The European Union and the People

The European Union and the People

This page intentionally left blank

The European Union and the People Mette Jolly

1

3

Great Clarendon Street, Oxford ox2 6DP Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York © Mette Jolly 2007 The moral rights of the author have been asserted Database right Oxford University Press (maker) First published 2007 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose the same condition on any acquirer British Library Cataloguing in Publication Data Data available Library of Congress Cataloging in Publication Data Data available Typeset by SPI Publisher Services, Pondicherry, India Printed in Great Britain on acid-free paper by Biddles Ltd., King’s Lynn, Norfolk ISBN

978–0–19–921307–8

1 3 5 7 9 10 8 6 4 2

To Charles

This page intentionally left blank

Preface

This book started its life as a Ph.D. thesis at The University of Nottingham in the autumn of 2000. Initially, my focus was on EU democracy and the difficulties in taking democratic government beyond the nationstate. What sort of institutions would be suitable, from where would legitimacy be derived, and would the publics be alienated from the political elites? After having done much research I realised that I needed to move my focus from the governing to the governed who, it seemed, had been overlooked. The results of this analysis are presented in this book. At this stage I must emphasise that the bulk of the book was completed before the French and Dutch referendums on the EU’s constitutional treaty and that in no way should this book be seen as an attempt at creating a ’constitution equivalent’ to the black box of a crashed airliner. The aim of this book is not to explain referendum outcomes nor to uncover reasons for levels of public support for European integration in general. Instead, the book seeks to explain why the nature of the governed must determine which policy areas remain the responsibility of the nation states and which can be delegated to the EU level. Over the years, I have discussed these issues with a number of academics, friends and taxi drivers. In particular, I would like to thank Andreas Bieler, Lori Thorlaksson and Arturo Alvarez-Rosete, all of whom read first versions of chapter two and provided valuable comments at that very early but important stage in the process. Matthjis Bogaards kindly read and offered detailed comments to a late version of chapter four, and Liz Monaghan provided comments to the completed draft. Finally, the reviewers gave valuable feedback for which I am grateful. I would like to thank my Ph.D. supervisors at The University of Nottingham, Paul Heywood and Erik Jones, for their help and guidance over the years. It was Erik who initially opened my eyes to the notion that European integration presented an interesting area of study, something I

vii

Preface

had hitherto found distinctly hard to imagine. I am particularly grateful to Paul for his guidance and encouragement throughout the process, especially during the phase when it seemed obvious to me that burning rather than completing the thesis was the more attractive option. Also, I would like to thank the editors at OUP for taking on the book and the editorial and production teams for being extremely helpful and flexible. More than anything, I thank Vernon Bogdanor for taking an interest in my work and for encouraging me to publish the script. Mette Jolly London, February 2007

viii

Contents

List of Abbreviations List of Figures List of Tables

xi xii xiv

1. Introduction 1.1. The European Union ‘at a Glance’? 1.2. Assumptions 1.3. The Structure of the Book 1.4. The Book in a Wider Context

1 1 13 17 20

2. Debating Democracy in the EU 2.1. Introduction 2.2. The Debate 2.3. Four Paradigms 2.4. Conclusion

27 27 28 45 54

3. The Demos 3.1. Introduction 3.2. Rule by the People 3.3. The No-Demos Thesis 3.4. Conclusion

61 61 62 73 86

4. Models of Democracy for the EU 4.1. Introduction 4.2. The Models 4.3. Summary and Preliminary Conclusions

93 93 94 129

5. The Framework for Empirical Testing 5.1. Introduction 5.2. Defining a Demos 5.3. Operationalizing and Measuring the Demos 5.4. Data, Indicators, and Variables

141 141 141 144 146

ix

Contents

5.5. Descriptive Statistics and the Presentation of Results 5.6. Conclusion

154 154

6. The Demos—Empirical Analysis 6.1. Introduction 6.2. Findings 6.3. Conclusion

157 157 157 227

7. Conclusions 7.1. The EU and Democracy 7.2. The Nature of the EU Demos 7.3. Models of Democracy 7.4. The Future of the EU

231 231 237 238 243

Bibliography Index

247 259

x

List of Abbreviations

EB

Eurobarometer

ECB

European Central Bank

ECJ

European Court of Justice

ECSC

European Coal and Steel Community

EP

European Parliament

EU

European Union

EU6

Belgium, Germany, France, Italy, Luxembourg and the Netherlands

EU9

Belgium, Denmark, Germany, France, Italy, Luxembourg, the Netherlands, Ireland and the UK

EU10

Belgium, Denmark, Germany, France, Italy, Luxembourg, the Netherlands, Ireland, the UK and Greece

EU12

Belgium, Denmark, Germany, France, Italy, Luxembourg, the Netherlands, Ireland, the UK, Greece, Spain and Portugal

EU15

Belgium, Denmark, Germany, Greece, Spain, France, Ireland, Italy, Luxembourg, the Netherlands, Austria, Portugal, Finland, Sweden and the UK

EU25

Belgium, Denmark, Germany, Greece, Spain, France, Ireland, Italy, Luxembourg, the Netherlands, Austria, Portugal, Finland, Sweden, the UK, the Czech Republic, Cyprus, Estonia, Latvia, Lithuania, Hungary, Malta, Slovakia, Slovenia and Poland

EU2004

The countries that joined the union in May 2004: the Czech Republic, Cyprus, Estonia, Latvia, Lithuania, Hungary, Malta, Slovakia, Slovenia and Poland

