In particular the 4th-last paragraph of the O. & S . contribution moves, in my view, to one of the underlying, barely recognised questions for everyone in capitalist societies : " : . . namely the methodological question of what preconditions are needed for an 'internationalised reproduction of capital' . In our view just as the process of constituting a total national capital was the result of functions of the national state, so even the tendency towards the constituting of a 'true historical world national capital' (Neus(lss) logically and historically presupposes supranational statehood . Such a supercession of the national state cannot therefore be thought of in terms of some economic mechanism, but implies politico-military action, which may well have its ultimate roots in economic relations, but which can in no way be described as an exclusively economic phenomenon ." Their concept of the "constituting of a true historical world national capital" implies, even if this is not specified, the regulation of an inter-



96

CAPITAL AND CLASS

nationalised trade union movement paralleling, but not basically challenging, the supranational state power of the "West" . It is precisely that conjuncture of trans-national power and international unionism accommodation which is now proceeding . This makes many points raised by O .& S . of considerable interest to me . I would like to touch briefly upon this background scenario of international relations that I share with O .& S . before coming to what they put forward as the mechanics of the new power within international unionism . From a study of the secret military and electronic surveillance/ missile guidance pacts - the ABCA Pact and the UKUSA Pact - which bind the five WASP nations (the US, the UK, Canada, Australia, New Zealand) and which then encompass the system of NATO pacts, those covering Latin America, those covering the Middle East and those of the earlier and later phases in South East Asia up to Japan, one must conclude that here is the military relations base of a consolidating supranational power . The central agreement of this network of contemporary power is the Quebec Agreement of 1943, which expanded into the ABCA and Four Powers' Pact, whilst extending its operations from nuclear to chemical and bacteriological warfare also . The US Army is the only authority recognising the secret pact's existence . Some CIA agents, like Marchetti, maintain that most of the agreement-structure is illegal and kept so deliberately because of CIA and US Army strategies . Moving from the military elements within the supra-nationality to communications, the National Security Agency (NSA) system of 2,200 electronic (also missile guidance and deflection) bases has operated effectively for 20 years no-one acknowledges the existence of the Pact and this is complemented by the COMSAT/INTELSAT system . (West Germany and Japan are reported unofficially to have been admitted as "second parties" .) (1) Within the US transnational corporation (TNC) there is a firm economic base of the new supra-national state system . It has an economic empire outside of the US exceeding the GNP of any other nation than the US itself. The U .S . TNC's have the basis for a "true historical world capital" and appear to have created some of the supra-national state regulatory organs for their exercise of power . Recognition of the existence of the supra-national state power or potential power leads us to look among the main U .S . power centres for some consciousness of corresponding power within U .S . unionism . Such a fusion of power and purpose is to be found within the U .S . AFL/CIO . The theoretical exposition of this new trade union power and policy can be seen in the studies by, for example, R . Cabot Lodge, for the U .S . National Policy Committee in 1958 . If O . & S . locate the World Corporation Councils (WCC's) within their concepts of the developing supra-national state power, I would not rule them out . Any form of collaboration at the top of the union-corporate pinnacles must be given its due importance . But O . & S ., in my view, give them an importance beyond the areas of collaborationism "at the top" . By 1960, the main ideologues of the role of international unionism within the supra-national state power saw the old system of international



DEBA TE :DOGMA TISM

97

trade secretariats (ITS's) with their headquarters in Washington, Denver, London and Geneva, as the all-important form of unionism for their economic and military expansionist needs . Notwithstanding the insights of Charles Levinson (0 . & S . p .57) about "the probable trade union route toward the creation of a global counterforce" and for "completely new trade union structures", it seems inescapable that most of the basic issues of the national working-classes and international working-class will be fought within the existing, traditional forms of unionism, within the sovereignty of the national or, perhaps, regional union centres - TUC, AFL/CIO, CETU, CGI L and LO&C - and within the sovereignty of the national state or an economic community of nations, and against the thrusts of the supra-national state through the ITS system . As a special arm of the supra-national state within the working-classes of the non-socialist lands, the CIA can now direct some of the assistance given to the TNC's during their world expansion into the extension of the trade union education programme, in order to make acceptable the new economic order . In their comments on "New Forms of Trade Union Internationalisation through Multinational Corporations" (0 . & S . section 2), our German colleagues are perhaps too uncritical of the World Corporation Councils . These WCC's are, to me, part of the middle-1960 "Detroit Psychology" which gave rise to the World Auto Councils . Must it be the case that the multinational corporations have created economic conditions making trade union internationalisation possible and desirable? Not necessarily, although some of the TNC's hack writers and theorists put forward such a view . In Australia, for example, some of the unionists most involved in the ITS-TNC-CIA operations internationally put forward the view of the TNC's that the barriers of the nation-states to the penetration of the TNC's must be eliminated . This progressive move, they argue, is the equivalent in Australia to the lowering of state barriers with the federation of the colony-states in 1900 . In spite of the fact that the WCC's evolved as part of the middlesixties 'Detroit Psychology', and also then extended to some of the European automobile industry centres, their acknowledged purposes - multinational bargaining - have not been realised . In their pursuit of what may have been even an acceptable programme to many national trade union leaders, the hopes placed on the WCC's may have kept the stronger of the national trade unions from the search for more basic and more successful programmes of international trade unionism . By holding out the possibility of international trade unionism, the ITS's and WCC's have simultaneously held back the development of stronger forms of working class organisation and smoothed the way for the further growth of the TNC's . The parallel operations of the U .S . TNC's, the ITS system and the special world agencies created by the conferences of corporation leaders like the "amazing" J . Peter Grace and the fifty that participated, the AFL/ CIO leaders - Meaney, Beirne, Lovestone, Buckley - and CIA clandestine operations men like Bissell, Dulles, Brown and Goldberg, have now become



98

CAPITAL AND CLASS

institutionalised in these new international operations . The American Institute for Free Labor Development (AIFLD), the American-African Labor Centre, the Asian-American Free Labor Institute (AAFLI) and the Inter-American Regional Labor Organisation (ORIT) became world agencies, related to specific regions, which embarked upon the worker indoctrination and organisation programmes needed by the U .S . TNC's . For the first five years of the system, the main group of U .S . TNC's paid 17 per cent of the costs of the huge operation . Now almost all the cost have been passed to the U .S . State Department, which channels the funds throughthe Agency for International Development (AID) into the ITS system into the individual U .S . unions which play key roles in the system, and then into the world regional centres like Geneva, Singapore, Washington, Nairobi . Some of the funding documents from U .S . Senate Committee investigations provide a small part of the total operations but, interestingly, have now been projected for the period 1977 to 1981 with a 10 per cent increase annually . From roughly 1958, there has been a fairly general consensus in the writings and statements from the U .S . National Policy Committee (NPC), the National Security Council (NSC) and the ideologues of the corporations and unions that the International Trade Secretariats (ITS's), many of which had their origins in Europe from 1890 to 1920, were to be the main form of international trade unionism . Economic penetration by the TNC's was seen as having two main supporting forces - the counter-insurgency forces developed under the military assistance programmes (MAP) and the new indoctrinated unionists from the education programmes organised by the AIFLD, AAFLI, AALC, and ORIT as well as by the individual ITS's, such as the PTTI, IMF, IFPCW, ITF and others (there are about 20 significant ITS's in all) . It has been remarked upon that the "dual" training programme resulted in the training of almost equal numbers of military counter-insurgents as of "educated" trade unionists . In Brazil, Guyana, San Domingo, Chile, and in African and South East Asian countries the roles of these trade union mercenaries is something quite new in international trade unionism . The U .S . State Department, CIA or OSS (Office of Strategic Services, as it was at the time), and the AFL (before the amalgamation with the CIO) having set up and firmly managed the International Conference of Free Trade Unions (ICFTU), used this base to extend its penetration and controls over the International Trade Secretariats . Their controls extended so far by about 1963, that some of the European union centres affiliated to the ICFTU began to establish the ECFTU and later established the European Trade Union Confederation (ETUC) which specifically rejected the word "free" in its title and rejected affiliations or formal relationship with either the ICFTU or the ITS system . Gradually, the ETUC began to develop co-operation with the CGT and CG IL on specific issues . "Most free and Christian unions in the EEC are now ETUC affiliates, as well as the Communist Italian CGIL . . . . There is speculation that the French Communist CGT will join ETUC . . ." (Rowan and Northrup 1975) .



DEBA TE: DOGMA TISM

99

At the same time the TNC-CIA-AFL/CIO system of strengthened and ex panded ITS's continues to grow in size (now more than 60 million affiliated unionists), exceeding the numbers in the ICFTU and approaching those of the World Federation of Trade Unions (WFTU) . There are few areas of the world today into which the TNC's pene-

trate which haven't sectors or some level of the capitalist mode of production . Those conditions, not specifically the arrival of the TNC, create the

prerequisites for unionism . O . & S ., I feel, rather over-simplify this process . A visit to and examination of some of these related factors of capitalist mode of production and unionism in some of the partly tribalised areas of Northern Australia and Papua-New Guinea, and then through some of the other cultures to the North, from Indonesia through to South Korea, illuminate these relationships in many, if sometimes curious ways . In some instances, the TNC's now moving their US and European and even Japanese plants into the 'new production' and 'free production' areas in Indonesia, Singapore, Taiwan, Hong Kong, South Korea and the Phillipines want unionism, but unionism of the kind their production methods need, and at other times and places oppose unionism of any kind totally .

Those objective demands for unionism, which is then contained, as a rule, within the ITS system, often envisage an alliance between plant or off-site union leaders and the TNC, which helps to discipline the work-force and applies pressure on the local government for what the TNC requires in its own and its employees' interests . The new production centres replicate many of the conditions of the old U .S . "company towns", combined with a controlled unionism . Wages are depressed by international standards and continue so . There are tax lures for the corporation and the management is able to summon police and the military for surveillance and control duties . If and when the union is institutionalised, its officials are part of the deal . Love of the union is imposed on the local culture-patterns . Where the TNC, on the advice of its anthropologists, decides there mustn't be any unionism, it takes about two years before the workers learn and by various ways convince the management and local authorities otherwise . By the end of five years the management begins to face the dilemmas : unionism of a somewhat independent kind has developed and demands for pay increases and better conditions are threatening output and profits, or new production areas have opened elsewhere and there a workforce without unionism or contained by company "unionism" is available . They usually move on . There is often a 'progress' of the U .S . TNC, particularly in electronics, into Mexico, Japan, and then to South Korea and, finally down to the bottom of the barrel, Indonesia . However, the importance O . & S . place on the WCC system, as a first expression of some counter-force to the TNC's, seems misplaced . In the creation of the WCC system, the leaders of the U .S . Union of Auto Workers, Walter and Victor Reuther, had a leading part . By the middle-1960s the Reuthers, who had been searching for a 'Detroit Psychology' that would meet the challenges of automation, job-disappearance and more revolutionary demands by some of the black-led locals of the UAW, were also being confronted by the challenges of the Japanese and European auto



100

CAPITAL AND CLASS

industries . Their internationalism led them to a union-corporation solution in which jobs of U .S . workers were preserved, but those of workers in the Brazilian, Australian and other GM and Chrysler plants were not hurt too much . At that stage the Reuthers were trying to disengage from the web of CIA activities - particularly their role as conduits of U .S . State DepartmentCIA funds - and were about to launch their big offensive against the AFL/ CIO over its tie-up with the CIA/TNC's and the network of world agencies they had created for the containment of the international trade union movement. Many consider the Reuthers to be the originators of the WCC idea : "At the 1964 Automotive Department meeting in Frankfurt, the IMF (International Metalworkers Federation) finally gave its formal commitment for the creation of world auto councils . Two years later at a meeting in Detroit, permanent councils were established for Ford, General Motors and Chrysler. Since then, six others have been organised : Volkswagen-Daimler Benz, and British Leyland (1966), Renault-Peugeot (1971), and Toyota and Nissan (1973) . A FiatCitroen council was formed in 1971 and became the Fiat Council in 1973 when the two companies dissolved their partnership agreement . Reuther made it clear at the 1966 Detroit meeting that the purpose of the auto councils was to further international union co-operation which would eventually lead to multinational collective bargaining . . . 'in Brazil or in Great Britain, in Canada or Australia, in Germany or in any other part of the world, he would know precisely what GM is doing about that kind of problem in every other part of the world . He should not be talking about vacation pay as though he was the only one dealing with that problem' ." (Rowan & Northrup 1975) . The WCC's have increased to 32 and have organisational relationships with 5 ITS's and meeting places in seven national centres . In spite of attempts by them to move to trans-national bargaining on pay and conditions and much persistence and patience displayed by the unions and ITS's involved, the TNC's refuse absolutely to bargain other than nationally . In some instances, powerful unions' leaders have been ignored and humiliated when, after travelling long distances, they tried to get some wage negotiations even started . In their power relationships with the TNC's through the WCC's, union leaders found themselves in a much weaker position than when the confrontation was between national unions and the single TNC firm . Their humiliating treatment in the attempts to secure trans-national bargaining seems to reflect these changes in the balance of power . Discounting some of the earlier estimates in an article in the Harvard Business Review (March 1974), Burton Bendiner, the IMF's World Auto Council Co-ordinator dismissed many of the earlier claims : "The work of the World Auto Councils has occasionally given rise to press reports that international collective bargaining is just around the



DEBA TE : DOGMA TI SM

101

corner or that global union contracts, covering all of General Motors Ford, Fiat or Volkswagen plants in a single country, are imminent . This is a naive and somewhat hasty conclusion . . . . (whilst insisting that) our main route at present involves greater co-ordinated action and co-operation among unions in resolving problems of labor organisation vis-a-vis the multinational automotive companies ." During the auto-slump of 1974 the UAW changed course and sought a highly protectionist Bill on U .S . car imports, and moved to virtual support of the AFL/CIO's restrictive Burke/Hartke Bill . Leonard Woodcock, president of UAW, who had favoured a shift of part of the German VW plant to the U .S ., visited Germany to announce a change in policy and to contend that it was the responsibility of the German Government to look after its workers . One interesting survey of opinion on international solidarity action by UAW workers showed that, whilst they would take solidarity action in disputes of other U .S . workers, their support for Canadians in a similar situation was somewhat lower, and that for Mexican workers in their own industry lower again . Support for overseas workers in GM or Chrysler plants fell away even further . In the electronics and telecommunications industries, where transnational bargaining through the WCC's and ITS's was thought to have excellent prospects, the IMF and ICE found absolute resistance from TNC's such as General Electric, ITT, RCA, Siemens, Ericssons, Philips, Brown Boveri and others . They were prepared to bargain regionally and, if forced to it, even nationally, but never internationally . Professors Rowan and Northrup (2) report : "The IMF has thus established the union organisational framework for bargaining on a multinational basis in the electrical industry as it has in automobiles . It has made what it regards as the initial step towards bargaining - limited contact between its multinational union committees and major companies . Will these admittedly `symbolic' meetings grow into something more substantial? Perhaps, but it does not appear that any such developments are on the near horizon ." (Rowan & Northrup 1975) . The European Metalworkers' Federation, as opposed to the International Metalworkers Federation, "has an avowed aim of pushing key European multinational concerns towards multibargaining arrangements . Among these are Philips and Fokker-VFW" . The EMF usually dissociates its activities from those of the ITS system . Four meetings have been reported between EMF and Philips since 1967 . EMF proposed a fifth meeting in 1973 where the issue of multibargaining was the main one . In preparation for a real showdown with Philips, the EMF allowed IMF to attend as an observer . Notwithstanding this, the Philips management objected to IMF attendance as a violation of their code of not dealing with matters on a trans-national basis, and emphasised that the management would move only to national collective



102

CAPITAL AND CLASS

bargaining. The Philips-EMF relationship seems to be stalled at a crossroads, and could only take an internationalist road through an intensification of the class struggle in some of the EEC states and its internationalisation in a dispute in the Philips-EMF plants in Europe and through its TNC empire . But given the increasing complexity of the international division of labour and increasing inequalities in th : development of centres of national capital, perhaps the vision of international bargaining has no divine inspiration at all . Perhaps the focus should be increasingly on regional bargaining . But national and regional bargaining bring the trade union leaders back to the real spectre that haunts them - intensification of the class struggle within its organically structured boundaries . Containment of the class struggle regionally, nationally and internationally has many expressions, not all as brutal as that by George Meaney, President of the ALF/CIO when, on one occasion, he was addressing a U .S . Senate Committee : "You can't dictate to a country from any angle at all unless you control their means of production . If you don't control their means of production, you can't dictate . Whether you control them through ideological methods or control them by brute force, you must control them ." (NACLA 1974) . But with the degeneration of the international trade union movement of the "West" so far, there may be no real answer to Meaney, or to the problems raised by O . & S ., from within the trade union structures . Perhaps the union movements today have only the capacity to open dialogues with the trans-national corporations through organs like the WCC's and continue the dialogues while the workers are accommodated to the new needs as seen, finally, by both parties in the dialogues - the ITS's and the TNC's . But the internationalisation of the class struggle is a quite different kettle of fish . It may be dependent on the forces generated in the areas where the international divisions and redivisions of labour throw up the most acute contradictions . And those areas are outside of Europe and North America and within the Movement of Non-Aligned Nations . It was interesting that at the Colombo 5th Conference of the Non-Aligned Nations Movement in 1976, some of these possible future contours of the international class struggle came to the fore . Some national leaders focussed directly onto the auto industry and Africa and how an inter-continental class struggle might develop within the New World Economic Order on which the Non-Aligned and some of the 105 assembling in Colombo made some progress . Coincidently, in a Melbourne conference on National Independence and Non-Alignment for Australia held only a month after the Colombo conference, this question of the trade unions in the advanced capitalist countries concerting their struggles with the Non-Aligned Nations Movement presented itself as a form, perhaps the necessary form, of an internationalised class struggle . Some saw this as awaiting only the interest and participation of unionists in the advanced capitalist countries .