EVS

European Value Surveys

QMV

Qualified Majority Voting

xi

List of Figures

2.1. Elements of democratic legitimacy

34

3.1. Two questions about the demos

68

5.1. Operationalizing the demos

154

6.1. Nationality vs. European in future (1994–2005)

158

6.2. Nationality vs. European in future with trend line for ‘nationality only’ (1994–2005)

159

6.3. Nationality vs. European in future with trend line for ‘nationality and European’ (1994–2005)

160

6.4. Nationality vs. European in future with trend line for ‘European and nationality’ (1994–2005)

161

6.5. Nationality vs. European in future with trend line for ‘European only’ (1994–2005)

162

6.6. Nationality vs. European in future with trend line for ‘don’t know’ (1994–2005)

163

6.7. Attachment to the EU

165

6.8. Attachment to own country

166

6.9. Attachment to region

167

6.10. Relationship between level of attachment to own country and level of attachment to the European Union (EU15)

168

6.11. Attachment to Europe EU25 (2005)

170

6.12. Attachment to country EU25 (2005)

171

6.13. Pride in nationality EU25 (2005)

172

6.14. Pride in being European EU25 (2005)

173

6.15. A shared European cultural identity?

175

6.16. Concern with fellow countrymen, Europeans and human kind—a comparison of means

178

6.17. A direct EU Tax?

179

6.18. Good and evil

182

xii

List of Figures 6.19. Freedom vs. equality

184

6.20. Should immigrants adopt the traditions and customs of the new country or maintain their own?

186

6.21a. Undesirable neighbours—homosexuals

187

6.21b. Undesirable neighbours—religion

188

6.22. Abortion

190

6.23. Why do some people live in need?

191

6.24. Religion and government

193

6.25. Democracy in our country

195

6.26. Is democracy the best political system?

196

6.27. Is membership of the EU a good thing? 1994–2006 with trend line for ‘A good thing’

199

6.28. Is membership of the EU a good thing? 1994–2006 with trend line for ‘A bad thing’

200

6.29. Is membership of the EU a good thing? 1994–2006 with trend line for ‘neither good nor bad’

200

6.30. Is membership of the EU a good thing? 1994–2006 with trend line for ‘don’t know’

201

6.31. Country’s membership good or bad—2006 figures (by country)

202

6.32. Country’s membership a good thing or other (2006 figures)

204

6.33. Feelings if the EU were to be scrapped

205

6.34. Age vs. good thing or bad thing

208

6.35. Age vs. feeling European

208

6.36. Quiz results vs. support for membership

221

6.37. Quiz results vs. feeling European

221

6.38. Quiz results vs. feelings about the veto

222

6.39. Quiz results vs. veto retained (combined categories)

223

6.40. Satisfaction with democracy in the EU

225

6.41. Satisfaction with democracy (own country)

226

6.42. Quiz results vs. trust in the EU

226

xiii

List of Tables

2.1. Four concurrent paradigms analysing democracy in the EU

55

4.1. Two contrasting models of the future EU-system (neo-Westphalian state vs. neo-medieval empire)

127

4.2. Models of Democracy for the EU

130

6.1. Year joined vs. European citizenship future feeling

164

6.2. Attachment to Europe vs. attachment to the EU

169

6.3. Measurements of transnational EC solidarity 1973–78

176

6.4. Prepared to make a personal sacrifice

176

6.5. Feelings if the EU were to be scrapped

206

6.6. Feeling European vs. membership good thing or bad thing

209

6.7. Benefited vs. good thing or bad thing

210

6.8. The meaning of being a citizen of Europe

212

6.9. The veto

213

6.10. Quiz questions and results 2006

218

6.11. Quiz questions and results 2004

220

6.12. Heard of the Council of Ministers vs. maintain veto

224

6.13. Trust in the Council of Ministers vs. maintain veto

224

xiv

1 Introduction

1.1. The European Union ‘at a Glance’? Since its birth as the European Coal and Steel Community, the European Union (EU) has been a source of public controversy and scholarly debate. Whilst giving rise to much hope and optimism (particularly in the early years), the EU has also caused much concern and criticism. Of course, there have been many reasons for this but almost all concerns link to one central issue, namely the puzzling idea that states would voluntarily give up formal sovereignty to a supranational institution.1 Students of politics realize that such state behaviour contradicts dominant international relations theories (perhaps with the exception of some strains of Kantian liberalism).2 However, one does not have to be familiar with international relations (IR) or political science terminology to feel puzzled by transnational political integration. It would appear that even fifty years on, the majority of people across the Union are still uneasy about the idea of pooled sovereignty. This can be seen from public opinion polls that consistently reveal a strong wish to maintain national opt-outs and veto rights. Despite considerable interest in travel for pleasure, and increasingly also work and studies (all of which may have been generated by increased opportunity), most people are not particularly interested in transnational civic participation. In an environment where elites increasingly see supranationalism as a viable means of tackling policy demands this is a problem because it creates an asymmetry between the civic and governmental levels of the polity. In the EU, the asymmetry between the governed and the governing manifests itself in three ways: lack of transnational identity, lukewarm support for the integration process, and widespread ignorance of the EU and its institutions. These are all related in that they each have the

1

Introduction

potential to affect the legitimacy of the Union. This book focuses on these elements with the emphasis on lack of a transnational civic bond between the EU’s citizens, a phenomenon also known as the no-demos thesis (Weiler, Haltern, and Mayer 1995) or the socio-psychological democratic deficit (Chryssochoou, Stavridis, and Tsinisizelis 1998). However, because it is meaningless to assess this issue out of context, the following paragraphs will address the relationship between the people(s) and the Union.

1.1.1. The EU and the People Considering the unusual institutional set-up, the complicated and manifold decision-making processes, and the fact that commentators struggle to agree as to where the power lies, it is hardly surprising that citizens find the EU distant and perhaps do not always see a connection between the goals of the Union and its everyday action as stated in the European Council’s Laeken Declaration, which is discussed below. Within the institutions, there is increasing recognition that the EU needs to be able to explain itself in simple terms, and yet, attempts to do so fail because the task is impossible. If explained in simple terms, the picture of the EU becomes blurred and ends up raising more questions than are answered. However, for a thorough and comprehensible explanation to be feasible, the prerequisites that an individual will need in order to grasp the mere basics are beyond what can be expected of people who do not spend the majority of their time involved in EU affairs either at the practitioner’s or the scholarly level. A primary example that illustrates this point is provided by the EU itself: On the recently revised EU Internet website, visitors are welcomed by an introduction to ‘The European Union at a Glance’, the first paragraph of which reads as follows: The European Union (EU) is a family of democratic European countries, committed to working together for peace and prosperity. It is not a State intended to replace existing states, but it is more than any other international organisation. The EU is, in fact, unique. Its Member States have set up common institutions to which they delegate some of their sovereignty so that decisions on specific matters of joint interest can be made democratically at European level. (http://europa.eu.int/abc/index_en.htm) (Italics added).

Despite the good intentions, one thing is confirmed by the quote: The EU cannot be grasped at a glance. Even in this relatively short paragraph, 2

Introduction

one could pinpoint material for a number of interesting and important research questions such as: Is the EU really a family or might that be taking the romanticism a bit too far? Is ‘peace and prosperity’ sufficient as a telos? Or is it empty rhetoric, as surely no political regime in its right mind would advertise working towards unrest and poverty? Furthermore: is the EU really unique, or is it perhaps the front runner in a new trend towards regional integration and single currency areas? Do states delegate some of their sovereignty as mentioned here, or do they in fact transfer some of their sovereignty? And, indeed, can sovereignty be split so that part of it is transferred and parts remain with the state—in other words, is there such a thing as partial sovereignty or does that undermine the meaning of the word? And finally, the question that inspired this book, namely, is it really possible—as promised in the quote—to make decisions democratically at the EU level? Not surprisingly, the answer to this question is complicated. Put simply, it all depends on the circumstances. There is a variety of reasons why it could be argued that making decisions democratically at the EU level is not possible, or at least not within the current vertical and horizontal institutional frameworks. But more fundamentally, the sociopsychological prerequisites for a sustainable, supranational EU may not be in place. This book investigates how the absence of a transnational demos affects the prospects for democratizing the European Union. The latter is receiving particular attention at the current time. This is partly because the constitution signed by the heads of state or government in 2004 has not been ratified because the conditions—that all member states must ratify for a treaty change to come into force—were not met and partly because there was already concern amongst EU decisionmakers that the relationship between this polity and its citizens were becoming strained. Hence, in 2001 the European Commission published a ‘White Paper on European Governance’ which acknowledged that people increasingly distrust institutions and that this problem is particularly acute at the EU level. In its executive summary the White Paper acknowledges that: ‘Many people are losing confidence in a poorly understood and complex system to deliver the policies they want. The Union is often seen as remote and at the same time too intrusive’ (European Commission 2001: 3).3 Even when addressing the gap between the Union and its peoples, it seems that the Commission does not acknowledge that perhaps policy is not the problem. There is clearly a tendency amongst practitioners to overestimate the value of outputs; however, when it comes to the 3

Introduction

European Union, the current legitimation problems are on the input-side (McCormick 1999: 148; Verdun and Stavridis 2001: 218). In fact, as is argued in this book, it is likely that too much output will serve to decrease rather than increase the legitimacy of the Union. An invitation to comment on the White Paper was issued on a designated Internet site. Contributions are published under the slightly misleading headline ‘Public Debate’. It is perhaps indicative of the relationship between the EU institutions and the people that the overwhelming majority of contributions published on the site are authored by either interest groups, governmental, non- or quasi-governmental organizations, or specialists.4 The White Paper established a number of principles such as the pledge to work for more openness in the EU policymaking process and also provided a road map for the subsequent Laeken summit, where European leaders initiated the convention for the future of the European Union which was later to agree the treaty establishing a constitution for Europe.5 The Laeken Declaration of December 2001 stated that: Within the Union, the European institutions must be brought closer to its citizens. Citizens undoubtedly support the Union’s broad aims, but they do not always see a connection between those goals and the Union’s everyday action. They want the European institutions to be less unwieldy and rigid and, above all, more efficient and open. Many also feel that the Union should involve itself more with their particular concerns, instead of intervening in every detail in matters by their nature better left to Member States’ and regions’ elected representatives. This is even perceived by some as a threat to their identity. More importantly, however, they feel that deals are all too often cut out of their sight and they want better democratic scrutiny. (European Council 2001: 20)

The European Council authorized a European Convention whose objective was to draft a new constitution for the Union. The Convention met from February 2002 and completed its work on 10 July 2003. The Draft Constitutional Treaty was publicized on 18 July 2003. This document was subsequently used as the basis for discussions on reform during the Intergovernmental Conference that opened in October 2003. Despite disagreements, the European Council approved the constitution on 18 June 2004 with relatively few amendments and the document was subsequently endorsed by the European Parliament on 15 January 2005 by 500 votes to 137, with 40 abstentions. Then the ratification process began. Of the twenty-five EU states some were to ratify by parliamentary

4

Introduction

vote and others needed to hold public referendums either as a result of national constitutional requirements or, as in the United Kingdom, at the discretion of the government (albeit due to political pressure). As is well known, many planned referendums did not take place because the treaty was rejected in both France and the Netherlands. European Union leaders reacted to this with a mixture of denial and disbelief, neither of which can possibly have endeared them to the public in any country, and many still refuse to declare the constitution dead although the rules on ratification of treaty changes are very clear. If one state cannot ratify the treaty cannot come into force. Nevertheless, the ratification process has continued although referendums have been postponed indefinitely in a number of countries, including Denmark and the United Kingdom (both countries where another ‘no’ would have been highly conceivable). After the initial panic had settled, the European Council on 16–17 June 2005 declared that they had agreed to hold a period of reflection which was to end by conclusion of the Austrian presidency in the first half of 2006. The idea behind this period of reflection was as follows: This period of reflection will be used to enable a broad debate to take place in each of our countries, involving citizens, civil society, social partners, national parliaments and political parties. This debate, designed to generate interest, which is already under way in many Member States, must be intensified and broadened. The European institutions will also have to make their contribution, with the Commission playing a special role in this regard. (Brussels, European Council, June 2005).

As its contribution to this, the Commission launched its Plan D for Democracy, Dialogue and Debate which aimed to assist member states in organizing national debates on the EU amongst civil society and citizens. This included Commissioners visiting member states and participating in discussions with national parliaments. At the time of writing the most recent council conclusions available are those that followed the end of the European Council June 2006. As regards the future of the EU Constitution, points 46–8 state as follows: 46. After last year’s period of reflection work should now focus on delivery of concrete results and implementation of projects. The European Council agrees a two-track approach. On the one hand, best use should be made of the possibilities

5

Introduction offered by the existing treaties in order to deliver the concrete results that citizens expect. 47. On the other hand, the Presidency will present a report to the European Council during the first semester of 2007, based on extensive consultations with Member States. This report should contain an assessment of the state of discussion with regard to the Constitutional Treaty and explore possible future developments. 48. The report will subsequently be examined by the European Council. The outcome of this examination will serve as the basis for further decisions on how to continue the reform process, it being understood that the necessary steps to that effect will have been taken during the second semester of 2008 at the latest. (10633/1/06 REV 1)

In other words, the period of reflection has been extended while statesmen, policymakers, academics, bureaucrats, think tanks, and ‘everyone and his dog’, as The Economist put it (The Economist, 4 May 2006), consider what to do, possibly with the notable exception of the EU citizen. The question of the no-demos thesis is directly linked to the issues of constitution-making, and on the basis of the discussions in the following chapters, I return to this subject in Chapter 7. In the following paragraphs I shall account in further detail for my assumptions and the methods I intend to use in order to answer the question.

1.1.2. Analysing the European Union—Methods and Approaches For many years, the ‘permissive consensus’ of the member state populations was taken for granted. In other words, elites assumed that the publics of the member states accepted the integration process without being particularly troubled by it. And although democracy was always an issue of concern to some (Rittberger 2003), it was not until the Maastricht ratification crisis in 1992 that it became a major factor in the public debate to the point where it seriously threatened the future integration process as well as pointed to a gap between a generally integrationist political leadership and a much more divided public’ (Wæver et al. 1993: 197).6 Thus, ‘the “permissive consensus” that accompanied European integration in its early stages is today greatly weakened’ (Schmitter 2003: 83).7 Before Maastricht, the public had been relatively unconcerned, the practitioner discourse had been dominated by matters of efficiency (output legitimacy), and academics had primarily been concerned with explaining what triggered and gave momentum to the integration 6

Introduction

process. Thus, from the 1960s to the early 1990s, the scholarly EU research was dominated by the rivalry between intergovernmentalism, which explained integration in terms of a convergence of perceived national interests, and neo-functionalism, which argued that once integration had commenced it was likely to gain a momentum of its own. Because the EU began as an agreement between states (i.e. the Treaty of Paris establishing the European Coal and Steel Community followed later by the two Treaties of Rome establishing the EEC and Euratom) it was natural that the research was dominated by the IR discipline. However, many now see the European Union no longer as a cooperation between states but an actual political entity of its own (if not a state, then at least a political system). Thus, accompanying deeper integration is a whole new way of analysing the European Union, namely as a political system rather than political systems interacting, (Hix 1994, 1999; Caporaso 1996).8 Based on Easton’s second definition of a political system—‘a means for resolving differences or as a set of interactions through which demands are processed into outputs’ (Easton 1964)—we can see that this development is completely justifiable.9 As a consequence, scholars from a variety of disciplines have begun taking an interest in the EU and it is now common to use comparative methods and political theory (normally applied in the national context) when analysing the Union. This sometimes causes problems because—at least in modern times—both discourses evolved around the nation state and if the EU is not a nation state it is possible that flawed analogies may affect the quality of the research. However, comparative approaches contribute to the insights provided by IR because they offer tools that are more suitable for analysing outcomes of integration as well as measuring other outcomes, such as democratic performance. Thus, where the IR discipline is perhaps more appropriate for the study of European integration, the comparative politics approach is more suited for studying European community politics (Hix 1994: 23). But despite the widely accepted notion that the EU is not a state, the statist analytical framework still continues to complicate the debate about the Union’s future. Despite the fact that many scholars begin their commentary by establishing that the EU is not a state, they often continue their analyses by consistently comparing the EU and its structures to those known from nation states. Or if not quite that, they evaluate the EU using concepts, standards, and terminology usually applied to the analysis of the state. Often this is an intuitive rather than a rational decision. Thus, we use the state model for comparisons, not because it is appropriate but because the territorial state as known since the Treaty of Westphalia 7

Introduction

has become the prototype from which criteria for all political systems are derived (Easton 1964: 109). Admittedly, it is probably safe to argue that the EU is not a state in the Westphalian sense. The EU does not have the monopoly on the legitimate use of coercion and violence within its geographical area and thus does not fulfil one of the most established criteria for what constitutes a state. But to go on from there to conclude that as a consequence the concepts and terms used when thinking of the EU system should not be the same as those used when thinking about domestic systems may be unreasonable. The quality of analysis is more likely to suffer if the EU is studied in isolation from other systems than if comparative methods are used in less than ideal circumstances (Sbragia 1992: 268). This is because such research is likely to produce descriptive rather than analytical accounts. Although the EU is not a state and although the EU may be sui generis, politics in the EU is not ‘inherently different to the practice of government in any democratic system’ (Hix 1994: 1) and therefore to ‘study the connection between political “inputs” and “outputs” on these issues, one would naturally use the discourse of “comparative politics”—the subfield of political science concerned with the study of the “internal politics” of “political systems”’ (Hix 1994: 1). For these reasons the theories, concepts, and knowledge drawn from the study of other polities are likely to be useful when analysing the European Union (Hix 1994: 24). Finally, it cannot be ruled out that the EU is a new type of state. In a recent introduction to Aristotle’s The Politics, the translator, Professor T. A. Sinclair, states that: . . . in the history of political thought the notion of a state is not a constant in the way that the notion of a triangle is a constant in the history of geometry. Wherever and whenever we read about the theory of the state we are reading about a conception of it current in the author’s time or else created by him. (translator’s introduction, Aristotle 1992: 24)

It follows from the above that one cannot—as is sometimes the case— jump from the ‘reasonable claim that the EU is a non-state political system (Hix 1999) to the conclusion that it can be legitimised by lesser standards than the state’ (Lord and Beetham 2001: 445). In other words, whether one considers the EU a state should not determine the democratic standards that the Union needs to apply. These depend on the powers that the polity has. Hence, it is necessary to establish standards of support and legitimacy that take into consideration the 8