DEBA TE:DOGMA TISM

103

In many details I may appear to disagree with Olle and Schbller, but our points of identity are the ones of real importance . I am indebted to them for having opened the issues in Capital & Class.

NOTES The writer is retired General Secretary, Australian Union of Postal Clerks and Telegraphists and officer of the Australian Council of Salaried and Professional Associations . 1 Further details on this basis of a projected supranational intelligence system (CIA-NSA) and their related secret pacts, appears in a forthcoming study on The CIA over Australian Labour (J .S . Baker and A .Cavanagh) . Professor Richard Rowan and Herbert Northrup of the Wharton 2 School, University of Pensylvania, would be to U .S ., Australian and many other students of unionism, the leading ideologues of the WCC system . Their University finances a 'global assembling' of all available information on the TNC's, the ITS's and the WCC's . Visiting Professors and reporters from many places keep them supplied with a large volume of intelligence material . Whilst I may be doing these analysts an injustice, the work of their School appears to be aligned with that of the Hoover Institute at Stanford on any and every movement in world communism and the production of the "Yearbook on International Communist Affairs" ; also that of the Centre for Strategic Studies at Georgetown University, which develops a vast study and intelligence apparatus on behalf of the U .S . TNC's . The starting point of its Director, Admiral Arleigh Burke, is that other nations have a responsibility to the United States! Any democratic freedoms are part of a 'responsible democracy' conceded by the U .S .!

BIBLIOGRAPHY NACLA 1974, North American Congress on Latin America, Report, VIII/6. Rowan, R . and Northrup, H .R., 1975, Multinational Bargaining in Metals and Electrical Industries : Approaches and Prospects, journal of Industrial Relations (Australia), March, Vol 17 No .1, p .1 - .



104

CAPITAL AND CLASS

TO : CAPITAL AND CLASS EDITORIAL COMMITTEE Dear Comrades, Can I raise, fraternally, a small question of presentation in Issue No .2? Simon Clarke and others use 'dogmatism' to characterise schools of thought. I think this is unscientific and will obstruct the growth of CSE on non-sectarian lines. The authors are in fact describing Stalinism . There are two reasons, I suspect, behind shunning the word itself . The first it that it has been used as a term of abuse and devalued by sectarian slanging . However, it has the merit that it is a materialist, analytic category . It describes a real relation between the world and a body of thought . It is widely accepted as defining an ideological trend which was initially developed and propagated by the Third International and Russian Communists during Stalin's regime . Some consider it a legitimate theoretical trend . Others stress its apologetic character . But all agree it is a well-understood, objective term . After all 'fascism' has become similarly devalued . Should we shun it? Would the word 'Stalinism' discourage participation by CP members? I think not . Those who accept the Stalin tradition have a right to see it called by its name . Those who are honestly critical of it have the right to distance themselves from it, which they cannot do unless it is called by

its name . The only ones repelled would be those who want to brush their heritage under the carpet . In this instance I don't think we help things by trying not to tread on corns . A major obstacle to the progress of Marxism is the refusal of critical trends in the CPs to discuss their heritage from a materialist standpoint . Nonsectarian debate with any current is only really possible to the extent that it confronts its own origins critically . The second possible reason for avoiding 'Stalinism' is Simon's desire to hunt the high seas of the socialist past for even bigger counter-revolutionary fish than the Stalinist leviathan . He writes : "The 'technicist' view of production was adopted by Stalin, through Plekhanov and Menshevism, from the classical revisionist theories of the Second International ." Dogmatism is a vast creature, stretching back through socialist history, the father and mother of all revisionism . We shall lance this Moby Dick in its technicist underbelly with the harpoon of the Capital-labour relation . Now there are many, many weaknesses in the Marxism of the Second International . But it is not true that either (i) all such errors were revisionist, or (ii) classical revisionism was a causal factor in the construction of Stalin's own peculiar school of thought . Yet Simon, in my view, fosters both these misconceptions . He borders, in fact, on an idealist approach to the development of theory . Accidental or superficial similarities in bodies of theory do not offer a materialist criterion for grouping or assessing them . The nub of the problem is this : who is, and who is not, a dogmatist? If a dogmatist is just someone who follows the tradition of Stalin, then let us say so . If it is someone in a larger tradition, which includes Plekhanov, Menshevism and the 'classical revisionism of the Second International', then we are all dogmatists .



DEBA TE: DOGMATISM

105

However, if a dogmatist is simply someone who expresses certain wrong ideas in common with certain other people, then why the genealogy? Why not say "here is a list of wrong ideas in Poulantzas ; here is another list of people with the same wrong ideas"? The two lists express no important historical fact . There is no causal link . Not only every revisionist school, but every advance in Marxism, must rest on past theory and must also break with it . The theory it rests on may have sprung from the pen of the greatest scoundrels in history, or its most celebrated heroes . Thus for the record, Lenin held that "the only Marxist in the international Social Democratic movement to criticise the incredible platitudes of the revisionists from the standpoint of consistent dialectical materialism was Plekhanov" (Marxism and Revisionism, 1908, CW Vol .] 5) . Has Lenin been swallowed by the Great White Russian Revisionist Whale? Unfortunately, to erect his schema Simon regrettably commits a genuine act of apologetics, and so does his own cause a disservice . He denies by omission the scientific contribution of Trotskyism . He writes : "The independence of [those Marxists who were not prepared to subordinate themselves to dogmatism] was maintained by its diversion of interest from political and economic concerns . It was dominated by the attempts to explain the apparent solidity of bourgeois domination by reference to specific superstructural features which varied from one country to another, thus constituting various schools of 'Western Marxism' . ( . . .) the 'Marxist' alternatives to dogmatism systematically avoided the fundamental theoretical issues which would have been raised by any direct challenge to dogmatism" . Now, if Trotskyism is outside dogmatism, which is suggested by the definition of dogmatism as the 'official Marxism of the orthodox Communist Parties', then this is a strange account . Trotskyism has its faults . But "diversion of interest from political and economic concerns" is hardly one of them . Most detractors maintain it never talks about anything else . And the last thing Trotskyism has done is "explain bourgeois domination by reference to specific superstructual features which varied from one country to another" . Again it is Trotskyism's opponents who characterise it by rigid and dogmatic adherence to international schemata of political and economic development . These facts do not show that Trotskyism is good, bad or ugly . But they do show it is absent from Simon's schema . His picture of trends in world Marxism omits Marxism's only world trend . But here is the problem : Stalinism cannot be assessed without reference to its birthplace in the gigantic faction fight of 1926-36, in which were concentrated all the contradictions of the world's first proletarian revolution, crystallising in the struggle between Trotskyism and Stalinism . In short, Simon has forgotten what the Stalinists least like to remember . It is he who "avoids the fundamental issues raised by a challenge to dogmatism". The problem is the attempt to trace theoretical descent by inheritance



106

CAPITAL AND CLASS

of ideas rather than community of practice : a method Simon shares with Poulantzas . In place of the real origin of Stalinism in the material contradictions of the Russian revolution, we get a metaphysical origin in the "revisionist theories of the Second International" . Of course different bodies of theory have an inner logical connection . But this does not prove a material or historical connection . This must be sought in the first instance in the sphere of real material social relations in history . The revisionism of the Second International reflected a real community of practice : the class collaboration of the emerging Social Democratic bureaucracy . Stalinism expresses a quite distinct community of practice . It boasts quite distinct theoretical inventions : the theory of socialism in one country, of the 'Third Period', and the Popular Front . None of these figure in classical revisionism or the theoretical armoury of Social Democracy . This brings me to my starting point : the danger of sectarianism . British Marxist circles have a tendency to replace political struggle with theoretical struggle . Behind it all there lurks a search for the Holy Grail of theoretical purity, by which all may be judged . Struggle around theory is of course valid . But I would dispute that the test of any revolutionary can be the theories which she or he espouses . I judge both comrades and class enemies by the political actions they commit, endorse or propose . The danger of any other approach is obvious : if the test of revolutionary worth is what you think or profess, an atmosphere emerges in which knowledge cannot advance for fear of holding wrong ideas . I do not accuse Simon of fostering this atmosphere . But I think his language leaves open a chink through which the vapours of sectarianism can rise . Because 'dogmatism' is an idealist category, it leads to definition by ideas . The only way to define a dogmatist, when it comes to it, is by dogmatic thought. In practice, everyone will be a private judge and jury . Materialist categories will not end debates . They can place debates on an objective foundation : the best way to avoid sectarian slanging, and to get good results at the end . In comradeship, Alan Freeman

ARCHIVE ABSTRACT LABOUR AND VALUE I N MARX'S SYSTEM I .I . Rubin We publish below an English translation of a lecture given at a meeting of the General Economics Section of the Institute for Economics, Moscow, by I .I . Rubin in May and June 1927 . The corrected transcription was first published in Under the Banner of Marxism . This translation has been made from the German, I .I . Rubin, S .A . Bessonov et al : Dialektik der Kategorien : Debatte in der UdSSR (1927-29) (VSA, West Berlin, 1975) . The lecture develops one of the main themes of Rubin's Essays on Marx's Theory of Value, thus providing a useful introduction to the latter work, while developing beyond it in important respects . The lecture aims to bring out more clearly than had the Essays the distinction between the social commensurability of labour that is characteristic of any society that is based on the division of labour, and the specific form in which this commensuration is achieved in capitalist society, the form of abstract labour. The lecture thus centres on a further investigation of the concept of the form of value and brings out particularly clearly the significance of the distinctions between value and exchange value, and between abstract and embodied labour, as well as contributing to the debate about the methodology of Capital. At the time the lecture was given these issues were of the greatest political significance, for in the period of the NEP, when Rubin's influence was at its greatest, the question of the applicability of the "law of value" under socialism, and so the proper understanding of the concept of value, was of the utmost importance to the future development of the Soviet Union . Despite the already intensifying repression it was therefgre still possible to examine the foundations of Marxism . With the abandonment of the NEP and the switch to the 'general line' such critical examination became increasingly intolerable for the Soviet state as heterodoxy in theory came to be identified with economic sabotage and foreign intervention in the subversion of established state policy . Thus, Rubin, along with other leading Marxist theorists whose works have recently been rediscovered (like Pashukanis and Volosinov) fell victim to the purges of the 1930s .



108

CAPITAL AND CLASS

Isaac Il'ich Rubin was born in 1896 . He joined the Bund in 1904 and became a member of its Central Committee . He later joined the Mensheviks and was elected to their Central Committee in the autumn of 1920 . Trained originally as a lawyer, his handbooks on conciliation and arbitration and on unemployment insurance were published by the Moscow Soviet in 1917-18 . He was imprisoned by the GPU in 1923-4 in the round-up of Mensheviks . On his release he dropped his political work to concentrate on his academic studies and teaching . From 1926-30 he was a Research Associate at the Marx-Engels Institute under Ryazanov, to whom he was very close, becoming one of the most influential interpreters of Marx's work . During this period he published several books : Contemporary Economists in the West ; Classics of Political Economy from the 17th to mid-19th centuries ; A History of Economic Thought; History of Class Struggles; Essays on Marx's Theory of Value, and co-edited Fundamental Problems of Political Economy. Rubin's former association with Menshevism led to his arrest in December 1930 and his inclusion in the Menshevik trial of March 1931, in which he was accused of collaboration with a supposed "Union Bureau of the Central Committee of the RSDRP" financed and directed from abroad . The charges were ludicrous, and the evidence internally contradictory and in a number of instances clearly false . However Rubin, like the other accused, made a full confession of his "guilt", a confession extracted by the use of sustained and extreme torture (see Medvedev, pp .1 32-6) . The trial was designed as an object lesson to the Bolshevik right opposition and to the Trotskyists . Rubin was included in order to incriminate Ryazanov, who was sacked and expelled from the party as a result of Rubin's "confession" "for treason to the Party and direct aid to the Menshevik interventionists" . In anticipation of his trial Rubin was denounced in Bolshevik, 2, 1930 by V . Milyutin and D . Bovilin as follows : "I .I . Rubin is the ideologist and representative of the ideology of the theoreticians of the Second International in Political Economy . . . It is a great error to describe I .I . Rubin as a fighter for orthodox Marxism against the 'social school', or to accept uncritically and with negligible reservations a number of his theoretical assertions as strictly Marxist ." To the article was appended a staterient announcing the complete cessation of press discussion of the views of Rubin and his followers . However it proved necessary to publish further diatribes by Bovilin in Pravda in its issues of 13/1/31 and 7/3/31, the first called "Let us tear out Rubinschina by the roots" . Rubin was sentenced to five years in jail . After three years in solitary confinement he was exiled to Turgai and then to Aktiubinsk, where he worked as a plan economist in the consumer cooperative and continued his own work . Arrested again in 1937 Rubin was transferred from Aktiubinsk and disappeared for good . As a recent Soviet philosopher, Rosenthal, has observed : "The Communist Party has destroyed this tendency, which is quite alien to Marxism, and assisted Soviet philosophers and economists to unmask its essence" (quoted Rosdolsky, p .570n) . We are very pleased to publish Rubin's "mischief" (ibid) here .



ARCHIVE:ABSTRACT LABOUR AND VALUE

109

REFERENCES Haimson, L .H . ; 1974, The Mensheviks, Chicago . Jasny, N ., 1972, Soviet Economists of the Twenties . Cambridge . Medvedev, R ., 1972, Let History judge . London . Rosdolsky, R ., 1977, The Making of Marx's Capital . London . Rubin, 1 .1 ., 1972, Essays on Marx's Theory of Value . Detroit .