Introduction

special nature of the EU without lowering the requirements at the same time.10

1.1.3. Legitimizing the European Union Perhaps because the EU is a relatively new political system, the main efforts have been towards increasing support for the regime, that is, the system itself. Hence, member states have tried to legitimize the EU in the eyes of their publics as a means of increasing capabilities and the discourse has been dominated by rhetoric such as the ‘national interest’. However, again because the EU has moved from an interstate cooperation further up the continuum towards an entity in its own right, support for the system is now sought at the individual level. There are many sources from where such support could be derived, however, at the individual level the distinction between affective and utilitarian support is useful in the EU context. Where affective support for political institutions is an ideological or non-material belief in the value of a political system, utilitarian support is the belief that the system advances the individual’s economic or political interest (Easton 1964; Hix 1999). Both are determinants as regards the perceived legitimacy of the system. And in fact, sometimes the distinction between support and legitimacy is unclear as scholars define the concepts differently. Chapter 2 provides a thorough discussion on sources of legitimacy for the EU particularly inspired by Weber’s three categories, namely, charismatic, traditional, and legal sources of legitimacy. However, at the present stage it is relevant to draw in Lipset’s definition because this establishes a link between political legitimacy, political identities and the ethos of a polity. According to Lipset ‘groups regard a system as legitimate or illegitimate according to the way in which its values fit with theirs’ (Lipset 1969: 79). It follows that in order to acquire what Hix and Easton define as affective support and what Lipset defines as legitimacy the EU needs discernible positive values with which citizens can identify. And although the formation of rights and citizenship were partly intended to increase the ‘sameness’ of Europeans and hence create a European identity and to an extent very much represent a value system, this is not sufficient as a foundation for a civic identity. This is because ‘the definition of the demos is connected to the telos of the polity’ (Weiler 1998: 232), that is, that further civic integration requires a positive end rather than the negative objectives of avoiding war and poverty. In this respect Lipset’s definition of legitimacy is useful. The definition establishes two primary conditions: (a) that people know what 9

Introduction

their own values are and (b) that they know what the polity’s values are. The EU’s objectives as mentioned above, that is, working towards peace and prosperity are—due to their vagueness and almost ‘commonsensical’ nature—not sufficient to constitute values. In the preamble to the Constitution, a range of values are mentioned that the EU builds on, that is, the respect for equality of persons, freedom, reason, the central role of the human person, and his or her inviolable and alienable rights, law, progress and prosperity for the good of all Europe’s inhabitants and peace, justice, and solidarity throughout the world (2004/C 310/01). In article 2 of the Constitution some of these values are repeated: ‘The Union is founded on the values of respect for human dignity, liberty, democracy, equality, the rule of law and respect for human rights’ (2004/C 310/01). Assuming that people are conscious of their own values, it should be relatively easy for them to make up their minds as regards the legitimacy of the polity (in Lipset’s definition). However, there are three reasons why this may not happen. First of all, most people are unlikely to ever read the preamble, let alone article 2. But even if people were able to derive the values of the Union indirectly (by evaluating the ‘behaviour’ of the institutions) the values are very general and not distinctly European but shared by many different cultures throughout the world. Finally, most people already have an established relationship with a polity that has similar values, that is, their state. Precisely because EU citizens are all guaranteed respect for their individual rights (at least formally) by their nation state they may not see the EU guarantee as an added benefit. And although it is perfectly possible for a person to be loyal to—or in Easton terms—support more than one political system at the same time (Easton 1964), such support is more likely to be sustainable if there is a relatively clear division of responsibilities between those systems or levels of government. And although it is evident that most people have multiple political identities (Easton 1964; Smith 1992) the populations of the EU member states have far stronger attachment to their nation state than the regional or EU level. For instance, when Europeans are asked how they define themselves, the large majority replies that they do so primarily in terms of their nation state. When one looks at their attachment to their region compared to their state and the EU, the attachment to nation state is far greater. Thus, although people are able to have a range of civic or political identities (in addition to their other identities as male/female, professional/blue collar, and so on), they are capable of having more than one political identity. But in Europe today the national identities are far stronger than any other: 10

Introduction Ethno-national communities are not automatically or necessarily the prime basis for society, but it is clear that ‘national identity’ when employed, is an extremely powerful mode of subjectivisation. People basically live with multiple identities and these do not necessarily have a clear or permanent hierarchy in relation to each other. But in specific situations, especially the closer one comes to war in either literal or metaphorical forms, the more there will be a hierarchy. In these conditions national identity is usually able to organize the other identities around itself. (Wæver et al. 1993: 22)

In a similar vein, Sbragia argues that The Union is in fact made up of nationalities which view each one another as foreign and react to European Union activities through the prism of nationality. While they are now foreigners who live peacefully side by side and are able to think of themselves as “European” in selected situations, their collective memories are such that to construct a transnational civil society is very difficult. (Sbragia 2002: 12)

For the EU this may present a problem because there appears to be a link at individual level between the strength of national identity and support for European integration. In an analogy to state-building, Ole Wæver claims that the ‘. . . more Corsicans feel Corsican, the less success for the French project’ (Wæver et al. 1993: 26). At the empirical level, Carey (2002) has shown a positive linear correlation between individual level strength of national identity and lack of support for European integration although— as is discussed in Chapter 6—others have reached different conclusions. Even if one were to credit the EU with the—relatively—peaceful period since 1945, as people get richer and the European wars hopefully remain in the past, the younger generations may begin to disassociate the Union with these achievements. Hence, Weiler argues that . . . the change in general public opinion is most interesting [. . . ]. There is, first, what one could term the paradox of success. In its foundational period, the European construct was perceived as part of a moral imperative in dealing with the heritage of World War II. Governments and states may have been happily pursuing their national interest, but the European construct could be cloaked with a noble mantle of newfound idealism. Within Europe war may be possible but it is certainly unthinkable, and with that huge success of its principal objective, what Europe is now presented as delivering is bread and circus. Remove the moral imperative, remove the mantle of ideals, and it is politics as usual with the frustrating twist that in Europe you cannot throw the scoundrels out at election time. So you try and throw the whole construct out. (Weiler 1998: 228)

11

Introduction

To a degree, this is already happening with the result that the missing ethos is becoming increasingly visible.11 In fact, the powerful momentum of European integration is not so much resident in enthusiasm for European Union but motivated in fear of unravelling what has already been achieved which again is bound in the fear of fragmentation and potential accompanied loss of security dynamics and economies of scale (Wæver et al. 1993: 197). This lack of a positive momentum translatable to the individual level constitutes a problem for the creation of a transnational identity.12 However, as is explained in Chapter 3, the EU struggles to define more concrete values for its polity because of the inevitable consequence of making a choice, namely that something else is not chosen. For instance, when some suggest Christianity as a religion and English as the EU language, it is implicit that French, Swedish, Spanish, etc. are not chosen, nor are Hinduism, Islam, and atheism. Unable to escape this all-or-no-values trap, the European institutions, particularly the Commission, have pursued two other strategies: The flags-and-balloons strategy and the utilitarian strategy.

1.1.4. The Flags-and-Balloons Strategy and the Appeal for Utilitarian Support After the Maastricht crisis and in an attempt to address the asymmetry between formal and social integration in Europe, worried EU officials adopted a ‘flags-and-balloons strategy’ (Wæver et al. 1993: 66) under the assumption that if loyalty is a question of identity, identity is a question of culture and culture is created by disseminating symbols and propaganda (Wæver et al. 1993: 66). Hence ‘. . . some strands of EC machinery work from this perspective with everything from balloons for the children with the EC symbol on, and putting the EU flag up in every possible place, to the EC passport and possible joint Olympics teams. . . ’ (Wæver et al. 1993: 66). Furthermore, it was expected that the interests of the community, expressed through the Commission, would slowly come to dominate the national interests and that this would help promote a sense of European identity (Taylor 1996: 141). This strategy has failed. People are feeling no more European and no more inclined to support EU membership than they were in 1993 (although they are not feeling less European or less likely to support European integration either). Accompanying the flags-and-balloons strategy has been an almost manipulative tendency to tone down integrationist ambitions. Hence, 12

Introduction

integrationists paper over changes by softening or blurring the discourse. Thus, ‘government’ becomes ‘governance’, ‘power’ becomes ‘competences’, ‘transfer’ becomes ‘delegation’, and pillars II and III are described as ‘decentralized’ policies—despite never actually having been ‘centralized’. However, as mentioned earlier, the strategy failed and subsequently ideas of a transnational culture have been toned down and the proclamations about culture (for instance in the Maastricht Treaty) now stress much more than previously the task of preserving national and regional variations (Wæver et al. 1993: 66). Instead, the EU institutions appear to have adopted a new strategy in that they are now counting increasingly on utilitarian support. Again, the EU’s website is a good indicator. On the index page, the left column addresses the visitor directly stating in line 2 that ‘Europe is fun’ as well as making a number of references to individual benefits and easy ways in which to contact Union institutions (the Ombudsman and the local MEP).13 This clearly indicates that a direct appeal is being made to Europeans both to understand how they benefit individually from EU membership and to engage with this new polity. Thus, it is clear, as mentioned earlier, that the idea of creating a European cultural identity has been somehow put back in favour of a more utilitarian and perhaps pragmatic appeal. To that extent, it is possible that the EU polity, in a democratic form, can be built around a political community based on a common perception that such a community benefits the people. Seen in this perspective the question is not so much whether the eventual European polity will be able to reproduce the same intensity of collective sentiment seen in many nation states, ‘but whether it can produce an encompassing system of stable and peaceful political relations without such a passionately shared identity or community of fate’ (Schmitter 2000a: 28). The question is how far and what types of supranational institutional arrangements this kind of support is able to sustain. These are some of the questions that are addressed in the following chapters.

1.2. Assumptions The following assumptions are implicit in the research question: Assumption 1: Democracy is superior to its alternatives It is assumed that democratic government is a good thing. Although many would be inclined to take this for granted, it is not a universally accepted 13

Introduction

truth that democratic government is superior to other types of government, such as for instance guardianship or a benevolent dictatorship.14 Assumption 2: Citizens of EU member states are entitled to be governed democratically I do not intend to proceed from the proposition that democracy is a good thing to the conclusion that all social systems should be governed democratically. Thus, I take it that there are other requirements for democracy in a political system than for instance in a privately owned company. In a legitimate political system, leaders are meant to look after the interests of the subjects. According to this philosophy there is a difference between a social system and a political system, that is ‘decision’-making’ and ‘political decision’-making’. However, the EU constitutes a political system and thus must be democratic, if we accept assumption 1. (This does not, of course, mean that the EU has to follow the same blueprints for democracy as nation states.) Furthermore, if it can be argued that the member states were democratic when they joined the Union, then it must follow that the EU ought to be democratic because otherwise the citizens have ‘traded for less’.15 Furthermore, in the previously quoted statement from the EU’s website, the EU itself claims that democratic decision-making is the ideal even at EU level. This is not to say that intergovernmental arguments of the democratic nature of the constituent parts are dismissed. Whether one takes the view that the EU is merely an interstate cooperation legitimized by its parts, a polity with strong supranational elements or even a new type of state, its citizens must be democratically governed because ‘its character as a non-state political system makes little difference to how the EU ought to be legitimated’ (Lord and Beetham 2001: 443). Assumption 3: A European democracy would have to be based on the principle of representation It is assumed that the democratic legitimacy of the Union shall be based on the principle of representation known in various forms from domestic member state systems and that such representation shall be equal, that is that citizens have an adequate opportunity, and an equal opportunity, for expressing their preferences (Dahl 1989: 109). This issue is addressed in detail in Chapter 2 and under the discussion of the network model in Chapter 4. In the terminology of Sartori, this means that the EU will be neither a direct democracy, a referendum democracy, nor a strictly electoral democracy.16 This does not mean that the EU should copy a particular system from a particular nation state. It may well mean that 14

Introduction

the EU will have to find a model of government which can rely on the democratic nature of the member states. Assumption 4: An enduring perceived lack of legitimacy can lead to disintegration, particularly in the presence of intervening factors such as a recession, or a security threat Apathy and lack of transnational identification have serious, negative consequences for legitimacy particularly when accompanied by lukewarm support. And this debate is not merely academic. A polity that lacks legitimacy is likely to be unstable, particularly if suddenly faced by a shock. And despite the fact that the EU is surviving and functioning at the moment, there is concern as to how well it would hold up if its members were to be hit by a serious and prolonged recession or a military crisis (Plattner 2003: 53). There have already been signs that in a crisis member states prefer to pull back from cooperation. In 2005, Italian reform minister Roberto Calderoli called for the return of the Lira (BBC News, 6 June 2005) and later that year, following the London bombings, France reinstated border controls despite being a contracting partner to the Schengen Agreement. [France is entitled to do so but the point is to illustrate that in a crisis the primary polity (the nation state) is likely to reclaim powers.] But more such situations could arise if hysteria developed over for instance, asylum issues, immigration, or even the threat to national identity that some groups may feel as a result of continued European integration.17 According to Wæver the reason is that the structures that have been removed as a result of integration have not been replaced because despite the fact that the states are changing, the nations remain the same. This is because cultural identity largely remains at the national level, even with a tendency to move down towards micronational ‘regional’ identity and the coupling of nation and state is weakened without a new synthesis being achieved at the European level (Wæver et al. 1993: 69). Thus, In the previous (pre-integration/‘modern’) situation when a nation/culture felt threatened (by immigration, foreign products, international cooperation), it could call on ‘its’ state to close the borders to immigrants, pursue a protectionist economic policy or withdraw from international cooperation. This seems no longer to be true. Border control and several forms of economic policy have already moved towards the EC level. (Wæver et al. 1993: 70)

If such feelings are looming underneath the surface, a shock could possibly trigger some negative developments. Although it may be an 15

Introduction

exaggeration, there is probably truth in Joseph Weiler’s claim that a public is prepared to accept anything ‘as long as there is enough butter on the bread’ (Weiler 1998: 225). The question is what happens when or if there is no longer ‘enough butter’. Thus, unlike the United States (a political system with which the EU is often compared) the EU will probably ‘continue to be marked by the presumption of freedom of exit’ (Stepan 1999: 33). But a crisis need not arise as a result of exogenous factors. A growing popular perception of the EU as an illegitimate polity may itself have serious consequences for the stability in the polity. By this I do not mean that there would be a hard security crisis or danger of regional war. But there are other serious consequences that can be caused by a widespread perceived lack of political legitimacy. These include electoral abstention, tax evasion, increased unpopularity of executives, increases in levels of litigation against authorities and increasing accusations of corruption as well as a blatant lack of adherence to rules. Of course, such matters are not in themselves destabilizing, although, if these factors are all present, the likelihood that states will be tempted to withdraw from the Union is higher. It is hard to say what developments potential exits could trigger, but to suggest that they could have a snowball effect is probably not entirely unrealistic. So far, only Greenland has left the Union with little wake to show for it.18 However, Greenland is not a state and even if it were, Greenland is not the United Kingdom or France. If one or more of the larger and richer countries were to leave, the Union would potentially lose net contributors as well as suffer reduced benefits of economies of scale, thus reducing the incentive for others to stay in. The Constitution did contain an exit clause with provisions for voluntary withdrawal (article 59) as well as clauses stating conditions whereby states could force another member to leave the Union (article 58). As this document has not been ratified it is the current treaties that regulate potential exit and there is disagreement as to what would happen if a state chose to leave. According to Majone most scholars believe such behaviour would be illegal (see Majone 2005: 216), however, it is difficult to see what the Union could do should a state decide to leave. Considering the fact that even when a state fails to comply in areas where legal instruments for ‘correcting’ their behaviour are in place (e.g. the Stability and Growth Pact), the Union would probably have to let a country leave if it wanted to. In order to accommodate inequalities as regards desires for integration of policy areas, a framework for enhanced cooperation exists which was 16

Introduction

also incorporated into the Constitution (article 43). Such a system (which is already in place in certain policy areas such as EMU and Schengen) is gaining popularity amongst many politicians. Indeed, in an interview with The Economist, former president of the European Commission, Jacques Delors, claimed that the enhanced cooperation model is the only viable solution for the EU. Delors expressed serious concern about the future stability of the EU and when asked if he would put the chances of the effective collapse of the EU as high as 50 per cent, his reply was ‘Yes’ (The Economist, 14 February 2004).