I .I . RUBIN : ABSTRACT LABOUR AND VALUE IN MARX'S SYSTEM TRANSLATED BY KATHLEEN GILBERT Comrades, I have chosen abstract labour and value as the theme of my lecture for two reasons : firstly, I know that the question of abstract labour and the form and content of value has been the subject of heated debate in your seminars . Because of this I decided to organise my lecture in such a way that I may deal with the problem of abstract labour in detail, while covering the question of value, its form and content at the same time . The second reason which persuaded me to select this theme is that it is the central problem- of all Marxist theory . We do not term the theory `the labour theory of value' for nothing - the name alone indicates that the main problem of the theory is the question of the reciprocal relationship between labour and value . What is the labour which creates or determines value, and what is the value which is created or determined by labour? That is the main problem of Marxist theory, which I hope to illuminate in my lecture . Before we move to the essential part of the question, I should like to make a few remarks on methodology . By what method do we intend to set about solving this problem? In the Introduction to a Critique of Political Economy (Introduction to the Grundrisse) Marx observed that an economic investigation can be conducted according to two methods : by the transition from the concrete to the abstract, and conversely by movement from the abstract to the concrete . The former, the analytical method, consists in taking a complex concrete phenomenon as the starting point of the investigation, and selecting a single, or several of the most important, characteristics, disregarding the multiplicity of its features, and so making the transition from the more concrete to the more abstract concept, to the simpler, or thinner concept, as Marx says . By further analysis we move on from this concept to an even simpler one, until we have reached the most abstract concepts in the particular science or the particular complex of questions, which interest us . To cite just one example as an illustration of the problematic we are dealing with, I may remind you of the reciprocal relation between the following concepts . The Marxian theory of value builds on the concepts : abstract labour, value, exchange value and money . If we take money, the



110

CAPITAL AND CLASS

most complex and most concrete aspect of these concepts, and by examining the concept of money make the transition to exchange value, as the more general concept underlying money ; if we then move from exchange value to value, and from value to abstract labour, we are moving from the more concrete to the more abstract concept, i .e . we are following the analytical method . But, Marx says, however necessary the use of the analytical method is in the first stage of scientific enquiry, it cannot satisfy us in itself, and it must be complemented by another method . Once we have traced the complex phenomenon back to its basic elements by means of analysis, we have to take the opposite direction and, starting from the most abstract concepts, show how these develop to lead us on to more concrete forms, more concrete concepts . In our case, this progression from the simpler concepts to richer and more complex ones would be the movement from abstract labour to value, from value to exchange value and from exchange value to money . Marx calls this method 'genetic', at one point, because it enables us to follow the genesis and development of complex forms . Elsewhere he terms it the dialectical . I hope we can also agree to describe the first method as the analytical, and the second (which includes both the analytical and the synthetic method) as dialectical . Marx indicates that he considers the dialectical method to be the only one which solves scientific questions satisfactorily . Accordingly, we have to subject the problem which interests us, the question of the relationship between labour and value, to investigation not only by the analytical method, but by the dialectical as well . Marx gives many examples to show in what respect the analytic method is inadequate . I should like to quote three examples here . Concerning the theory of value, Marx says "Political economy has indeed analysed, however incompletely, value and its magnitude, and has discovered what lies beneath these forms . But it has never once asked the question why labour is represented by the value of its product and labour time by the magnitude of that value ." (Capital I, p .80) . In another passage, devoted to the theory of money, Marx says : "In the last decades of the 17th century it had already been shown that money is a commodity, but this step marks only the infancy of the analysis . The difficulty lies, not in comprehending that money is a commodity, but in discovering how, why and by what means a commodity becomes money ." (Capital I p .92) Here, as we see, the dialectical method differs once again from the analytical . Finally, at a further point while discussing religion, Marx repeats the idea which he has stated before, that it is obviously much easier to discover by analysis the core of the curious religious conceptions, than conversely, it is to develop from the actual relations of real life the corresponding forms of those relations . The latter method is the only materialistic and consequently the only scientific one (Capital / p .372 note 3) . Following Marx, we must solve our problem in this way . Our task does not only consist in showing that the value of a product can be attributed to labour . We must also show the converse. We must reveal how people's productive relations find their expression in value .



A RCHI VE : ABSTRACT LA BOUR AND VALUE

111

This is the basic statement of the problem, which must be considered the most methodologically correct from the Marxian standpoint . If we put the question in this way, we take not the concept of value as the starting point of the investigation, but the concept of labour . We define the concept of labour in such a way that the concept of value also follows from it . The requirements of the methodology already give us some indications as to the correct definition of the concept of labour . The concept of labour must be defined in such a way that it comprises all the characteristics of the social organisation of labour, characteristics which give rise to the form of value, which is appropriate to the products of labour . A concept of labour from which the concept of value does not follow, and particularly a concept of labour in the physiological sense, i .e . the concept of labour which lacks all the features which are characteristic of its social organisation in commodity production, cannot lead to the conclusion which we seek from the Marxian standpoint of the dialectical method . In the following I shall try to show that the difference in conception between the sociological and the physiological understanding of abstract labour can in part be explained precisely by the distinction between the two methods, the dialectical and the analytical . Although the physiological conception of abstract labour can stand its ground more or less successfully from the standpoint of the analytical method, nevertheless it is doomed to failure from the start from the standpoint of the dialectical, since one cannot obtain from the concept of labour in the physiological sense any notion of value as the necessary social form of the product of labour . So we have to define labour in such a way that from it, from labour and its social organisation, we may understand the necessity of value as the basic social form which the products of labour assume in commodity production and the laws of the movement of value . Moving on to the analysis of labour, we will start with the most simple concept, with the concept of concrete or useful labour . Concrete labour is seen by Marx as labour in its useful activity, as labour which creates products which are necessary for the satisfaction of human needs . Labour viewed from this material technical side represents concrete labour . It is obvious that concrete labour does not interest us in the least, so long as we are speaking of the individual, of Robinson Crusoe overcoming nature, since the object of our science is not the production of a single individual, but social production, the production of a whole group of people which is organised on the basis of a specific social division of labour. The system of the social division of labour is the totality of the various concrete kinds of labour, which are unified in a determined system and complement one another materially . So we have made the transition from concrete labour in general to the system of the social division of labour, as the totality of the various concrete kinds of labour . We have to inquire more closely into the concept of the social division



112

CAPITAL AND CLASS

of labour since it plays a key role in the understanding of the whole of Marx's theory of value . Marx says that the system of the social division of labour can occur in two-fold form - as he terms it - as a system which is mediated through exchange and as a system which has no need of such mediation, for example the natural economy of a large clan or of a socialist community etc . We may look first at the system of organised social division of labour which has developed without exchange . So long as one speaks of an organised system of the social division of labour, we have not only concrete material-technical labour, but social labour as well . In Marx, the concept of the social division of labour is on the border between the concept of concrete useful labour, and social labour in social production . On the one hand, at the beginning of the section on the two-fold character of labour (Capital I p .41 f), Marx exaamines the social division of labour as the totality of the concrete modes of labour . Elsewhere in Capital, particularly in the chapter on "Manufacture", (Capital / p .350ff), he examines the system of the social division of labour from the standpoint of the human relations of production which characterise this system . In organised production, the relations among people are relatively simple and transparent . Labour assumes a directly social form, i .e . there is a determined social organisation and determined social organs, which distribute the labour among the individual members of the society, whereby the labour of each person enters directly into social production as concrete labour with all its concrete material characteristics . The labour of each person is social, specifically because it differs from the labour of the other members of the society and represents a material complement to them . Labour is directly social in its concrete form . At the same time it is also divided labour . For the social organisation of labour consists in labour being distributed among the individual members of the society, and conversely the division of labour being the act of a social organ . Labour is both social and divided, and possesses these characteristics in its material technical, concrete or useful form also . Let us now ask this question : is the labour in an organised community also socially equated? Do we find a process which we could describe as a social process of equation of labour in this community? There are various views on this particular problem . Some economists maintain that this kind of social equation of labour already exists in any production community, which is based on the division of labour, and in a form which does not differ in essence from the equation of labour in commodity production . Other economists take the opposite view, saying that the process of social equation of labour is a process which is only appropriate to commodity production and occurs in no other form of production . In particular, these economists deny the possibility and necessity for social equation of labour in a socialist economy . I have suggested a middle road in my book . I pointed out that every production which rests on the division of labour has recourse to social equation of the labour of different kinds and different individuals, to some



ARCHIVE: ABSTRACT LABOUR AND VALUE

113

extent and in one form or another . I also pointed out in connection with this that this equation of labour acquires a very particular social form in commodity production and therefore makes way for the appearance of a completely new category, that of abstract labour . I think that Marx regarded the question in this way, although we have no clear statement by him on the subject . I know of one very explicit observation, which dates already from the first edition of 'Capital'. There he says : "In every social form of labour the labours of the various individuals are related to each other also as human labours but here this relation itself counts as the specifically social form of the labours" (Des Kapital, 1st edition p .238) . We will analyse the end of this sentence at a later point . For the present, I only want to establish that Marx clearly thought that in every social form of labour, the labour of single individuals is related as human labour . It is correct that extreme adherents of the physiological version could maintain that Marx meant here only the physiological equality of the various kinds of labour . But this interpretation seems to me too farfetched . Both the actual sense of the particular sentence, which speaks of the "social form of labour", as well as its relation to many other places in Capital, indicate that Marx meant here the process of social equation of labour . I think it is necessary to add a certain qualification to the formula that social equation of labour occurs in any social form of production . I think that in the ancient family, for instance, where the labour was divided between man and woman and was tied to the representative of each sex, where the change from male labour to female did not exist and was even forbidden, the process of social equation of labour could not take place, even in embryonic form . Further, in social organisations which were based on extreme inequality of the various social strata (e.g . slavery), the social equation of labour could only occur for the members of a specific social group (e .g . for slaves or for a specific category of slaves) . Even the concept of labour as such, as social function, could not be acquired in this kind of society . If we then leave aside social organisation which was based on extreme inequality of the sexes or of individual groups, and turn to a large community with division of labour, e .g . the kind found in the large family associations of the Southern Slavs - I think that here the process of social equation of labour was necessary . It becomes all the more necessary in a large socialist community . But this process of the equation of labour in an organised community differs essentially from the process which occurs in commodity production . Let us actually imagine some socialist community where labour is distributed among the members of the society . A determined social organ equates the labour of different kinds and of different individuals, since without this organ there could be no economic planning . But in a community of this kind the process of equation of labour is secondary and only complementary to the process of socialisation and division of labour . Labour is primarily social and divided . The characteristic of socially equalised labour belongs here as derivative or supplementary . The main characteristic of labour is its social and divided aspect and its socially equated aspect is an additional feature .



114

CAPITAL AND CLASS

I may take this opportunity to say that for the sake of clarity I would find it useful to distinguish between three concepts of equal labour : 1) physiologically equal labour 2) socially equated labour 3) abstract labour, as used by Marx, or preferably, abstract universal labour (a term which Marx uses in the 'Critique') The physiological homogeneity of the various modes of labour existed in all historical epochs, and the possibility that individuals may change over from one occupation to another is the prerequisite for any social division of labour . Socially equated labour is characteristic for all systems with the social division of labour, that is not only for commodity production, but, for instance, for a socialist community . Finally the third concept of labour, as abstract universal, is characteristic only for commodity production . We will come onto this concept . So far we have only discussed the second concept of labour as socially equated and divided . Let us take a look at the changes which will take place in the organisation of labour in our community, if we imagine it not in the form of an organised whole, but in the form of a combination of individual production units of private commodity producers, that is, in the form of commodity production . In commodity production we also find the social characteristics of labour, specified above, which we observed earlier in an organised community . Here too we will find social labour, divided labour and socially equated labour; but all these socialisation processes, processes of equation and division of labour, occur in a totally different form . The interrelation between the three characteristics is now completely different, primarily because in commodity production the direct social organisation of labour is missing, and labour is not directly social . In commodity production, the labour of an individual, a single commodity producer, is not directly regulated by the society, and in itself, in its concrete form, it does not yet belong to social production . Labour only becomes social in commodity production when it assumes the characteristic of socially equated labour ; the labour of every commodity producer only becomes social by virtue of the fact that his product is assimilated with the products of all the other commodity producers, and the labour of a specific individual is thus assimilated with the labour of all the other members of the society and all the others kinds of labour . There is no other characteristic for the definition of the social character of labour in commodity production . There is no previously conceived plan for the socialisation of the division of labour, and the only indication that the labour of a particular individual is included within the social system of production is the exchange of the product of a specific labour for any other product . So in comparison with the socialist community, the characteristics of social labour and of equated labour have exchanged roles in commodity production . Previously, the characteristic labour as equal or equated was the result of the secondary process, of the derived act of a social organ, which socialised and distributed labour . Now labour only becomes social in the form in which it is equated with all other kinds of labour, and becomes thus socially equated labour .



ARCHIVE : ABSTRACT LABOUR AND VALUE

115

I should like to quote a few statements by Marx which should confirm this . The most unequivocal example can be found in the 'Critique' where Marx says that labour only becomes social "by assuming the form of its direct opposite, of abstract universal labour" (p .34), that is, the form of equation with all other kinds of labour . "Abstract and in that form social labour" - Marx frequently characterises the social form of labour in commodity production with these words . I may also call to mind the well known passage from 'Capital' which states that in commodity production "the specific social character of private labour carried on independently, consists in the equality of every kind of that labour by virtue of its being human labour" (Capital/ p .74) . And so in commodity production the emphasis of the social characteristic of labour shifts from the attribute of socialised labour to that of equal or socially equated labour, which only becomes socially equalised labour through the equation of the products of labour . The concept of the equality of labour plays an important role in Marxian value theory precisely because in commodity production labour becomes social only in its quality of being equal labour . Like the characteristic of social labour the characteristic of divided labour also follows from the equality of labour in commodity production . The division of labour in commodity production does not consist in its conscious distribution corresponding to determined, previously expressed needs, but is regulated by the principle of the equal advantage of production . The division of labour between individual branches of production takes place in such a way that in all branches of production, the commodity producers receive an equal sum of value through expenditure of an equal quantity of labour . We established the three characteristics of labour as being social labour, socially equated labour and divided labour . All these characteristics also appertain to labour in a socialist society, but completely change their character and their interrelationship as compared with commodity production . The three characteristics of labour which we listed here are the basis from which the three aspects of value develop . Marx considers value as the unity of the form of value, the substance of value and the magnitude of value . "The crucially important task however was to discover the inner necessary interrelationship between the form of value, the substance of value and the magnitude of value" (Kapital 1st ed.p .240) . The unity of the form, substance and magnitude of value reflects the unity of labour as social, socially equated and quantitatively divided . In commodity production, the relations of labour and of production are "objectified" and the social characteristics of labour assume the form of "objectified" attributes of the product of labour . The "form of value" iss the social form of the product of labour, which reflects the particular social character of labour in commodity production . "The substance of value" represents socially equal labour. And finally the "magnitude of value" is the expression of the social division of labour, or more precisely of the quantitive side of the process of division of labour . The threefold character of labour, which we have suggested, helps us to explain the relationship which exists in the Marxian system between form, substance, and magnitude of value . In particular this division clarifies

116

CAPITAL AND CLASS

some problems of the construction of Marx's section on the 'Fetishism of Commodities' . Allow me to read out this section from the second paragraph : "For, in the first place, however varied the useful kinds of labour or productive activities, may be, it is a physiological fact that they are functions of the human organisation, and each such function, whatever may be its nature or form, is essentially the expenditure of human brain, nerves, muscles etc . Secondly, with regard to that which forms the ground-work for the quantitative determination of value, namely, the duration of that expenditure or the quantity of labour, it is quite clear that there is a palpable difference between its quantity and quality . Lastly, from the moment that men in any way work for one another, their labour assumes a social form" (Capital I p .71) . In the three points quoted, Marx indicates that we can observe the three characteristics of labour, social, equal and quantitatively divided, not only in commodity production, but also in other forms of production . But, says Marx, "whence, then arises the enigmatical character of the product of labour, so soon as it assumes the form of commodities?" And he answers himself : obviously precisely from the form of commodities, in which the three characteristics of labour are already transformed, "reified", in the value of the products of labour . "The equality of all sorts of human labour is expressed objectively by their products all being equally values ; the measure of the expenditure of human labour power by the duration of that expenditure, takes the form of the quantity of value of the products of labour ; and finally, the mutual relations of the producers, within which the social character affirms itself, takes the form of a social relation between the products ." (Capital/ p .72) In these three points Marx already speaks of the substance, the magnitude and the form of value . His reasoning can be traced particularly clearly in the first edition of 'Capital', where the three sentences quoted are immediatley followed by a whole page on the substance, magnitude and form of value . In the second edition the comments referring to the substance, magnitude and form of value are apparently omitted by Marx . In reality they were only deferred . The three paragraphs which precede the analysis of the various forms of production (Robinson's production, medieval production etc .) are devoted to the substance, the magnitude and the form of value (1) . We have now reached the conclusion that equal labour can mean firstly physiologically equal labour, which we have only briefly considered ; secondly it can signify socially equated labour, and this kind of labour exists not only in commodity production, but also, let us say in a socialist community or another large community which is based on the social division of labour ; and finally there is abstract universal labour, that is, socially equated labour in the specific form appropriate to commodity production, labour which becomes social and divided only by the process of social equation . Only this socially equated labour can be described as abstract or abstract-universal . We should mention here that Marx makes several allusions to the three kinds of equation of labour in the 'Critique of Political Economy', that is to physiological, social equalisation in general