1.3. The Structure of the Book In a review of the literature on the democratic deficit (Chapter 2) I identify four different paradigms within the general debate about democracy in the EU. The first paradigm is the efficiency paradigm that analyses to what extent (if any) there exists a trade-off between democracy and efficiency in a polity, that is, to what degree could a high output satisfy legitimacy requirements. The second paradigm is the vertical democracy paradigm, which discusses the relationship between state level and EU levels of government, more precisely, whether the EU has perverted national democracies or in fact brought increasing democracy to citizens due to the increased capabilities facilitated by pooled sovereignty. The third paradigm, the horizontal democracy paradigm, analyses the division of powers between EU institutions. Finally, the socio-psychological paradigm debates what the lack of a demos means for the prospects of democracy in the European Union. It will be argued that the socio-psychological dimension is the most fundamental problem. In Chapter 3 this argument is refined and the concept of ‘demos’ is analysed. It is argued that there is more than one side to a demos, namely a legal as well as a social meaning. The legal dimension can be created from the top via the formation of institutions, including rights and citizenship. However, the social dimension consists of elements such as a common history, language, identity, and solidarity (towards other individuals as well as the polity). These are obviously far harder to create. Furthermore, in addition to the legal versus social typologies, it makes sense to make a distinction between two categories of properties that constitute the social demos. Weiler names the first group the ethnocultural conditions (Weiler, Haltern, and Mayer 1995). These are matters such as language, history, political habits, ethnic origin, and religion. I have named the 17

Introduction

second dimension the subjective emotional elements such as solidarity and identity. To some extent, the two dimensions are interrelated in the sense that high ethnocultural homogeneity would probably tend to increase the intensity of the subjective emotional identification. Because of the close links between concepts such as demos, society, and identity the chapter also discusses various sources of common identities, threats to identities, and the difference between the demos and the nation.19 Thus, when looking for ways of increasing democratic legitimacy in the EU, it is important that the peculiar nature of the people is kept in mind. Hence, Chapter 4 evaluates eight models of democracy suggested in the literature. Because the boundaries between descriptive and normative arguments are often unclear in the literature, I separate the descriptive and the normative evaluation of each model under the rationale that the fact that a model perhaps serves as a fitting description should not lead one to argue that it also represents the most favourable scenario. Hence, after an evaluation of the descriptive strengths of each model, I assess the extent to which they address the Union’s socio-psychological deficit. The models are the parliamentary model, the federal model, the presidential model, the consociational model, the confederal model, the two-tier model, the network model, and the neo-medieval empire model. Although, this is a relatively broad range of models, I am aware that the list is not exhaustive. Most notably, it leaves out deliberative models of democracy and to an extent ignores the idea that alternative sources of representation could compensate for a possible legitimacy gap. Such models consider ‘political participation by citizens in a broad sense which is not limited to participation in strictly political institutions (voting)’, hence ‘enabling more effective citizen participation in the governing process’ (Curtin 1999) based on the notion that organizations of the civil society can contribute to good governance in the EU (De Schutter 2002). The reasons for the omission are relatively straightforward. The idea that civil society can contribute to legitimacy in the EU has been successfully challenged by a number of scholars. Paul Magnette has argued that ‘these participatory mechanisms constitute extensions of existing practices, and are underpinned by the same elitist functionalist philosophy. They remain limited to “stakeholders” and will not improve the “enlightened understanding” of ordinary citizens and the general level of participation’ (Magnette 2003: 144). Rumford rejects the idea from another perspective—similar to the no-demos thesis, namely that contemporary European society is characterized by division without overall unity (see 18

Introduction

Rumford 2001). In a book which argues that the EU has no people, any suggestion that civic society could contribute to democratization would be bizarre. Furthermore, even if a ‘civil society model’ were considered it could never be an alternative to any of the suggested models because the EU still needs institutional and legal structures. Having said that, the network model as conceptualized by Leonard (see Chapter 4) is built on many similar ideas, that is, that individuals can acquire capability to act publicly through voluntary associations and movements (see Kaldor 2003: 285). But only a minority of people are members of such organizations and there are signs that they may actually contribute to the democratic deficit rather than the opposite precisely because of the limited membership. On the other hand, it will not be categorically denied that such interaction can at least create an additional link between the governed and the governing. But such links are precisely that, that is, additions to the institutional and legal framework and could exist—to varying degrees—in any of the models discussed in Chapter 4.

1.3.1. Conceptualizing the Demos In Chapter 5, one suggestion as to how the demos concept could be operationalized is offered. This chapter builds on the conclusions reached in Chapter 3, that is, that the social dimension of the demos can be divided into two categories, the ethnocultural dimension and the subjective– emotional dimension. Since the ethnocultural dimension is relatively well documented in Europe, this book deals mainly with the subjective– emotional dimension. Chapter 5 addresses potential problems with operationalizing this dimension, such as problems with validity and reliability, and on the basis of the literature derives four indicators of a demos, these being the ‘democratic self-consciousness of the collective citizen body’, ‘adherence to shared values’, the degree of a ‘desire to shape a common future’, and ‘public awareness of the transnational process’.20 In Chapter 6, the EU is tested against this framework, meaning that each indicator is measured on the basis of a range of proxy variables. Chapter 6 attempts to show this in an empirical analysis using survey data. It is not claimed that measuring a demos (a demos is seen as a matter of degree as is democracy) is possible with great accuracy. However, by using survey data, the chapter seeks to analyse the empirical reality in detail rather than merely making the statement that the EU has no demos. A detailed account of the process is provided in Chapter 5. The aim of the empirical analysis is not to make predictions about outcomes of possible 19

Introduction

future referendums, be they about the Euro or about the Constitution. Neither will I seek to explain support or lack of support for European integration.21 Rather, the aim is to offer some pointers as to the degree we can talk about a demos forming at EU level based on the assumption that like democracy itself a demos is not merely an absolute. Despite relatively widespread support for integration, there is no evidence to show that a European demos is forming. I conclude that seen in the light of the absent demos, majoritarian government is not—at present—feasible at EU level, if such government has to be legitimate. However, the operationalization of the demos shows that it may be possible to establish a transnational demos based on shared values which could open up new opportunities for legitimate transnational government even of a majoritarian nature. This could be the case if it were possible to create one political identity whilst maintaining numerous cultural identities.22 However, due to the strength of the national identities and the lack of transnational European solidarity it is important that those policies subject to supranational decision-making are policies which do not require such solidarity. In practical terms this means redistributive policies. Furthermore, keeping national vetoes in certain areas could also serve to both take into consideration the absent demos and calm those voters who are worried about giving up this last ‘handbrake’. This means that the EU will have to abstain from deepening integration in a number of areas such as direct personal income taxation and social policy.

1.4. The Book in a Wider Context In addition to being a study of the conditions in Europe and of the options available when it comes to the EU’s future, it is my belief that the discussions here are relevant at a more general level in that they reflect the challenges inherently involved when attempts are made at taking democracy beyond the nation state.23 The fact that political leaders increasingly see regional and international spaces as superior to national ones with regards to problem solving occasionally leads to ‘linkage deficits’ in the chain of representation, particularly, of course, in those cases where new layers of political elites are created. And since such linkages are a major component of democracy, when they are deficient, democracy is deficient (EtzioniHalevy 2002). Hence, the European problem with democracy applies in varying degrees to all forms of transnational government, only it is more visible, and certainly more acute, in the EU due to this organization’s 20

Introduction

advanced state of legal and institutional integration and the high number of policy areas in which the EU level is involved. When democracies take part in such international cooperation (potentially but not necessarily the first stages of integration), the decision-making process becomes remote from the people whose mechanisms of scrutiny are subsequently diminished.24 This challenge to modern, transnational government is summarized by Dahrendorf: We shall find that the hardest principle to apply beyond the level of the nation state is the leading role of the people, their sovereignty, the possibility of expressing their own opinion and their own will. Ways can be found to guarantee change without violence; means can be found to set up an effective system of controls. But we shall not find ways to make the people’s voice heard. This is the fundamental problem that democracy has to tackle throughout the world in the next few decades. For people want to have their say, and we cannot even imagine how to do that at the supranational level, except through street demonstrations or through the media, undoubtedly influential methods but with highly doubtful legitimacy, or else through discussion on the Internet, which is important but certainly not democratic, if only because many people, starting with myself, do not take part. (Dahrendorf 2003: 107)

This problem is linked to another issue, namely the question of the size of the polity, because increasing the scale of democracy magnifies the already significant utopianism of the democratic ideal (Dahl 1989) amongst other things because the larger the polity, the more abstract becomes the notion of the people (Sartori 1987: 25). Hence, all things being equal the larger the polity the harder it is to democratize. This is an old argument that also preoccupied classical philosophers: ‘experience has also shown that it is difficult, if not impossible, for a populous state to be run by good laws; at any rate, we know of no state with a reputation for a well-run constitution that does not restrict its numbers’ (Aristotle 1992: 404). Unlike Plato, who calculated the optimum number of citizens to be 5,040 (quoted in Dahl and Tufte 1973: 5), Aristotle does not mention any particular number but argues that for a state to be of optimum size it ‘must have the largest population consistent with catering for the needs of a self-sufficient life, but not so large that it cannot be easily surveyed’ (Aristotle 1992: 405). The reason is that Aristotle sees it as imperative that the citizens of the state are familiar with each other so that they know what kind of people their fellow citizens are (Aristotle 1992: 405). Of course, the fact that representation has replaced direct democracy has enabled larger democratic polities but the question remains whether even 21

Introduction

these have a natural limit. In other words, is there such a thing as the ideal size of a polity? Dahl and Tufte (1973) argue that this is not the case, claiming that: ‘No single type or size of unit is optimal for achieving the twin goals of citizen effectiveness and system capacity’ the reason being that democratic goals conflict and ‘no single unit or kind of unit can best serve these goals’ (Dahl and Tufte 1973: 138). In their recent volume, Alesina and Spolaore address this conclusion and comment that: For us, this is a valuable insight insofar as it stresses the impossibility of achieving an absolute best in which two goals are perfectly attained at the same time. However, if we want to consider the extent to which a goal can be achieved (more or less effectiveness, more or less capacity), the presence of trade-offs between goals does not mean that the optimal point on the trade-off has not been or cannot be reached. (Alesina and Spolaore 2003: 7)

These authors argue that the optimal size of a country is determined by a trade-off between the benefits of the size and the costs of heterogeneity. Focusing on public goods, the authors argue that when economies of scale are possible large polities make sense.25 However, the fact that people have different preferences has to be balanced against the advantages of economies of scale. The authors conclude that Certain policy prerogatives can be delegated to subnational levels of governments, and some international organizations and unions of countries have assumed certain policy domains where international spillovers and international economies of scale are important. (Alesina and Spolaore 2003: 217)26

Hence, environmental policy makes sense at EU level because this area enjoys both economies of scale and high level of heterogeneity whereas for instance foreign policy—despite the economies of scale—ought to remain an area for national governments due to the divergent interests of the national states. Having said this, the authors are not blind to the consequences for democratic legitimacy: In a multinational world of highly integrated democracies only a few political institutions can ‘control’ the provision of supranational public goods. Supranational institutions typically do not have the same political legitimacy of national parliaments. As a result any functions they delegate may raise complex issues of legitimacy for democratic theory. (Alesina and Spolaore 2003: 224)

It is probably fair to say that at the present time the EU represents the most advanced attempt at creating (as in most deeply integrated) such a transnational polity.27 Hence, it is likely that observers of other polities 22

Introduction

composed of numerous nations may eventually have much to learn from the EU experience. At this stage, however, it is still not clear what sort of lessons they will be.

Notes 1. In informal terms the notion of a completely sovereign state is only theoretical (see Watson 1992; Krasner, 1999). 2. Unless of course one takes the view that European integration is security driven and that the EU is the emergence of a new competing pole on the global scene (see Wæver et al. 1993: ch. 4). 3. An analysis of the White Paper goes beyond the scope of this chapter. The point here is to illustrate that the Commission recognized the problem. For detailed commentaries on the White Paper see Wincott (2001) and Sbragia (2002). 4. On the designated site, the Commission published reactions to the White Paper submitted between 25 July 2001 and 31 March 2002 stating that unless ‘explicit requirement was made for confidential treatment, substantial reactions to the White Paper, that have been sent to us either by post or to our electronic mailbox at [email protected] have been published here on this site’ (Commission 2002). There are 191 entries on the site of which only 16 are posted by individuals, the large majority of whom are experts or politicians. They include: three professors, five other academics, a member of the Northern Ireland Assembly (who is also an academic), one independent consultant specializing in government–business relations, one employee of Punto Europa (an Italian information centre on the European Union and part of the European Union’s information network), one writing on behalf of an interest group, the vice-president of the Italian branch of the European Movement and the Mayor of Stuttgart. Two contributors do not mention any professional affiliation and one of those presents himself as a ‘European Union citizen’ (http://europa.eu.int/comm/governance/debat_en.htm). 5. This book refers to the Treaty Establishing a constitution for Europe as ‘the Constitution’. 6. According to one early commentator, the European ‘project’ was always an elite enterprise: ‘. . . organized propaganda for economic integration in Western Europe has had big funds placed at its disposal—to a great extent paid, directly or indirectly by American subsidies—and it has filled space in the newspapers, sponsored studies, issued booklets, and distributed posters; it has financed journeys and conferences and run exhibitions. “Movements” have been started, but they have been confined largely to a small fraction of ideologists and to specialized politicians and government officials. They have never reached out to the peoples, and the courage has been lacking to attempt to change their attitudes on the material matters that

23

Introduction

7.

8. 9.

10.

11.

12.

13. 14. 15.

16.

24

are at stake. The propaganda has been kept on a most general plane and never faced the awkward practical issues’ (Myrdal 1956: 65). In a similar vein, Weiler argues that ‘Maastricht, refreshingly, gave lie to years of a Eurobarometer ostrich syndrome. It is clear that Euroscepticism is not just another English vice’ (Weiler 1998: 227). Lindberg and Scheingold had used the expression ‘political system’ about the EC almost thirty years earlier (see Lindberg and Scheingold 1970: ch. 3). This is slightly more questionable if one were to apply Easton’s other definitions ‘a set of interactions through which valued things are authoritatively allocated for a society’ (1964: 153) and ‘a means through which the resources and energies of society are mobilized and oriented to the pursuit of goals’ (1964: 153). These definitions are harder to apply to the EU due to the insignificant level of redistribution undertaken by the system. For instance, if one were to use Fukuyama’s definition of state-building—‘the creation of new governmental institutions and the strengthening of existing ones’ (Fukuyama 2004: 7) it would follow that the EU can be analysed as a type of state despite not being a Westphalian state. The Commission has tried to compensate for this by treating the young to a range of educational literature highlighting the virtues of European integration. Particularly intriguing is the tale of The Raspberry Ice Cream Wars (now out of print). This is a comic strip about three children, thoughtfully named Christine, Max, and Paul who are living in a peaceful Europe with no borders until one day when ‘a mysterious home page on the Internet pitches our three heroes into a land long before our time. Here, there are still guards at the city gate and every summer The Raspberry Ice Cream War breaks out anew. The people in this country need a good lesson in democracy and Europe. Christine, Max, and Paul arrive just in time to help.’ (http://europa.eu.int/comm/publications/young/index_en.htm). This is not to say that all polities need an ethos or an American dream. When a polity is old enough for people to stop asking questions about its right to exist, the need for such unifying ideals translatable to the individual level diminishes. http://europa.eu.int/index_en.htm. For a detailed discussions on these issues see Dahl 1989. There are, of course, cases where one could argue the opposite, i.e. that some countries only became democratic, or remain democratic, because of the pressure from the EU. Examples are Spain, Greece, and Portugal (Plattner 2003: 46). This argument also applies to at least 8 of the 10 members who joined in May 2004. Although elections will be the means whereby the people choose their representatives, the ‘representative’ label is more suitable than the ‘electoral’ label. This is due to the fact that an electoral system does not necessarily create representatives (Sartori 1987: 113).