ARCHIVE : ABSTRACT LABOUR AND VALUE

117

and social equalisation in commodity production . Marx does not draw any absolutely clear distinction it is true, but we should point out that he does distinguish three terms : human labour, equal and abstract universal labour . I would not maintain that these three terms coincide with those which we characterised earlier as physiologically equal labour, socially equalised and abstract labour, but there are some points of contact nevertheless . In dealing with the problem of abstract labour, we cannot therefore stop at the preliminary characteristic of labour as physiologically equal, nor the characteristic of labour as socially equated . We have to make the transition from both these characteristics to a third, and investigate that specific form of equated labour which is peculiar to commodity production, that is, the system of the social division of labour based on exchange . Consequently, not only are the followers of the physiological conception of abstract labour mistaken in our opinion, but also those comrades who understand abstract labour in general to mean socially equated labour independent of the specific social form in which this equation occurs . We must add, that the two concepts of labour, physiologically equated and socially equated, are frequently confused, and not distinguished from one another sufficiently clearly . The concept of abstract universal labour naturally implies the physiological equality and the social equation of labour, but apart from these it also contains the social equation of labour in the quite specific form which it takes in commodity production . We could give many quotations from Marx himself to show how he is crudely misconstrued by the followers of the physiological conception of abstract labour. I should like to read just one very characteristic quotation here . In his short sketch of Franklin's views Marx says that Franklin unconsciously reduced all the forms of labour to one aspect, being uninterested in whether the labour was that of a shoemaker, a tailor, etc . Franklin believed that value is determined "by abstract labour, which has no particular quality and can thus be measured only in terms of quantity ." Franklin recognised abstract labour . "But", Marx added, "since he does not explain that the labour expressed in exchange value is abstract universal social labour, which is brought about by the universal alienation of individual labour, he is bound to mistake money for the direct embodiment of this aliented labour" . (Critique p .56-57) . It is obvious here that Marx is contrasting abstract labour with abstract universal labour . The abstract universal labour which is embodied in value is the labour which is specifically appropriate to commodity production . We now reach the conclusion : that if we analyse the problem of the relation between labour and value from the standpoint of the dialectical method as well as the analytical, then we must take the concept of labour as the starting point and develop the concept of value from it . If we follow the analytical method, start out from value and ask ourselves what lies beneath this concept, we can certainly say that physiologically equal labour and socially equated labour are concealed beneath the value of products . But neither answer will be adequate, since there is no way to make the transition from physiologically equal labour or from socially equated labour to value. In order to arrive at the concept of value dialectically from the concept



118

CAPITAL AND CLASS

of labour, we must also include in the concept of labour those features which characterise the social organisation of labour in commodity production and necessitate the appearance of value as the particular social form of the product of labour . Consequently this concept of abstract universal labour must be far richer than both the concept of the physiological equality of labour and the concept of the social equation of labour in general . II We moved from physiologically equal labour to socially equated labour, and from socially equated to abstract universal labour . We enriched our definition of labour by new characteristics in the three stages of our investigation and only when we moved on to the third stage and defined labour as abstract universal, from which the category of value must necessarily follow, was it possible for us to move from labour to value . We could define abstract labour approximately as follows : Abstract labour is the designation for that part of the total social labour which was equalised in the process of social division of labour through the equation of the products of labour on the market . In my book 'Essays on Marx's Theory of Value' I gave more or less this definition . I think it is necessary to add that the social nature of abstract labour is not limited by the fact that the concept of value necessarily follows from this concept . As I have already outlined in my book, the concept of abstract labour leads unconditionally to the concept of money also, and from the Marxian standpoint that is entirely consistent . In reality we defined abstract labour as labour which was made equal through the all round equation of all the products of labour, but the equation of all the products of labour is not possible except through the assimilation of each one of them with a universal equivalent . Consequently the product of abstract labour has the ability to be assimilated with all the other products only in the form that it appears as universal equivalent or can potentially be exchanged for a universal equivalent . One can see particuarly clearly in the 'Critique of Political Economy' that the concept of abstract labour is inseparably tied to that of the universal equivalent for Marx . There Marx approaches the study of abstract labour as follows . As in 'Capital', he starts out from the commodity or value, and uncovers analytically the abstract universal labour which lies beneath value (Studienausgabe p .235) . After he has moved by analysis from the equality of values to the equality of labour he goes on to a detailed sociological characterisation of this equal labour, of the "social categories of labour", "social . . . in the particular sense" which is appropriate to commodity production . (Critique p .31) . In commodity production the social character of labour is expressed by "the labour of the individual assuming the abstract form of universal labour, or his product assuming the form of universal equivalent (Critique p .33-34) . "The universal labour-time finds its expression in a universal product, a universal equivalent ." (Critique p .32) . "The labour of



ARCHI VE : ABSTRACT LABOUR AND VALUE

an individual can produce exchange value only if it produces

119 universal

equivalents" (Critique p .32) . As we can see, Marx links the category of abstract labour inseparably with the concept of the universal equivalent, or money . We therefore have to carry the social characterisation of abstract labour still further and deeper, and not confine ourselves to the assimilation of labour through the equation of its products . We must add that labour becomes abstract through being assimilated with a particular form of labour, or through theassimilation of its product with a universal equivalent, which was therefore regarded by Marx as the objectification or materialisation of abstract labour . From this standpoint, an interesting parallel between Marx and Hegel opens up here . The term 'abstract universal' itself, as we know, is reminiscent of Hegel, who distinguishes the abstract universal from the concrete universal . The distinction between the two can be reduced to the fact that the concrete universal does not exclude the differences between the objects which are included within this universal aspect, while the abstract universal excludes such differences . In order to understand why Marx describes the equated labour of commodity producers as the abstract universal, we have to compare the process of equation of labour in a socialist community with the process of equation of labour in commodity production . We will notice the following distinction . Let us assume that some organ compares the various kinds of labour one with another in a socialist community . What happens here? This organ takes all these kinds of labour in their concrete useful form, since it links them in precisely this form, but it abstracts one of their aspects and says that these kinds of labour are equal to each other in the given circumstances. In this case the equality appears as a characteristic of these concrete kinds of labour, as a characteristic which was abstracted from these forms ; but this universal category of equality does not destroy their concrete difference, which manifests itself as useful labour . In commodity production comparison of this kind is impossible, since there is no organ which consciously equates all these kinds of labour . The labour of a spinner and that of a weaver cannot be equated, so long as they are concrete useful labour . Their equation results only indirectly through the assimilation of each with the third form of labour, namely

'abstract universal' labour (cf.

Critique) .

This determined kind of labour

is 'abstract universal' (and not concrete universal) precisely because it does not include the distinctions between the various concrete kinds of labour but precludes these divergences : this kind poses all the concrete kinds of labour in that it appears as their representative. The fact that in this case Marx intended the distinction between the abstract universal and the concrete universal, which occurs in Hegel, can be seen clearly in the first edition of 'Capital' where in general the traces of Hegelian concepts and Hegelian terminology stand out far more distinctly than in the second . Here there is a paragraph which reads : "Within the value-relation and the value expression included in it, the abstractly general accounts not as a property of the concrete, sensibly real ; but on the contrary the sensibly-concrete counts as the mere



120

CAPITAL AND CLASS form of appearance or definite form of realisation of the abstractly general . . . This inversion, by which the sensibly-concrete counts only as the form of appearance of the abstractly general and not, on the contrary, the abstractly general as property of the concrete, characterises the expression of value . At the same time, it makes understanding it difficult ." (The Value Form, pp .] 39-140) .

At another point Marx says : "It is as if together with and besides lions, tigers, hares and all the other real animals, which as a group form the various genuses, species, subspecies, families etc of the animal kingdom, there also existed the Animal, the individual incarnation of the whole animal kingdom ." (Kapital 1st ed . p .234) . To decipher this statement by Marx, we must say that in commodity production the abstract universal really appears not as characteristic or attribute of the concrete, the sensuous-real (i .e . of the concrete modes of labour), since in order to abstract the specific universal features from these concrete modes of labour, it would need a unified organ, which does not exist in commodity production . The concrete kinds of labour are therefore not assimilated one with another through abstraction of some universal characteristics, but through comparison and equation of each of these kinds with a particular determined concrete kind which serves as phenomenal form of universal labour . In order that concrete labour becomes universal, universal labour must appear in the form of concrete labour, "if the individual's labour time represents universal labour time, or if universal labour time represents individual labour time" (Critique p .32) . It is only in the light of these comments by Marx, which show clear traces of Hegel's influence, that we can understand the passages from the Critique which we mentioned earlier, in which Marx says that labour only becomes social in commodity production by assuming the form of abstract universality . This idea is generally, related to Marx's views on bourgeois society . In his earlier works, in the 'German Ideology' for example, he expresses the idea that in bourgeois society, where a central social organisation of production is lacking, the representation of the social interest always falls to some single organisation, to a group of people, to a single class . This single social class declares its partial interests to be the interests of the whole society and lends its ideas 'the form of universality' . The particular interest is expressed as the general interest and the general as the dominant (German Ideology, Marx/Engles Collected Works Nol . V p .60) . If we compare these remarks by Marx in the Critique with those statements where he says that social labour assumes "the abstract form of universality" and that the value of a commodity assumes the form of a particular determined commodity, the form of money, then the close ideal relationship of these concepts becomes evident . To conclude the problem of abstract labour, I must take up two criticisms, which have been made against me, in the article by Daschkowski (2), and by various other comrades .



ARCHIVE : ABSTRACT LABOUR AND VALUE

121

The first criticism was that I apparently seek to substitute for abstract labour the process of abstraction from the concrete characteristic attributes of labour, that is, that I seek to replace abstract labour with the social form of the organisation of labour . Admittedly, a substitution of this kind, if it had really occurred, would deviate from Marxist theory . But we maintain that the character of people's relations of production in commodity production unconditionally means that labour, both in its qualitative and its quantitative aspect, finds its expression in value and in the magnitude of value of a commodity . If instead of abstract labour we take only the social form of the organisation of labour, it would only help us to explain the 'form of value', i .e . the social form, which a product of labour assumes . We could also explain why a product of labour assumes the form of a commodity which possesses a value . But we would not know why this product assumes this given quantitatively determined value in particular . In order to explain value as the unity of the form of value, the substance of value and the magnitude of value, we have to start out from abstract labour, which is not only social, and socially equated but also quantitatively divided . One can find formulations in Marx himself, which, if one chose, would be sufficient reason to say that Marx substituted the social form . of labour for labour itself. Since it would be tedious to refer to the various points in Marx, I should just like to mention one passage which, if written by anyone but Marx, would sound heretical . The sentence runs : "The labour which posits exchange value is a specific social form of labour" (Critique p .36) . In the same place Marx says in a footnote that value is the social form of wealth . If one combines these two statements, then instead of the thesis that labour creates value, we have the thesis that the social form of labour produces the social form of wealth . Some critic would well say that Marx replaces labour completely with the social form of labour : which Marx obviously did not intend . I should now like to turn to the second criticism . It has been said that my explanations give rise to the impression that abstract labour is only produced in the act of exchange . One could conclude from this that value also is only created in exchange, whereas from Marx's standpoint, value and consequently abstract labour too must already exist in the process of production . This touches on the profound and critical problem of the relations between production and exchange . How can we resolve this difficulty? On the one hand value and abstract labour must already exist in the process of production, and on the other hand Marx says in dozens of places that the process of exchange is the precondition for abstract labour . Allow me to quote a few examples. I should like to come back to Franklin . Marx says : "But since he does not explain that the labour contained in exchange value is abstract universal social labour, which is brought about by the universal alienation of individual labour . . ." etc . (Critique p .56) . Franklin's main mistake consequently was that he disregarded the fact that abstract labour arises from the alienation of individual labour . This is not a question of an isolated comment by Marx . We will show that in the later editions of 'Capital', Marx increasingly stressed the idea that in commodity production only exchange reduces concrete labour to abstract labour .



122

CAPITAL AND CLASS

To return to our earlier comments : "Hence when we bring the products of our labour into relation with each other as values, it is not because we see in these articles the material receptacles of homogeneous human labour . Quite the contrary : whenever, by an exchange we equate as values our different products, by that very act, we also equate as human labour, the different kinds of labour expended upon them ." (Capital I p .74) . In the first edition of 'Capital' this sentence had a completely opposite meaning . Marx wrote : "When we bring our products into relation with each other as values to the extent that we see these articles only as material receptacles of homogenous human labour . . . " etc . (p .242) . In the second edition Marx altered the' sense of this sentence completely, fearing that he would be understood to mean that we consciously assimilate our labour as abstract labour in advance, and he emphasised the aspect that the equation of labour as abstract labour only occurs through the exchange of the products of labour . This is a significant change between the first edition and the second . As you will know, Marx did not confine himself to the second edition of the first volume of 'Capital' . He corrected the text subsequently for the French edition of 1875, and wrote that he was making corrections which he was not able to make in the second German edition . On this basis he assigned to the French edition of 'Capital', an independent scientific value equal to the German original . (cf Capital I p .22) . In the second edition of 'Capital', we find the famous phrase : "The equalisation of the most different kinds of labour can be the result only of anabstraction from their inequalities, or of reducing them to their common denominator viz . expenditure of human labour power or human labour in the abstract" (cf . Kapital p .87) . In the French edition Marx replaces the full stop at the end of this sentence with a comma and adds " . . . and only exchange produces this reduction, by bringing the products of the most diverse kinds of labour into relation with each other on an equal footing" (Le Capital I p .70) . This insertion is highly indicative and shows clearly how far removed Marx was from the physiological conception of abstract labour . How can we reconcile these observations by Marx, of which there are dozens, with the basic thesis that value is created in production? This should not be too difficult . The point is that the comrades who discussed the problem of the relationship between exchange and production did not in my view distinguish sufficiently clearly between the two concepts of exchange . We have to distinguish exchange as social form of the reproduction process from exchange as a particular phase of this reproduction process, which alternates with the phase of direct production . At first glance, exchange seems to be a separate phase in the process of reproduction . We can see that a process first takes place in direct production and is then followed by the phase of exchange . Here, exchange is separate from production, and counterposed to it . But exchange is not only a separate phase in the process of reproduction, it stamps the whole process of reproduction with its specific mark and represents a particular



ARCHIVE : ABSTRACT LABOUR AND VALUE

123

social form of the social process of production . Production based on private exchange : Marx frequently characterised commodity production with these words . To make this point clearer, I will quote Marx's words from the third volume of the 'Theories of Surplus Value' that "Exchange of products as commodities is a method of exchanging labour, [it demonstrates] the dependence of the labour of each upon the labour of the others, [and corresponds to] a certain mode of social labour or social production" (Theories of Surplus Value vol .3 p .129) . Here too we find a statement which explains why Marx regarded exchange as a social form of labour : "The whole economic structure of society revolves round the form of labour, in other words, the form in which the worker appropriates his means of subsistence" . (Theories of Surplus Value p .414) . Let us ask now in exactly what form the labourer acquires his means of subsistence in commodity production . We repeatedly find the following answer to this question in Marx : In commodity production the only form of appropriation of products is the form of their alienation and, because the form of the appropriation of products is the form of social labour, so alienation, exchange, is a determined form of social labour which characterises commodity production . If one takes into consideration that exchange is the social form of the production process itself, the form which stamps its mark on the course of the production process itself, then many of Marx's statements become completely clear . When Marx constantly reiterates that abstract labour only results from exchange, he means that it is the result of a given social form of the production process . Labour only takes the form of abstract labour, and the products of labour the form of values, to the extent that the production process assumes the social form of commodity production, i .e . production based on exchange . Thus exchange is the form of the whole production process, or the form of social labour . As soon as exchange really became dominant form of the production process, it also stamped its mark on the phase of direct production . In other words, since today is not the first day of production, since a person produces after he has entered into the act of exchange, and before it also, the process of direct production also assumes determined social characteristics, which correspond to the organisation of commodity production based on exchange. Even when the commodity producer is still in his workshop and has not yet entered into a relationship of exchange with other members of the society, he already feels the pressure of all those people who enter the market as his customers, competitors or people who buy from his competitors, and ultimately pressure from all the members of the society . This link through production and these production relations, which are directly regulated in exchange, continue to be effective even after the specific concrete acts of exchange have ceased . They stamp a clear social mark both on the individual and on his labour and the product of his labour . Already in the very process of direct production itself the producer appears as producer of commodities, his



124

CAPITAL AND CLASS

labour assumes the character of abstract labour and the product assumes the character of value . Here it is necessary to guard against a mistake which is made by many comrades . Many think that because the process of direct production already has a particular social characteristic, the products of labour, and labour in the phase of direct production, must also possess precisely these social characteristics which they possess in the phase of exchange . Such an assumption is totally false, even though both phases (production and exchange) are closely connected to each other, nevertheless the phase of production does not become the phase of exchange . There is not only a certain similarity between the two phases, there is still a certain distinction too . In other words, on the one hand, we recognise that from the moment when exchange becomes the dominant form of social labour, and people produce specifically for exchange, that is in the phase of direct production, the character of products of labour can already be regarded as values . But the characteristic of the products of labour as values is not yet that which they assume when they are in fact exchanged for money, when, in Marx's terms, the 'ideal' value has been transformed into 'real' value and the social form of the commodity is replaced by the social form of money . The same is also true of labour . We know that commodity owners in their acts of production take the state of the market and of demand into account during the process of direct production, and from the start produce exclusively in order to transform their product into money and thus also transform their private and concrete labour into social and abstract labour . But this inclusion of the labour of the individual in the labour mechanism of the whole society is only preliminary and tentative . It is still subject to a strict test in the process of exchange which can give positive or negative results for a particualr commodity producer . Thus the labour activity of the commodity producers in the phase of production is directly private and concrete labour and only indirectly or latently, as Marx puts it, social labour . Thus when we read Marx's work, and particularly his descriptions of the way in which exchange influences value and abstract labour, we must always ask what Marx had in mind in a particular case - exchange as a form of the production process itself, or exchange as a separate phase counterposed to the phase of production . In so far as exchange as a form of the production process is concerned, Marx distinctly says that without exchange there is neither abstract labour nor value, that labour only assumes the character of abstract labour with the development of exchange . Marx's views are quite clear and I have developed them in my book . Where Marx refers to exchange as a separate phase counterposed to the phase of production, he says that labour and the product of labour possess a determined social character even before the process of exchange, but that this character must yet be realised in the process of exchange. In the process of direct production labour is not yet abstract labour in the full sense of the word, but has still to become abstract labour . Numerous statements to this effect can be found in Marx's work . I should like to quote just two passages from the 'Critique' .



ARCHIVE : ABSTRACT LABOUR AND VALUE

125

"But the different kinds of individual labour represented in these particular use-values, in fact, become labour in general, and in this way social labour, only by actually being exchanged for one another in quantities which are proportional to the labour-time contained in them" (Critique p .45) . Elsewhere Marx writes : "Commodities now confront one another in a dual form, really as usevalues, and nominally as exchange values . They represent now for one another the dual form of labour contained in them since the particular concrete labour actually exists as their use-value, while universal abstract labour time assumes an imaginary existence in their price . . ." (Critique p .68) . Marx maintains that commodities and money do not lose their differences because of the fact that every commodity must unconditionally be transformed into money . Each is in reality what the other is ideally, and ideally what the other is in reality . All Marx's writing on this show that we must not approach this problem too linearly . We should not think that because commodity producers are already linked to one another by determined social relations in the process of direct production, therefore their products and their labour already possess a directly social character . The labour of a commodity producer is directly private and concrete labour, but together with this it acquires an additional 'ideal' or 'latent' characteristic as abstract universal and social labour . Marx was always amused by the Utopians who dreamed of the disappearance of money and believed in the dogma that "the private labour of a private individual contained in (a commodity) is immediately social labour" (Critique p .86) . We thus come to these conclusions : Abstract labour and value are created or "come about", "become" in the process of direct production (Marx used the expression "werden" more frequently for this process) and are only realised in the process of exchange . III We have spoken up till now of abstract labour . I should now like to move on to value . Our task is the same in regard to the problem of value as it was with abstract labour . I tried to show that within the concept of abstract labour we must also include the characteristic of the social organisation of labour in commodity production . In the same way I should like to show that within the concept of value we must necessarily include the social form of value, the social form which the products of labour assume in commodity production . The task which lies before us is to introduce social form into the concept of abstract labour and the concept of value . How is value usually defined, as distinct from exchange value? If we take the most popular and widespread conceptions, we can certainly say that value is usually understood as the labour which must



126

CAPITAL AND CLASS

necessarily be expended for the production of a particular commodity . The exchange value of a particular commodity is understood as the other product or other sum of money, for which a particular commodity is exchanged . If a particular table was produced in three hours' labour and is exchanged for three chairs, then one usually says that the value of the table is equal to three hours of labour, and finds its expression in another product, which is different from the table itself, that is, in these three chairs . The three chairs represent the exchange value of the table . In this kind of popular definition it is usually unclear whether the value is determined by the labour or whether the value is the labour itself. Naturally from the standpoint of Marx's theory it is correct to say that value is determined by labour. But then the question arises : What is this value, which is determined by labour? We cannot usually find any adequate answer to this in popular scientific explanations . Hence the readers form the impression that the value of a product is nothing other than the labour which must be expended in its production . The deceptive impression of the complete identity of labour with value is created . This idea is very widespread in anti-Marxist literature . One can say that the majority of the misunderstandings and misinterpretations which we came across in anti-Marxist literature rest on the false assumption that for Marx labour is also value . This false impression frequently arises from the lack of understanding of the terminology and the train of thought in Marx's work ; for instance Marx's famous words that value is 'congealed' or 'crystallised' labour is usually construed to mean that labour is also value . This misconception is fostered by the ambiguity of the Russian verb for 'represent' (darstellen) . Value 'represents labour' . But the Russian translation can be read not only as meaning that value is the representative or expression of labour - the only conception which is consistent with Marx's theory, but also as meaning that value 'is' labour . This idea is very widespread in the critical literature directed against Marx, and is obviously wrong . The critics who interpret Marx's statements that labour constitutes the substance of value to mean the complete identity of the two concepts, do not notice the fact that in this case Marx borrowed Hegel's terminology . Anyone who knows Hegel's 'Logic' with the theory of essence, will remember that Hegel uses various terms when he attempts to clarify the relationship between two objects, one which determines and one which is to be determined . He first says that an object appears as the essence of the other, then he defines it as the ground for the latter object, next he describes it as content as distinct from form, later he regards this same object as substance, as cause and finally he moves on to consider the interrelation between two objects . It is an interesting fact that in Marx's works, the whole scale of expressions which we meet in Hegel can be found, now applied to labour . Labour is also described as the essence of value, and as its ground, its content, its substance and its cause . We have to link all these expressions with the methodological principles on which Hegel's theory is based, and it then becomes clear that Marx's thesis that



ARCHI VE: ABSTRACT LABOUR AND VALUE

127

labour is the substance of value, can in no way be interpreted to mean the complete identity of the two . In my book I advanced this particular thesis in the chapter on the content and form of value . I was mainly trying to show that labour is only substance of value, but does not yet represent value . In other words, when Marx's critics say : 'In Marx's writings the substance of value is labour, consequently labour is value', it must be emphasised that labour is only substance of value, and that in order to obtain value in the full sense of the word we have to add something to labour as the substance of value, namely the social form of value . Only then do we obtain the concept of value in the sense in which it is found in Marx's work . What then does value represent as the unity of the content or substance (i .e . labour) and the form of value? What is this value as distinct from exchange for Marx? To find an answer to this problem we have to ask the question : How does Marx move from exchange value to value? Why does he find it necessary to form a new and more abstract concept of value, in conjunction with exchange value which appears in reality in the act of exchange? You will probably know that Marx had not yet made any clear distinction between exchange value and value in the 'Critique of Political Economy' . In the 'Critique' Marx begins his interpretation with exchange value, and from there passes on to value (which he calls exchange value) . This transition is entirely imperceptible, smooth and apparently self evident . In 'Capital' Marx makes this transition completely differently and it is very interesting to compare the first two pages of the 'Critique' with those of 'Capital' . The first two pages in both books correspond completely ; in both alike the exposition begins with use value and moves on to exchange value . The sentence that exchange value at first sight presents itself as a quantitative relation, as proportion, is found in both books but from then on the texts begin to diverge . While Marx passes imperceptibly from exchange value to value in the 'Critique', in 'Capital' the opposite is the case, as if he intends to linger on this point, foreseeing the objections from his opponents . After the sentence mentioned above, Marx comments : "Hence exchange value appears to be something accidental and purely relative, and consequently an intrinsic value, i .e . an exchange value that is inseparably connected with, inherent in commodities, seems a contradiction in terms" (Capital / p .36) . Let us take a closer look . As we can see, Marx had in mind an adversary who wanted to prove that nothing exists beyond relative values, that the concept of value in political economy is utterly superfluous . Who was this adversary to whom Marx was referring? I would rather not commit myself so precisely, but I assume that this adversary was Bailey, who tried to prove that the concept of value in general is unnecessary in political economy, and that we should confine ourselves to the observation and investigation of particular proportions, in which the various commodities are exchanged . Bailey met with great success with his superficial but witty critique of Ricardo, and attempted to undermine the foundations of the labour theory of value . He maintained



128

CAPITAL AND CLASS

that we cannot speak of the value of a table, but that we can only say that the table is exchanged for three chairs on one occasion, for two pounds of coffee on another occasion etc . The magnitude of the value of the table is purely relative and varies in different cases . From this Bailey drew the conclusion which led him to deny the concept of value where the concept of value differs from the relative value of a particular product in a given act of exchange . Let us imagine the following case : the value of a table is equal to three chairs . After a year this table is exchanged for six chairs . We think we can say that although the exchange value of the table has altered, its value has remained unchanged, only the value of the chairs has fallen to half their former value . Bailey finds this assertion meaningless . If the chairs' relation of exchange to the table has changed, then the table's relation of exchange to the chairs has changed, and the value of the table consists only in this . In order to refute Bailey's theory, Marx thought it necessary to develop the thesis that we cannot understand exchange value unless it is traced back to an underlying unity of value . The first section of the first chapter of 'Capital' is devoted to establishing a basis for this idea, of making the transition from exchange value to value and from value to the unity which lies behind it, to labour . The second section is an extension of the first, in that it simply explains the concept of labour in more detail . We can say that Marx makes the transition from the diversity which is observable in the sphere of exchange values to the underlying unity behind all exchange values, that is to value (and ultimately to labour) . Here Marx demonstrates the incorrectness of Bailey's conception of the possibility of confining our investigation to the sphere of exchange value . In the third section Marx retraces the journey and explains how the unity of value of a specific product is expressed in its various exchange values . Previously Marx had moved from diversity to unity ; now he moves from unity to difference . Earlier he refuted Bailey's theory ; now he supplements Ricardo's theory, in which the transition from value to exchange value was missing . To refute Bailey's theory Marx had to develop Ricardo's theory further . In fact, Bailey's intention of proving that no value exists except exchange value was made easier by the one-sidedness of Ricardo, who could not show why value appears in a determined form of value . Marx was therefore confronted with two tasks : 1) to prove that behind exchange value we have to discover value and 2) to prove that value leads necessarily to different forms of its manifestation, to exchange value . In this present lecture I should like to deal only with the former task, as it is my concern to clarify the concept of value . A complete eludication of the concepts of exchange value and money would take me beyond the confines of my theme . How then does Marx makes the transition from exchange value to value? Critics and commentators on Marx usually suppose that his main argument consists in the well known comparison of corn and iron, on the third page of the first volume of `Capital' (Capital / p .37) . When one equates corn and iron, Marx concludes, then there exists in equal quan-



ARCHIVE: ABSTRACT LABOUR AND VALUE

129

tities something common to both, the two things must be equal to a third and this third thing is their value . This is usually thought to constitute Marx's central argument and the critical blows of his adversaries are usually directed against this argumentation . There is no work hostile to Marx which does not make some reference to Marx's attempt to prove the necessity of the concept of value by a purely abstract analysis . But they completely overlooked this fact : the paragraph which deals with the comparison of corn with iron is no more than a conclusion following on from the previous paragraph, which is usually disregarded, not only by the critics but by commentators on Marx also . The previous paragraph reads : "A given commodity, e .g . a quarter of wheat is exchanged for 20 pounds blacking, 1 .5m silk or Y2oz gold etc ; in short for other commodities in the most different proportions . But the exchange value of the quarter of wheat remains unchanged, and is expressed only in the blacking, the silk and the gold . Consequently the exchange-value must contain something distinguishable from these phenomenal forms ." (Capital / p .37) . Marx worked on this paragraph with care and gave different variations in various editions . We quoted the passage in the Russian translation of the German edition which was edited by K . Kautsky . We can follow the reasoning even more clearly in the second edition of `Capital', where the end of this passage reads : "But since x blacking, y silk or z gold etc . each represent the exchange value of one quarter of wheat, x blacking, y silk, z gold etc . must as exchange values be replaceable by each other or equal to each other . Therefore first : the valid exchange values of a given commodity express something equal ." (Capital l p .37) In other words two commodities which are equal to our given commodity, the wheat, are equal to each other . If we take this conclusion into consideration, as emphasised by Marx in the variations quoted, we can see that the next paragraph follows in logical sequence . If follows from this that one and the same commodity can be expressed in the most different use-values . In the paragraph quoted, Marx comes to the conclusion that two commodities, which are exchanged for one and the same commodity, or are equal to a third, are equal to one another . From this follows also with logical necessity the converse conclusion, which Marx reaches in the next paragraph : if two commodities are equal to one another, then they are equal to a third . It is this thought which Marx expresses in the paragraph where he compares the wheat with the iron . Thus Marx's thesis that two commodities which are equal to one another must also be equal to any third is simply a logical conclusion of the previous thesis, according to which two commodities which are equal to a third, are equal to each other . The true sense of Marx's argumentation consists in the statement of a well known fact about commodity production, the fact that commodities



130

CAPITAL AND CLASS

can be equated with each other and that a specific commodity can be assimilated with infinite numbers of other commodities . In other words, it is the concrete structure of commodity production which forms the starting point of all Marx's reflections and in no way the purely logical comparison of two commodities . Marx thus starts out from the fact of the universal equalisation of all commodities with each other, or from the fact that every commodity can be compared with a vast number of other commodities . Nevertheless this assumption alone is not adequate for all the conclusions Marx draws . There is another tacit assumption underlying these which Marx expressed elsewhere . The second assumption consists in this : we assume that the exchange of a quarter of wheat for any other commodity, is an exchange which is governed by a known regularity (Gesetzmassigkeit), and the regularity of these acts of exchange is due to their dependence on the process of production . We have to reject the notion that the quarter of wheat can be exchanged for any random quantity of iron, coffee etc . We cannot agree with the assumption that the proportions of exchange are laid down each time in the act of exchange itself, and so have a completely accidental character . We maintain that all these possibilities for the exchange of a specific commodity with another, are governed by a determined regularity which is based in the process of production . In this case Marx's whole argumentation takes the following form : Marx says : let us take not the accidental exchange of two commodities wheat, and iron, but exchange in the form in which it actually occurs in commodity production, and then we will see that each object can be universally equated with all other objects ; in other words, we can observe countless numbers of proportions of exchange of a given product with all others . But the proportions of the exchange are not accidental, they are regular, and their regularity is determined by causes which are grounded in the process of production . Thus we reach the conclusion, that independently of the fact that the value of a quarter of wheat is expressed on one occasion as two pounds of coffee, on another as three chairs etc ., the value of a quarter of wheat remains one and the same in all the different cases . If we were to assume that a quarter of wheat has a different value in each of the infinite number of proportions of exchange - and Bailey's assertions amount to this then we would be acknowledging that complete chaos reigns in the phenomenon of price formation, in that sublime phenomenon of exchange of products, through and by means of which a universal inter-relation of all modes of labour is established . We can draw certain conclusions from the train of thought which led Marx from exchange value to value . I came to one conclusion earlier, when I referred to the fact that Marx makes commodity production with its universal equation of all products the starting point of his enquiry, an equation which is closely connected with the course of the production process . Marx does not set out from the contrived example of a random comparison of two commodities, nor from a purely logical analysis of all the characteristics which they may have in common, but from the real



ARCHIVE : ABSTRACT LABOUR AND VALUE

131

form of the exchange of products which is characteristic of commodity production . Our second conclusion comes down to this : when Marx compares wheat with iron, he finds in both something 'common' and in this 'common' factor he recognises the value of the products . In the popular literature, one cannot find a clear answer to the question as to what is the 'common' factor in the exchangeable products to which Marx refers . Sometimes it is correctly seen as value, sometimes though it is identified with labour . If we turn to Marx, we find a clear answer to the question, on the fifth page of 'Capital' : "Therefore, the common substance that manifests itself in the exchange value of commodities, whenever they are exchanged, is their value ." (Capital / p .38) . Marx therefore does not move directly from exchange value to labour . From exchange value he moves to the concept of value and then only by further analysis, from the concept of value to labour . Strictly speaking there are three stages in the chain of reasoning, as it moves from exchange value to value and from value to labour. The conclusion I should like to draw from this, comes down to the fact which we discussed previously, - that the concept of value must be strictly distinguished from the concept of labour, although there is a tendency, particularly in popular interpretations to explain them as identical . But what then is this value, which we obtained by abstraction from the concrete proportions of exchange, in which our quarter of wheat is equated with other products . Although we are now abstracting from those concrete products, for which our quarter of wheat is exchanged, nevertheless we do not abstract from the social form of value, which this quarter of wheat possesses, that is, we hold that our quarter of wheat has the capacity to be exchanged in a determined proportion for any other product which exists in the particular society . Further, we consider the product's capacity for exchange to be its characteristic feature, which is subjected to determined laws, and is in particular closely linked with the conditions of manufacture of a specific product . In other words, no longer does the concept of the social labour necessary for its production alone form part of our concept of the value of wheat . The concept of social labour which assumes 'material form', the form of a particular property of a product-, is also included together with the 'content of value' and the 'form of value' . I should like to give one quotation to show that Marx distinguishes value from labour as the content of value . "Every product of labour is, in all states of society, a use value ; but it is only at a definite historical epoch in a society's development that such a product becomes a commodity, viz ., at the epoch when the labour spent on the production of a useful article becomes expressed as one of the objective qualities of that articles i .e . as its value ." (Capital p .61) . Thus the content of value (i .e . labour) and the social form of value are also



132

CAPITAL AND CLASS

included in the concept of value . What then is this 'form of value' which as distinct from exchange value is a part of the concept of value itself? I should like to give one very clear definition of the form of value from the first edition of 'Capital' : "The social form of the commodity and the form of value or form of exchangeability are therefore one and the same" (Studienausgabe p .235) . As may be seen, the form of value is the description of the form of exchangeability or the social form of the product of labour which contains the capacity to be exchanged for any other commodities, in so far as this capacity is determined by the quantity of labour necessary for the production of a specific commodity . In this way, we did not depart from the social form of the product of labour when we made the transition from exchange value to value . We have only abstracted from that concrete product, in which the value of the commodity is expressed, but we never lost sight of the social form of the product of labour . Our conclusion can also be formulated thus : Marx analyses the 'form of value' separately from exchange value . In order to introduce the social form of the product of labour in the concept of value itself, we were forced to split or divide the social form of the product of labour into two forms : into the form of value and into exchange value, the former meaning the social form of the product which has not yet concretised in a specific object, but represents as it were the abstract character of a commodity . I have also explained this distinction between the form of value and exchange value in my book . There I considered them both as qualitative and quantitative aspect of exchange value, it is true . I did this mainly because in some places in Marx's work, the terms form of value and exchange value are scarcely distinguished from one another . A complete identification of the form of value with the qualitative aspects and of exchange value with the quantitative cannot be regarded as correct, since both concepts must be considered both from their qualitative as well as from their quantitative side . The question does not bear directly on our theme and I will therefore not spend any more time on it . I will simply note that this division of the social form of the product into the form of value and exchange value is extensively dealt with in my book . I had to introduce the characteristics of the social form of the product of labour into the concept of value itself, and thus demonstrate the inadmissability of an identification of the concept of value with the concept of labour, an identification frequently made by popular scientific interpretations of Marx's theory . In other words : I had to demonstrate that value arises not only from the substance of value (i .e . labour) but also from the 'form of value', and in order to introduce the form of value into the concept of value itself, I had to distinguish it from exchange value, which Marx considers separately from value . I had to divide the social form of the product into two parts : into social form, which has not yet acquired a concrete appearance, and into that form which has already acquired a concrete and independent character. Now that the distinction between the form of value and exchange value has been clarified, I should like to turn to the concept of value and



ARCH.'VE:ABSTRACT LABOUR AND VALUE

133

develop the relationship between its various aspects : between the content or substance of value and the form of value . What relation exists between labour and that social form of value with which we have dealt? The general answer to this question runs : the form of value is the adequate and exact form of the expression of what is contained in value (i .e . labour) . In order to explain this idea, we must come back to an earlier example : a table was exchanged for three chairs . We say that this process of exchange is subject to a determined regularity, and dependent on the development of and the alterations in the productivity of labour . But exchange value is a social form of the product, which not only expresses the alterations in the labour, but also conceals and obscures these very changes . It obscures them for the simple reason that exchange value is the relation between two commodities, between the table and the chairs, and therefore the alteration of the proportions of exchange between these two articles gives us no information about whether the labour expended on the making of the table has actually altered . If the table can be exchanged for six chairs after some time has elapsed, then the exchange value of the table has altered, while the value of the table itself may not have changed one iota . In order to examine the process whereby the change in the social form of the product depends on the quantity of labour expended in its making, in its pure form, Marx had to separate the phenomenon as it exists into two parts. He [ -.ad to cut it across and say that we must study separately those causes which determine the value of the table, and those which determine the value of the chairs, and that one and the same phenomenon of exchange (the fact that the table can now be exchanged for six chairs instead of for three) can either be caused by reasons connected with the table, or by reasons stemming from the conditions of the production of the chairs . To examine the activity of each of these causal chains separately, Marx had to split the fact of the change in the exchange value of the table into two parts and assume that these changes are exclusively caused by reasons effective on the side of the table, i .e . through a change in the productivity of the labour necessary for the production of the table . In other words he had to assume that all the other commodities for which our table is exchanged maintained their original value . Only on this assumption does the change in the value of the table follow from the change in the quantity of 1 . .bour necessary for its production, and the social form of labour proves to be a more precise and adequate expression of the content of value or the substance of value (that is of the quantity of labour expended in the process of production .) The determination of value as unity of content (i .e . labour) and social form of value, carries the following advantages . We can break with the widespread identification of value with labour straight away, and so determine the relation of the concept of value to the concept of labour more correctly . On the other hand we can also determine the relation between value and exchange value more correctly . Formerly, when value was regarded simply as labour and had not yet assumed more precise social characteristics, this value was on the one hand identified with labour, and on the other hand separated from exchange value by an abyss . Economists



134

CAPITAL AND CLASS

often saw only labour in the concept of value and could not make the transition from this concept to the concept of exchange value . Now, regarding value as the unity of content and form, we link value through its content with the preceding concept, with labour ; on the other hand though, we link the concept of value through the form of value with what follows, with exchange value . In fact when we maintain that value is not labour in general, but labour which has assumed the form of the exchangeability of the product, then we necessarily have to make the transition from value to exchange value . Thus the concept of value is inseparably linked with, on the one hand, the concept of labour, and, on the other, with the concept of exchange value . But the inseparable connection of all these concepts should not lead to their identification with each other . We regard value as social labour which has assumed the form of an 'objectifed' property of the product of labour, or as the property of the product to be able to be exchanged for any other product, in so far as this property of the product depends on the quantity of social labour necessary for its production . In conclusion I should like to point out that the ability to split the social form of the product into two parts (the form of value and exchange value, the former itself belonging to the concept of value, while exchange value is only a phenomenal form of value) possibly recalls an analogous procedure in Hegel's writing . Although Marx does not refer anywhere to a connection between his concept and Hegel's philosophy, one can find an essential similarity between the division of the social form in Marx's work, and Hegel's theory of the 'doubling of the form' . I should like to quote a few lines from the so-called small 'Logic' by Hegel : "The essential point to keep in mind about the opposition of form and content is that the content is not formless, but has the form in its own self, quite as much as the form is external to it . There is thus a doubling of form . At one time it is reflected into itself ; and then is identical with the content . At another time it is not reflected into itself, and then is the external existence, which does not at all effect the content ." (Hegel's Logic, Clarendon Press, Oxford 1975 p .189) I think the distinction Marx made between the form of value, which is included in value itself, and exchange value, which represents something 'external', 'undetermined' in relation to value, bears some similarity with the doubling of form which we find in Hegel . I now come to the last part of my lecture which concerns the question of the content or the substance of value . All Marxists agree that labour constitutes the content of value, but the problem lies in determining what kind of labour we are speaking of . The previous part of the lecture should have convinced us as to the variety of different concepts which can be concealed in the word 'labour' . What kind of labour therefore constitutes the content of value? Most readers will have taken me to mean that by the content of value I understand labour in its material technical form . I admit that this interpretation is justified since approximately these formulations may be found in my book 'Essays on the Theory of Value' . Nevertheless I



ARCHIVE : ABSTRACT LABOUR AND VALUE

135

must recall that in my book, in the one chapter on the content and form of value, one can find not one, but three formulations which could show that by the content of value I did not mean labour which is studied exclusively from its material technical aspect (3) . There I wrote "Labour as the substance of value is not seen by Marx as a determined quantity of labour, but as something 'independent and absolute', as something accumulated in the product and materially objectified . This labour is examined from the standpoint of the process of division of social labour among the individual branches of production and taken as part of the total social labour in its relation to the latter, as to the whole" . Elsewhere I quoted Marx's words on value as "form, in which the proportional division of labour is expressed ." Lastly, the final conclusion of the chapter reads : "Considered from the qualitative aspect, the relation between labour as 'substance of value' and 'form of value' signifies the relation between the process of division of labour and its specific social, and exchange form" . (3) These references should justify my conclusion that I did not take the content of value to mean labour considered exclusively from the material technical side . Rather my conception approximated to the concept of socially equated and divided labour discussed earlier . But this concept, which can be found in many places in my book, was not adequately explained, and needed important corrections . In the present lecture I have drawn a sharp distinction between socially equated labour in general (which exists not only in commodity production but also, for example, in socialism) and abstract-universal labour as labour which is equated in the specific form appropriate to commodity production . Let us now ask : does Marx understand the content of value to mean socially equated labour in general or abstract universal labour? In other words, when we refer to labour as the content of value, do we include in the concept of labour all those characteristics which we incorporated above in the concept of abstract labour or do we conceive of labour in the sense of socially equated labour which does not incorporate those characteristics which characterise the social organisation of labour in commodity production . Does the concept of labour as the content of value coincide with the concept of abstract labour which constitutes value, or is the character of the former concept broader than this? At first glance one can find arguments in favour of both interpretations of the 'content of value' in Marx's writings . On the one hand one can find arguments which apparently mean that by labour as the content of value, we should understand something more limited than abstract labour, that is, labour without all those social characteristics, which appertain to it in commodity production . What arguments can we find to support this solution to the problem? Marx often meant by content of value something which can not



136

CAPITAL AND CLASS

only acquire the social form of value but another social form too . Content is understood as something which is capable of assuming differing social forms . It is precisely this ability which distinguishes socially equated labour but not abstract labour i .e . labour which has already assumed a definite social form . Socially equated labour can assume the form of labour organised in commodity production, and the form of labour organised in, for instance, a socialist economy . In other words in this case we are conceiving of socially equalised labour in its abstract form and disregarding those modifications which are brought about in the content itself (i .e . labour) by one or other of its forms . Does this concept of the content of value exist in this sense in Marx's work? we can now answer this question positively . Think for example of the passage where Marx says that "exchange value is a definite social manner of expressing the amount of labour bestowed upon an object ." (Capital I p .82) . Labour is clearly being considered as abstract content here, which can assume either one or another social form . When Marx writes in his well known letter to Kugelmann on 11th July 1868 that the social division of labour manifests itself in commodity production in the form of value, he is again regarding socially divided labour as the content, which can assume this or that social form . In the second paragraph of the section on the fetishism of commodities Marx explains directly that we can find the "content of the determining factors of value" not only in commodity production but also for example in a patriarchal family or on a medieval estate . Here, as we see, labour represents a content which can assume various social forms . We may now put forward an argument in favour of the opposing thesis, according to which we have to see abstract labour as the content of value . Firstly, we find a few statements by Marx confirming this, e .g . "(Commodities) relate to abstract human labour as to their common social substance" (Studienausgabe p .235) . This statement leaves no room for doubt that abstract labour is not only a creator of value but also substance of value or content of value . The same conclusion may be reached on the basis of methodological considerations . I demonstrated earlier that in commodity production, socially equalised labour assumes the form of abstract labour, and value as the social form of the products of labour arises necessarily only from this abstract labour . It follows from this that the concept of abstract labour directly preceded the concept of value in our system, and that would show that we must interpret precisely this concept of abstract labour as the basis, content, or substance of value . But one must also not forget that in the question of the relation between content and form Marx took not Kant's but Hegel's standpoint . Kant regarded form as something external in relation to content and as something which joins on to it from the outside . From the standpoint of Hegelian philosophy, content does not represent something which form attaches to from the outside, rather the content itself in its development gives birth to this form, which was contained within this content in concealed form . The form arises necessarily from the content itself.



ARCHIVE : ABSTRACT LABOUR AND VALUE

137

This is the main thesis of Hegelian and Marxist methodology, a thesis which stands in contradiction to Kantian methodology . From this standpoint, the form of value also must arise of necessity from the substance of value, and consequently we must view abstract labour as the substance of value, in all the fullness of its social features which are characteristic for commodity production . Finally for our last argument, we will point out that when we take abstract labour as the content of value, an essential simplification of the whole Marxist system is achieved, since in this case labour as content of value is not distinguished from the labour which creates value . So we have reached the paradoxical conclusion that at one point Marx acknowledges socially equalised labour as the content of value, and at another he acknowledges abstract labour as this content . How can we resolve this contradiction? It seems to me that the contradiction disappears if we remember the distinction between the two methods, the analytical and the dialectical, which I discussed at the beginning of my lecture . If we set out from value as a determined social form, and ask ourselves what is the content of this form, it will become apparent that this form only expresses the fact in general that social labour was expended : value proves to be form, which expresses the fact of the social equation of labour, as a fact which not only occurs in commodity production, but can also occur in other kinds of production . By proceeding analytically from the finished form to its content, we have found socially equated labour as the content of value . But we reach another conclusion, when we take not the finished form as starting point, but the content itself (i .e . labour) from which the form (value) must necessarily arise . In order to make the transition from labour, regarded as content to value, as form, we have to include in the concept of labour the social form of its organisation in commodity production, i .e . recognise abstract universal labour as the content of value . It is possible that the apparent contradiction in the definition of the content of value in Marx's work can be explained precisely by the distinction between these two methods . If we now summarise the interpretation discussed in our lecture, we can say that the following five concepts are the basic concepts on which Marxian theory of value and money rests : (1) the relations of production of the commodity producers, (2) abstract labour, (3) value, (4) exchange value and (5) money . Engels pointed out in his article on Marx's 'Critique of Political Economy', that Marx's contribution consists in showing us the whole system of the bourgeois economy in its inner interrelations (Critique p .226) . Applied to these five categories, Marx's contribution consists in showing the inner inseparable interrelations between all these categories . Unfortunately this interrelation was frequently lost sight of by readers of Marx and these categories were each considered separately . Let us recollect how the relationship between the five categories has usually been envisaged . Let us begin with the relations of production of the commodity producers . This concept was known to all Marxists . It was generally known that the theory of the production relations between people is the



138

CAPITAL AND CLASS

basis of Marxian economic theory . But no one made sufficient attempts to show clearly how these categories arose from people's production relations . There was therefore a complete break between the first and second concepts when we made the transition to abstract labour . Abstract labour was defined as physiologically equal labour, that is, the form of the production relations between people as commodity producers had been completely dismissed . We forgot this form and suddenly found ourselves in the sphere of physiologically equal labour, which is the same in all historical epochs . Making the transition from the concept of abstract labour to the concept of value, it must be said that these two concepts were always closely connected in Marxist literature . It would actually be very strange, if the adherents to the labour theory of value did not link the concept of labour with the concept of value . But this connection was paid for very dearly in that value was almost identified with labour and it was not clear in what way value is actually distinguished from labour . There was a break again in the next transition, from value to exchange value . Value was identified with labour, and so we did not know how exchange value arises from value either. Lastly, the relation between the concept of exchange and the concept of money was always very consistent in Marxist literature already since Marx emphasised this relation and substantiated it in particular . Thus the five categories we listed were split up into three groups . In the first group were the production relations of commodity producers, in the second, abstract labour and value, and in the third, exchange value and money . The system was only interrupted in two places, at the point where we have to move from the relations of production to abstract labour, and then again from value to exchange value . These interruptions disappear when we regard abstract labour as labour which possesses a determined social form, and value as the unity of content and form . Through these two reformulations we now obtain an uninterrupted logical interrelation of all the categories listed . A determined form of the production relations of people as commodity producers gives rise to the concept of abstract labour . From abstract labour in commodity production, viewed not as physiologically equal labour but as socially equated labour in a specific form, the concept of value emerged of necessity . The concept of value, considered as unity of content and form, is linked through its content with the preceding concept of abstract labour and through its form with the following concept of exchange value . Finally, the development of exchange leads of necessity to value . It would be contrary to my intention, if the interrelation between these categories appeared as some logical self-progression of concepts, which each give rise to one another . The close interrelation of the concepts which follow on from one another logically is explained by the fact that all these concepts are built up from the concept of the relations of production, between people as commodity producers . This concept conceals a multitude of real social relations between people, which consistently conflict and develop uninterruptedly . The economic categories express "forms of existence, determinations of existence, often just individual aspects of this given society" (Grundrisse, p .106) . The logical



ARCHIVE : ABSTRACT LABOUR AND VALUE

139

unity of the economic categories is due to the real unity of this society, the actual object of our study .

FOOTNOTES 1

2

3

A whole paragraph is devoted to the substance of value, beginning with the words : "Hence, when we bring the products of our labour into relation with each other as values, it is not because we see in these articles the material receptacles of homogenous human labour ." (p .74) . The following paragraph is devoted to the magnitude of value, and the next to the form of value . This refers to an article by I . Daschkowski "Abstraktuy trudi eknonomitscheskije kategorii Marksa" ("Abstract Labour and Economic Categories in Marx") in Pod Znamenem Marksizma 6, Moscow 1926 . These passages were apparently omitted from the German edition of "Studien zur Marxschen Wertheorie", and cannot be traced in the English edition .

REFERENCES Marx . K ., A Contribution to the Critique of the Political Economy, Lawand Wishart, London 1971 . Marx, K ., Grundrisse . Pelican, London, 1973 . Marx . K ., Capital, Volume I . Progress Publishers, Moscow, 1965 . Marx . K ., Le Capital, Garrier-Flammarion, Paris, 1969 . Marx . K ., Das Kapital, First edition 1867, Marx Engels Studienausgabe 1966, Vol . 11 . Marx . K ., Das Kapital / Marx Engels Werke, Berlin 1956 - Bd . 23 . Marx . K ., The Value Form, Appendix to Chapter I of the first edition of Vol . I of Das Kapital (1867) . Translated in Capital and Class No . 4 . Marx . K ., Theories of Surplus Value, Lawrence and Wishart, London 1972 . Marx, K, and Engels, F ., Collected Works, Lawrence and Wishart, 1976 . Rubin, I .I .,Essays on Marx's Theory of Value . Black Rose, Detroit 1972 .

REVIEWS

CONSIDERATIONS ON WESTERN MARXISM By Perry Anderson NLB (London, 1976), pp . 125, £4 .00 Reviewed by Michael Barratt Brown This essay constitutes a remarkable act of the most honest and honourable self-criticism . At first I did not believe it and supposed that I was being asked once again by the Editor of N .L .R . to compare the hopelessly empirical English radical tradition with the studiously theoretical continental tradition of Marxist scholarship . The Afterword clearly revises many N .L .R . editorial judgements but the whole text is in effect a critical reassessment of the body of western Marxist thought which succeeded the first race of giants of classical Marxism and which Perry Anderson has heroically sought for 15 years to bring to the attention of the benighted Anglo-Saxons in the British Isles . Apart from Sweezy and Keynes (who are treated almost synonymously) only two Anglo-Saxons (or Celts) appear in this book . William Morris and Raymond Williams are generously commended - in footnotes . Readers of New Left Review will have known for a long time that this is all very good for us in challenging our inveterate, insular empiricism, which is enriched only by our infatuation with historiography . There is not of course any doubt about the extraordinary range and depth of intellectual inquiry achieved by each one of Perry Anderson's innovators of Western Marxism whose writings are summarised here with consummate lucidity . Some of these even the English have heard of Gramsci, Colletti, Lukacs, Marcuse, Sartre ; others may be less well knownKorsch, Adorno, Lefebvre, Benjamin, Delta Volpe, Goldmann . What have they to say to us then? With the single and outstanding exception of Gramsci, Anderson has now concluded that the answer is quite extraordinarily little except in the field of aesthetics . Economics and politics, it appears, were abandoned



REVIEWS

141

by them for philosophy and aesthetics . His explanation is that theory and practice must go together . The mutually reinforcing power of Stalinism and of imperialism in the west has distorted and held back every revolutionary upsurge . Lacking a contemporary mass revolutionary movement, revolutionary thought retreats to academic chairs, shrivels and dies . The cutting edge of classical Marxism - from Marx and Engels to Lenin and Trotsky - was fired and forged in revolutionary heat . In a further touchingly self-critical 'Afterword' Perry Anderson concedes that this explanation overvalues 'activism' in any and every current conjuncture, devalues the study of history and might imply uncritical acceptance of the whole of classical Marxist theory . After such an open act of confession it would seem ungracious to cavil . But doubts remain on two scores . The first relates to the discussion of the hegemony exercised by capital over labour during the last 50 years . Perry Anderson has now explored Gramsci's concept of a hegemonic system of bourgeois power established not by force alone but through the cultural and ideological subordination of the exploited classes . He still appears to miss the crucial subordination of the labour process and the whole gamut of production relations . In discussing Adorno and Horkheimer he notes their insistence that 'the price of domination over nature, of which man himself was inseparably a part, was a social and psychic division of labour that inflicted ever greater oppression over men even as it created ever greater potential for their liberation . Subordination of nature proceeded pari passu with consolidation of classes, and hence subordination of the majority of men to a social order imposed as an implacable second nature above them . The advance of technology hitherto only perfected the machinery of tyranny .' Despite this perceptive summary he seems to regard this rather as a contribution to the ecological debate than as an insight into the essence of capitalist hegemony . The second relates to the pivotal association in his essay of revolutionary theory and revolutionary practice . Despite the warning in his Afterword there is an assumption throughout the essay that the working class only exists when it is engaged in mass revolutionary action and only when it is truly so engaged can Marxist thought develop . There is surely a chicken and egg problem here which is only resolved by recognising the dialectical interaction of revolutionary thought and working class actions that are small in scale and fall short of Lenin's truly revolutionary movement. Revolutions, especially revolutions of thought, are not only movements but processes extending over a considerable period of time and they do not occur only in the minds of the great thinkers but in the contributions of many lesser mortals . In the final section of his essay, Perry Anderson has already suggested some of the unfinished business on any agenda of Marxist theory - the contemporary laws of motion of capitalism and imperialism and the forms of crisis specific to them ; the nature of bourgeois democracy and the meaning and position of the nation in a world divided by classes ; the dynamics of the bureaucratic states that have emerged from socialist revolutions in backward countries ; the revolutionary strategies capable of overthrowing the modern bourgeois nation state ; the institutional forms of



142

CAPITAL AND CLASS

socialist democracy that could then arise . This is indeed the agenda, but is there really nobody thinking about it in Britain? There is, despite his protestation, a certain blindness in Perry Anderson's writings to the importance of political economy and to the English and Scottish contribution to it - a blind eye that fails to see, for example, a figure like Maurice Dobb to whom members of the CSE owe so much . As a Conference of Socialist Economists we are concerned primarily with the first items on Perry Anderson's list . He remarks in passing that 'work has only just begun on these issues' . The sole references in the text are to Ernest Mandel's writings. None of us, I suspect, would wish to claim more than that we were good journeymen in a great tradition . It may be that giants will grow from among our numbers . But I suspect that the days of the universal, well-nigh omniscient scholar-prince or intellectual and revolutionary leader are gone . Their place will be taken by teams and groups, not of narrow specialists but of men and women who have studied across the disciplines and married their common studies with working class actions at all levels . The peculiar dichotomy in Perry Anderson's writings between his deep respect for the few great thinkers and for the 'masses' when 'they themselves speak' comes from the peculiar Olympian view that he takes of the earth's surface . From great heights an all-embracing, synoptic view can be obtained and Perry Anderson's writing often brilliantly illuminates the main features of the landscape - but men and women, except in masses, appear as pygmies . We may concede that 'there were giants on the earth in those days' ; and still pray that he should 'make no more giants but elevate the race .'

THE BREAK-UP OF BRITAIN By Tom Nairn NLB (London, 1977), pp .368, £7 .50 Reviewed by lain Macwhirter, Edinburgh CSE This book is a product of the setbacks experienced by the working class and the Left since the optimistic days of 1974 . With mass unemployment returning, the welfare state being systematically dismantled, significant defeats on the industrial front, and the failure of Marxism to get out of the universities and become a 'material force', frustrated radical intellectuals are naturally looking for something to break the log-jam . In the U .K . context, nationalism is an attractive candidate because, as Nairn says, it appears to be " . . . the principal factor making for a revolution of some sort . . ." (p .89) . The Break-up of Britain deals with relevant questions and, on the analysis of 'neo-nationalism' in Scotland, the author has useful things to say . There are also chapters on culture and politics in Wales ; the impasse in Northern Ireland ; Enoch Powell's nationalism ; European integration ; and the decline of the British State . (Incidently, the centrepiece, 80 page analysis, "The Twilight of the British State" has already appeared in New Left Review no . 101 /2) . However, the book's overall significance is that it seeks to accommodate Marxism to nationalism, and as such its analysis is as potentially dangerous, in the context of working-class retreat, as any of its historical forerunners .



REVIEWS

143

In his last essay , "The Modern Janus", Nairn offers his solution to "Marxism's great historical failure . . ." ( p .329) over the national question . He attributes this failure to Marx's Eurocentrism and his lingering attachment to Enlightenment ideas of Progress . Thus Marx predicted a universalisation of class struggle with capitalism's subsumption of the world in the same way that Kant has foreseen a diffusion of civilisation . Both were utopian . Had Marx witnessed the form in which capitalism's expansion across the globe historically took place he would have identified "world poltical economy" as the "proper object" of historical materialism ; " . . . the only genuine 'structure' which can be held to explain the assorted 'superstructures' of capitalist reality" (p .352), and he would have seen that the contradictory uneven development of capitalism leads to national struggle being the principal contradiction and 'motor of historical development', not the empty abstraction of class struggle . Under the subheading, 'Nation Triumphs over Class' Nairn concludes that : "The story of uneven development is one of how the unheralded contradiction I have tried to outline has enveloped and repressed the other contradiction upon which Marxism laid such stress : the class struggle" (p .353) . Recognition of this primacy of national struggles is, Nairn claims, the only way to a rebirth of historical materialism from dogmatism . Many will be reminded here of the various 'Third Worldist' ideas floating around on the radical left . Nairn in fact disavows Third Worldism's model of bourgeois and proletarian nations several times in his book . But he does so on the grounds that : "Its primary fault may be rather that it takes no proper account of what one may call the 'middle class' of the world order" (p .183) . He means here a "petit bourgeoisie" of "relatively overdeveloped" nations and regions, like Catalonia, Northern Ireland and possibly Scotland : " . . . in sensitive zones of a larger political economy, alongside or in between powerful neighbours" (p .185), and developed through possession of a strategic resource of a 'cross-roads' trading position . Nairn therefore remains firmly within the 'nation-class' problematic, merely compounding the initial fallacy . The starting point of all Third Worldist notions is the rejection of, or confusion over, the analysis of the capitalist mode of production . There is consequently no theory of accumulation as 'form-determined' class struggle . The 'economic', the sphere of production, is effectively de-politicised and de-historicised in these theories, and conflict appears only at the level of distribution . Nairn's particular sleight of hand is to read 'capitalism' and the sociologist's concept of 'industrial society' as synonymous, and thereafter to analyse uneven development without reference to accumulation . Thus, social class relations can be dismissed as being in any way the fundamental contradiction (fundamental because inscribed in the very form of extraction of surplus-labour in this mode) in favour of nationalist struggles arising from the uneven development of 'industrialisation'. Nowhere does Nairn discuss the analysis of the mode of production . Consequently, his dismissal of class struggle as mere dogmatism is indistinguishable from bourgeois criticism of Marxism, and is emphatically not materialistically based .



144

CAPITAL AND CLASS

I have focused on Nairn's dubious 'materialism' because it seems important to stress that he is not simply working on the raw nerve of Marxism's abstractness . Marxists do, often, fail to appreciate the complexities of class struggle as it works itself out within concrete social formations. This arguably has led to debacles on the national question . But Nairn is going much further than just pointing this out . He is trying to insinuate nationalism into socialist theory as such . And his analysis of the neo-nationalisms threatening the integrity of the post-imperial British State - the main focus of the book - is informed by this nationalist perspective and departure from any rigorous materialism . Nairn feels deeply pessimistic about the possibility of independent working class action . He refers bleakly to " . . . the receding horizon of the socialist revolution and the permanent difference this has made" (p .87) . His analysis of the crisis of British capitalism and the sclerosis of the British State barely involves the working-class at all, let alone the tendency of the rate of profit to fall . Consistent with his lack of appreciation of accumulation itself as class struggle, Nairn's analysis is exclusively in terms of intra-bourgeois conflicts and processes . He describes the 'patrician' character of the British State, tracing this quality back to the specific class compromise achieved in the 17th Century between the landowning aristocracy, and the emergent bourgeois . This, he argues, led to an agrarian elite's occupancy of the political, adminstrative and 'civil' institutions, which was perpetuated by the era of 'City-based' Imperialism, and which is chiefly responsible for Britain's industrial decline today . "The pioneer modern liberal-constitutional State never itself became modern ; it retained the archaic stamp of its priority" (p .72) . Any modernising, technocratic 'revolution from above' is blocked by this pre-modern, transititional aspect of the British State . The main point of the analysis seems to be to encourage comparison with the absolutist dynasties which lingered on in central Europe until the early years of this century : "Although not of course an absolutist state, the Anglo-British system remains a product of the general transition from absolutism to constitutionalism : it led the way out of the former, but never genuinely arrived at the latter . . .It is a basically indefensible and unacceptable relic, not a modern state form" (p .75) . Neo-nationalisms seeking to extricate themselves from this sinking ship become therefore : " . . . relatively left-wing causes saving themselves from central reaction" (p .90) . Nairn appeals here to Lenin's support for the nationalisms that broke up the central European empires in the First World War . But a vital difference, surely, is that Lenin's argument was based on tactical considerations guided by proletarian internationalism, whereas Nairn's is based on the point of view of the irreducible nation entailed by his rejection of class struggle . Nowhere does Nairn attempt to make a convincing argument that separatism, under modern conditions, would benefit the working class . This review has focused mainly on the theoretical issues raised by only two of Nairn's essays . Space precludes examination of the other contentious theses in this book : Nairn's call for an independent Northern Ireland and rejection of Republican claims as "extravagant" ; his doubts about racism as a fascist phenomenon ; and his notorious support for the E .E .C .



REVIEWS

145

Similarly, this review cannot deal with the positive aspects of the work . This is unfortunate because Nairn's initial question : " . . . why has the threat of secession apparently eclipsed that of class struggle in the 1970's?", was well raised, if unsatisfactorily answered . Also, more specifically, his historical study of the unique compromise between Lowland Scottish bourgeois elements and English Imperialism achieved by the Act of Union, which left an array of characteristically Scottish institutions of 'civil society' - church, legal system and education - is essential for any evaluation of the strength of the current nationalist revival there . However, fashionable 'neo-Gramscian' analysis of superstructures and 'civil society' is mere sociology if the 'anatomy of civil society' (the capital relation and the structural contradictions arising on the basis of capital accumulation) is left out of the account . Nairn's dubious materialism, and more or less complete rejection of a working-class stand-point in this book needs to be made clear - especially given the number of radical intellectuals and journalists who are already using its arguments to justify support for 'progressive nationalism' north of the border .

A WORKER IN A WORKER'S STATE By Miklos Haraszti Penguin Books (London, England) 1977, pp .] 75, £0 .85 Reviewed by Janet Asquith Those of us who have never experienced life on the factory floor can get some understanding of it through the eyes of others . Factory life does not figure greatly in literature, maybe because the only way to tolerate it is by compressing it into a small compartment of your life and shutting and bolting the door . No one would willingly and freely reflect on factory life unless they were distanced from it : the time and motion 'expert' behind the screen of 'scientific management' ; the foreman behind the office door ; the sociologist behind questionnaires and 'methodology' ; or the revolutionary, finally, who can place before it the hope and will for something different . Haraszti's book is a compulsory addition to the short list of vital works on factory life : Upton Sinclair, F .W . Taylor, Braverman, the recent book by Beynon & Nichols, Living with Capitalism . . . What is different about Haraszti's book is that it is a report from the battlefront, with no punches pulled, no screens to distance us from what's going on . The title of the British edition is misleading and fatuous in the extreme : the original title, precisely and coldly accurate, was Piecework . The analysis of piecework forms the core of the book . The picture of the system is built up gradually by examining the role of each element in turn, always from the observation point of the machine operator, Haraszti himself. First the job itself, then the hourly rate, the setters, the piece-rates, rate-fixing and 'looting', the factory organisation, managers, inspectors, foremen, the collective agreement, bonuses, and so on . It is a mosaic,



146

CAPITAL AND CLASS

incomplete and fragmented : that is how it is to the machine operator . The utter degradation of the piecework system is fully revealed : the way in which it cripples every person involved, their creativity, their capacity for self-respect, communication with others . The book will teach you nothing directly about the class nature of East European societies or the Hungarian economic reforms, but indirectly, it teaches almost all you need to know . Haraszti is of course no average machine operator : he is a socialist, a sociologist in the best sense of the word, a poet, expelled from the university, jailed once and almost again for writing this book . With nice irony, he used the classic techniques of the Hungarian rural sociography of the 1930s so revered by the official cultural authorities . The controlled fury of the writing is remarkable also . Everyone should read it .

THE LOCAL STATE : THE MANAGEMENT OF CITIES AND PEOPLE By Cynthia Cockburn . Pluto (London, 1977), pp .207, £2 .95 paperback Reviewed by Jennifer Dale Cynthia Cockburn's study, based on three years research in Lambeth, is situated in the context of a crisis at the level of the local state . The massive increase in state expenditure on welfare has led to pressure from the central government for more centralised and cost-effective control of the local budget, whilst at the same time local authorities are expected to deal with the new urban contradictions thrown up by the outward movement of industry and the decay of inner city areas . One major theme of this study is the way the state attempts to alleviate this crisis through managerial innovation . The opening chapter provides a useful descriptive summary of the importation of corporate management techniques from business enterprises into local government . With their emphasis on identifying goals and making "rational" choices between alternative ways of meeting them, they have the effect of centralising and de-politicising decision making. Whilst these developments may have been a necessary part of gearing state expenditure more closely to the needs of capital - replacing old department rivalries with what O'Connor refers to as a "Class conscious political directorate" - they created new problems for the local state in Lambeth, which faced a mounting social crisis, especially in the sphere of housing, which spawned new forms of working class action, in particular squatting . The remoteness of the officers and councillors in Lambeth from the working class meant that, despite their sophisticated managerial structures, they could not, in fact, manage, and were forced to adopt a new strategy of establishing neighbourhood councils which would channel the struggles of the working class over housing and community services in the direction desired by the local state . Community action is thus seen as a mode of struggle which, even if it



REVIEWS

147

overcomes ambiguities as to its class context, can easily be incorporated by the state . The second major theme of the book is the presentation of an alternative strategy of struggle, defined not in relation to territorial communities, but in relation to the defence and transformation of those state services, and the family, which contribute to the reproduction of labour power . Such struggles could unite the clients of social services, state workers and women in the home, thus linking the spheres of production and reproduction and avoiding the "urban managerialist" paradigm into which the struggles of so many community groups fall . This book has many merits . The history of managerial innovation in Lambeth is well documented, and the arguments are presented in an attractive and clear way . It may well become the standard radical text for courses on local government and community work and its lucid summary in chapter 2, for example, of the essentials of a Marxist theory of the state and state expenditure make this analysis accessible to many who never turn the pages of New Left Review or try to fathom the debates between Poulantzas and Miliband . On the other hand, the imbalance in the treatment of the two major themes is disappointing . The research material on Lambeth relates almost exclusively to the first theme of local authority management, which is richly documented . The discussion of struggles in the sphere of reproduction is, by contrast, compressed into one chapter, and despite its many interesting ideas, remains schematic . This is a pity, because one is left wondering whether local authority management really is that significant . Corporate management techniques have led to changes in the internal organisation of local authorities, but in terms of impact on day to day town hall policy, the increasing central government control over local authorities, especially through the introduction of cash limits, would seem rather more significant . In similar vein, the book shows clearly that the "community approach", exemplified by neighbourhood councils, community development workers and government "small area" poverty programmes are all part of a management exercise - but considering their paltry scale, one wonders whether they should be seen as important when considering issues such as the incorporation of urban working class struggles . The second theme of the book on the struggle around the social wage, is on the other hand clearly of crucial importance, so that it was disappointing to find it so undeveloped and inadequately documented compared with the rest of the book . Finally, local government has received relatively little attention from Marxists, in terms either of analysis or of political intervention . Cynthia Cockburn suggests that there is really no difference between local government and the remainder of the state apparatus at local level . Her aim is, correctly, to de-mystify the Fabian notions of local democracy and municipal socialism . But in dissolving local government into the remainder of the local state, certain historical experiences of class struggle at the local level are written out, and the opportunities that socialists are now beginning to see for taking the struggle over the cuts and social services into the electoral arena at local level are ignored . Cynthia Cockburn poses the right questions - how to defend the gains made by the working class in the



148

CAPITAL AND CLASS

sphere of reproduction, whilst promoting demands and forms of struggle that point in the direction of emancipation from the domination of capital - but clearly much remains to be done in developing the analysis that can inform this strategy .

MEDICINE UNDER CAPITALISM By Vicente Navarro Croom Helm (London, 1976), £7 .95 hardback Reviewed by Mike Bury During recent years the writings of Vicente Navarro on medicine have drawn increasing attention from researchers in the health field, whilst remaining relatively unknown outside . It is therefore to be welcomed that a selection of his most important essays has been brought together in this volume, although they have all been previously published . The aim is to reach a wider audience and I, for one, hope it is successful . Navarro originates from Spain, where, after graduating in Medicine and Surgery he worked as a physician in coal mining districts, until his political activities against Franco led, in 1963, to his departure . He then worked in France, Italy, Sweden, and Britain, where he studied social policy . He finally settled in the United States where is he now Professor of Health and Social Policy at the Johns Hopkins School of Public Health . He has lectured widely and edits the impressive (though costly) Inter-

national journal of Health Services . The most important contribution that these essays make is to emphasize health as a major component of the social fabric, often neglected in social research and theory as worthy of treatment from a Marxist perspective . Health and medicine are explored as part and parcel of prevailing social relations, as both examplifying and influencing crucial aspects of contemporary social reality . Materials drawn from Latin America in particular give a cutting edge to his expositions (especially when one realises he was an advisor to the Allende Government) . His discussion of work and the structure of power which underpin persistent differences in health experience under capitalism, underlines the necessity to re-assert basic Marxist categories so frequently missing in the pragmatics of health policy or in sociological research . A well argued critique is also developed to take account of the 'radical alternatives' to modern medicine, particularly those of writers such as Illich . Navarro's essay on "The Industrialisation of Fetishism" has rightly won him a great deal of acclaim, inside and outside the medical field . Against romantic anti-capitalism and idealism he asserts the primacy of class relations in determining both the distribution and quality of health care . The analysis provides a useful starting point in tackling the growing wave of 'self-help' and 'community' approaches to health, as well as the naive elitist theories of the medical profession itself .



REVIEWS

149

Having said this in a strong recommendation of the book, one or two reservations must be made . Firstly, many of the points are made at a highly generalised level . References are made to the 'capitalist crisis' or the 'economic crunch' with very little specificity . Assertions about the crisis of legitimacy and profound weakening of the capitalist system suffer from the same over-estimation (though perhaps not the over-writing) that can be found so often in polemical expositions on the left . Secondly, there is a tendency to eclecticism in many of his essays and it proves to be a difficult task to pin him down . Navarro places himself in a tradition which is difficult to piece together . In the last section of the book he says "I aim at developing a Marxist theory of the State . . ., which has been greatly influenced by the contributions of Gramsci and of Milk band, Offe, Gough, O'Connor, Poulantzas, Althusser and others . . . " In fact a theory of the state is not really developed and the contradictory view of the influences he cites are left somewhat in the air . Concepts such as "Lumpenbourgeoisie" and "praxis" are used with little critical hesitation and at one point (p .201) he advances a criticism of Poulantzas's position on the state, "which I find linear, non-dialectical, deterministic and most - un-Marxian" without really saying why . In addition there is at times a superficiality in the analysis and a reliance on references to voting patterns and public opinion polls to sustain points which really require far more development. With these reservations, Navarro is to be congratulated (and I hope read) in his vigorous and lively treatment of a field too long neglected in Marxist thought .

ANTI-SAMUELSON By Marc Linder (in collaboration with Julius Sensat, Jr) . Urizen Books, 2 Vols, f3 .90 each volume Reviewed by Francis Green As anybody who has ever taken a formal course in Economics knows, Samuelson's Economics constitutes the bible of post-war bourgeois economics, of the neoclassical/neo-Keynesian synthesis . While of course it has its modern rivals, Samuelson's text came out for the first time in 1948, thereby allegedly announcing the solution to the problems of capitalism, and, ten editions later, it has become one of the world's best-ever selling textbooks . Its contribution to the propagation of bourgeois ideology can only have been immense . So it is particularly welcome that Linder has attempted to provide economics teachers on the left with a systematic critique of this text . Described on the cover as a 'monumental critique', Linder's AntiSamue/son certainly achieves monumental proportions, in terms of quantity of material covered . Taking the two volumes together there are over 900 pages . In this space Linder takes each of Samuelson's chapters and subjects

150

CAPITAL AND CLASS

them to a critique, step by step, and section by section . For this reason it has to be read along with Samuelson, rather than as a substitute for ever reading Samuelson at all . Correspondingly, there are four parts to the book . The first considers basic ideological concepts, the second is on Keynesianism ; the third covers the whole field of microeconomics, while the fourth covers imperialism and the world market . The theoretical approach is explicitly Marxist . Faced with such a long book it is pointless to try to describe and assess the analysis in detail . So I shall instead point to some of the guiding threads in his approach, and critically discuss some aspects of Linder's 'anti-textual' approach . Linder's method leads him to return again and again to four criticisms of Samuelson . First of all there is his ahistorical approach, which Linder always attempts to remedy himself in each application . Next, there is the criticism that Samuelson misuses one's pre-scientific notions about the economy ; or, rather, that he never transcends these notions . Third, and associated with this, there is Samuelson's eclecticism . Fourthly, as a bourgeois economist, his arguments are always limited to the sphere of circulation . Moreover, this is not surprising since he has no concept of the mode of production . There are some aspects of Linder's approach which I found particularly useful . For example, his use of statistical material to counter Samuelson's diagrams and tables is good . For it is true that Samuelson has an 'incredibly sloppy scholarly method' (p .xi) . He generally fails to attribute the sources of his statistics and his use of them is uncritical . Linder has put plenty of work into tracing and checking the statistics where possible, and producing his own material . Especially good here I found were the chapters on income distribution and on the trade cycle . Another aspect to be recommended is Linder's method of setting the modern theory in their historical context - both by discussing earlier economists, and more broadly by bringing in the wider economic situation . In this context, there is a good section on Keynesianism and its origins, and, in passing, on the role that Samuelson's text played in promoting it . And in addition I would say that Linder is generally to the point when he analyses the specific ideological purposes of Samuelson's chapters . But I have some criticisms to make . The main one is that is is a difficult book, the reading of which is only made easier if one is already familiar with many of Marx's concepts . The majority of Samuelson readers are those who follow one year (or two semester) courses in Economics, and I suspect that all but a few could be put off by its difficulty . My point is not that all books should be easy . Rather it is that Linder does not seem to have made sufficient concessions . Throughout the book he uses Marx's value concepts to criticise Samuelson . The reader will often come across references to the organic composition of capital, to the production of relative and absolute surplus value, and so on . Dotted around the book are spots where he takes time off following Samuelson's text to explain some piece of Marxism - as with for example, the section on money . However, I imagine that it would be difficult for the reader to be able to bring together these pieces of explanation into an intellectual whole . Moreover - and in this context this is important - there is no index to guide you .



REVIEWS

CAPITAL AN

151

So there is a problem in knowing how to read the book . One cannot really avoid reading it alongside Samuelson, chapter by chapter . But then in the Anti-Samuelson points of criticism may be made which may be difficult to follow because they are either explained in another part of the book (but you do not know where) or else they are not explained at all . Apart from this problem of presentation, the book also suffers from the problem of 'barrage criticism' : i .e . he criticises Samuelson for almost every section of every chapter of the whole book . This not only makes his own book very long, it also results in him making, as I detect in some places, unnecessary and incorrect criticisms . For example, he seems to rest his critique of Samuelson's standard U-shaped average cost curve by reference to one piece of evidence which suggests that all average cost curves are decreasing . This is ad hoc empiricism . As an alternative he suggests that : 'since the MC curve is L-shaped, the maximum profit is not reached by equalising MC and P, but rather is obtained by the greatest production at the lowest costs of production' . (Vol .11 p .151) . Does that mean at infinite output? Another consequence of barrage criticism is that criticisms are made with which no neoclassical economist would disagree . For example, in Chapter 13 on Supply and Demand, the point is made that if there are no limits on the expansion of firms with high productivity then in the longrun only supply determines price . But this is hardly a devastating critique of supply and demand curves . You would find this result in any microeconomics textbook . Here it is confusingly presented as a critique of supply and demand (the confusion being compounded by putting price on the horizontal axis and not saying so, contrary to normal practice) . (Later on in the same chapter, things improve when, in his discussion of Samuelson's applications of supply and demand, Linder returns to his main critical threads, and attacks here the ahistorical discussion of agriculture .) These points aside, the Anti-Samuelson could be of much value to the careful user . As long as it is supplemented by alternative reading material, the book could well be used to good effect by the individual chapter . Alternatively it could be used to follow right through the Samuelson course, even though it would then require continual help and explanation from a left teacher. On the whole, given the dearth of material with which it is possible to struggle against bourgeois economics in the bourgeois academies we can only welcome this book ; and at only f3 .90 per volume it's pretty good value for money .



Revolutionary Communist Papers Number One contains a full critique of the politics of the Revolutionary Communist Group - from which the comrades who formed the RCT were expelled in November 1976 - and outlines the central tasks facing the revolutionary movement today . Published by the Revolutionary Communist Tendency Articles on internationalism & party-building ; Stalinism & the British Communist Party Subscription for four issues UK £2 .00 UK Library £3 .75 Overseas £3 .00 Overseas Library £6 .00 Issue Number One UK £0 .50 (plus 13p postage) Overseas £0 .75 (plus 10p postage) Make cheques, postal orders and international money orders payable to RCT Association, and send to : BM RCT, London, WC1 V 6XX

is the first Anxvncan journal to develop a

NEW GERMAN CRI'Il(IUI•H comprehensive discussion It German politics, social theory . literature and art on an international level .

ne

ggr n

n

'

~

,

Pwm'nM MW vN„I, .nn 1- U NdIM.PI

L

critique german' criti ue

1~

s

NM

A ,' MMnm Nd,ur

nwJNm . r IL~n AsP.m

N

A••• a PP.$ .PSO~.n. .nJ P .Ptll.wl,w i Marker •n Y l~iu •+NUw• fMU ,:menrlNn In tM ....Im, . . F .nn, FMrr fM L., n m R ••m.nu m. fp •ra•. nPP , . vu ul Pstrl me UrrvY, NIUMI,~.Pw~

NEW GERMAN CRITIQUE, The Universuy o WisconsinMilwaukee, German Dept ., Box 413, Milwaukee, WI 53201 . Published three times a year . Annual subscription $6 .00 individuals, $12 00 institutions . Foreign $1 .00 extra . Back issue, $2 .50

PLUTO 4 PRESS MARXIST ECONOMICS FOR SOCIALISTS : A CRITIQUE OF REFORMISM John Harrison A member of the editorial committee of Capital and Class, John Harrison introduces the basic concepts of Marx's economics in simple language - as political ideas, part of a constant battle against reformism : from Proudhon in the 1840s, through Bernstein in the 1890s, to Crosland and Holland in Britain in the 1950s and today . £2 .40 paper £4 .80 cloth

ECONOMICS OF THE TRANSFORMATION PERIOD Nicolai I . Bukharin A key text on the transition from capitalism to socialism as a world system, published together with Lenin's marginal notes . £4.50 limited paperback edition Unit 10 Spencer Court, 7 Chalcot Road London NW 18LH

tl URIZEN BOOKS MARX ON MONEY Suzanne de Brunhoff In this excellent introduction to Marx's theory of money Suzanne de Brunhoff pulls together everything Marx wrote for publication on the subject. £2 .70 paper £4 .80 cloth

ANTI-SAMUELSON VOLUMES1&2 Marc Linder In his step by step critique of Samuelson's economic theory Marc Linder's two-volume study provides an essential alternative reader for every economics student. £3.90 paper £8 .70 cloth for each volume Distributed by: Pluto Press, Unit 10 Spencer Court, 7 Chalcot Road, London NW 18LH 01-722 0141

NOTES FOR CONTRIBUTORS

Capital & Class welcomes all contributions within the general scope of the . CSE, not only long articles of a theoretical nature but also short notes and polemics, correspondence, and pieces long or short which have a topical political relevance . The Editorial Committee is responsible for producing Capital & Class . The members of the EC are elected at the annual conference, half being replaced each year so that no-one serves for more than two years . The Editorial Board has now been replaced by an open delegate body which will discuss all general policy questions, which should include the relation of the journal to other CSE activities . The Editorial Committee hopes to develop a closer liaison with general CSE activities through working groups and local groups, who are invited to nominate members to act as points of contact and to attend periodical editorial meetings . Material submitted to Capital & Class is assigned by the reviewing editor to at least 4 people for comment, usually two members of the EC and two others . Frequently, papers submitted will have been discussed at a CSE group or dayschool, and might have been revised after discussion there . However, the Editorial Committee of Capital & Class likes to be involved at an early stage rather than simply being presented with a final version to accept or reject . Papers submitted to the EC may be referred for comment to a CSE group if they have not already been presented at such a group meeting . All submissions should go directly to a member of the EC or to John Harrison, Oxford Institute of Economics & Statistics, St Cross Building, Manor Rd ., Oxford . Material submitted in draft form should be typed single-spaced and wherever possible 5 copies should be provided. The final copy which goes for typesetting must be typed very clearly, doubie-spaced with very wide margins . Writers should try to avoid unnecessary jargon or other complexities of style ; for our basic format in using references and footnotes please refer to a current issue .