Introduction 17. Or perceived to be a result of European integration. 18. Having been granted ‘home rule’ from Denmark in 1979, Greenland left the union in 1982 following a referendum in which 52% of the voters favoured withdrawal. 19. Although the terms are used interchangeably (including occasionally in this book), it is important to keep in mind that Europe and the EU are not the same thing. Many would probably argue that they say ‘Europe’ when what they really mean is the EU. (On the other hand, in the United Kingdom, people— including academics—often use the term Europe to mean the continent, i.e. all European countries, other than the United Kingdom.) Still, even if one were to argue that a European demos exists one could not automatically expand this argument to say that a European Union demos exists. Because if there were a European demos, surely that demos would include for instance the Norwegians and the Swiss. But Norway and Switzerland are not in the European Union, so part of the demos would be outside the political system. 20. The categories are derived from Chryssochoou 2001. 21. In fact, measuring the possible presence of a demos and support for European integration at the same time would be a rather peculiar project seeing that the presence of a demos would logically mean that social integration had already taken place in which case measuring the support for its legal basis might be slightly besides the point. 22. A distinction suggested by Wæver et al. (1993). 23. The ultimate stage of such supranationalism would be the creation of a global state, or a Kantian world government with one global constitution encompassing all humans. 24. In an almost completely comprehensive international cooperation (such as the UN) where most members are not democracies, the democratic process is even further diluted (Etzioni-Halevy 2002). 25. The authors talk about public goods in the broader sense, including a monetary and fiscal system, tax collection, a legal and judicial system, infrastructures, communication, law and order, public libraries, national parks, and embassies. 26. Methodologically, the book is an epitome of rational choice. It is not a book about European integration as such but a theory of transnational integration in general which focuses on utility maximization. Hence, the authors do not bring into the equation the ideological ideal behind the ‘European project’, i.e. the wish to avoid war on the continent. However, at the pragmatic level it brings valuable insights into what can be expected to be sustainable in terms of transnational political integration. Furthermore, the trade-off between economies of scale of public goods and heterogeneity means that one of the conclusions is that some policies are better decided at a microlevel and others at macro-level. This conclusion is very much in tune with the European integration project. However, the policies that according to

25

Introduction this model are suitable for the EU level do not correspond perfectly to those policies that are in fact decided at EU level. This illustrates the conflict between the ideological intentions of the integration project and the rational choice ideas of the common market. 27. According to Dahl: ‘Only the European Community shows much sign of harboring a supranational growth gene. There the incipient institutions of a “democratic” transnational political community are faintly visible’ (Dahl 1989: 320).

26

2 Debating democracy in the EU

2.1. Introduction In order to understand the significance of the no-demos thesis and its consequences for the EU’s democratization prospects, it is necessary to see it in the context of the wider debate on democracy in the EU. This literature is vast and scholars use an array of different approaches, identify a wide range of problems and reach a number of different—sometimes opposite—conclusions as regards the state of democracy in the EU. Some conclude that the EU is not suffering a lack of democracy (any more than could be expected and should be accepted). Others argue that the EU does have a democratic deficit but that this can be solved via institutional reform. Finally, the third group argues that the EU indeed has a democratic deficit but doubt that this can be resolved in the short term because the EU lacks the social foundations needed in democratic political systems. It is easy to see how such different conclusions affect the prospects of reaching satisfactory solutions. What may be a solution to the second group of scholars—such as for instance increasing the powers of the EP—would worsen the problem as seen by the third group because of the loss of democratic legitimacy a parliament arguably suffers if it is not representing one people. The first part of this chapter presents the methodological problems faced by scholars (particularly the inevitable problems caused by contested concepts) and discusses the central arguments put forward in the literature. In the second part of the chapter, it is argued that four concurrent paradigms can be identified within the general debate on EU democracy. The first paradigm is the efficiency paradigm, which asks to what extent, if at all, output can satisfy legitimacy requirements. The second paradigm is preoccupied with vertical democracy and the link between the EU and national levels of government. The

27

Debating Democracy in the EU

third is the horizontal democracy paradigm, which discusses the balance of power between EU institutions. And finally, the socio-psychological paradigm asks what the lack of a European demos means for the future democratization of the EU. This chapter shows how these debates, which are concerned with crucially different aspects of democracy and/or of the EU itself, are taking place alongside each other and that real dialogue exists only within each debate and not between them, hence the term ‘paradigm’.

2.2. The Debate The academic debate about democracy in the EU has a number of particular characteristics. First of all, it is interdisciplinary both in the sense that scholars from various disciplines address the issue but also in the sense that there is a rather high degree of communication between the various disciplines, these being mainly law, political science or international relations, economics, international political economy, and sociology. The broadness of the debate ensures a wider perspective although it also means that the literature on the subject is substantial, and more importantly, that approaches chosen by scholars not only vary according to disciplinary paradigms but also disciplines per se. The danger is that with the many different angles even within individual disciplines, the debate becomes unproductive because scholars are all preoccupied with the areas that they have identified as the ‘problem’. So not only does disagreement exist about potential solutions but these disagreements are intensified by the fact that the problem has not been clearly identified. In other words, the preoccupation of different scholars with different elements of the so-called ‘democratic deficit’ means that few have a picture of the ‘whole’. Lord (1998: 11) uses an analogy by Donald Puchala to describe this problem: a group of blind men approach and touch an elephant in order to find out what it is. Each person investigates a different part and therefore they all arrive at different conclusions.1 According to Lord, Puchala’s analogy is also applicable to current research into the EU’s alleged democratic deficit. The fact is that not only do scholars define the deficit differently but the result of looking at isolated parts often is that what one theorist would consider a reasonable solution to a particular problem would worsen the problem as identified by another scholar. The best example of this are the discussions about the role of the EP. According to many scholars, the EP is the only real source of democratic legitimacy 28

Debating Democracy in the EU

in the EU. Therefore, the EP should have more powers, such as the prerogative to dismiss individual Commissioners rather than just dismiss the Commission in its entirety. Other scholars hold that this would increase the EU’s democratic deficit, arguing that the EP is not the EU’s true source of legitimacy. Thus, what constitutes a solution to some, will inevitably add to the problem as defined by others. Furthermore, not all scholars subscribe to the view that the EU actually lacks democratic legitimacy, as some argue that the assumption that the EU has a democratic deficit stems from a tendency amongst scholars to compare the EU to an abstract, perfect model of democracy rather than empirical studies of the workings of political systems (see Moravcsik 2002: 621). This issue is discussed below. Secondly, the debate on democracy in the EU is relatively new. Having emerged during the 1980s, the debate intensified in connection with the ratification crisis of the Maastricht Treaty in 1992. The reason democracy in the EU received relatively little attention prior to this point in time was the ‘strong technocratic orientation and element of intergovernmental cooperation’ which had so far characterized the EU (Andersen and Eliassen 1996: 253). Prior to the Maastricht ratification crisis, it was generally held that the integration process derived its legitimacy from the democratic character of the member states (Majone 1998: 12). This was justified in ‘the principle that a body that is composed of legitimate governments is itself legitimate’ (Lord and Beetham 2001: 444). Furthermore, a so-called ‘permissive consensus’ was taken for granted by the elite and other main actors (Lindberg and Scheingold 1970), although whether this was really a permissive consensus or—as argued by Elmar Brok (2001: 31)—rather a case of ‘benevolent indifference’ remains an open question. There would appear to be much truth in Moravcsik’s claim that ‘[t]raditionally, the EU has been justified largely by its output-economic and, according to some, geopolitical benefits for its member states and citizens’ (Pollack et al. 2000: 6)2 although Moravcsik contends that ‘polarization of elite and public opinion is eroding the broad consensus that supported integration for the past 3 decades’ (Moravcsik 1998: 4). This view is echoed by Schmitter who argues that ‘Europeans feel themselves, rightly or wrongly, at the mercy of a process of integration that they do not understand and certainly do not control—however much they may enjoy its material benefits’ (Pollack et al. 2000: 2). Furthermore, the terminology is not intersubjective (an indicator of the pre-paradigmatic situation). The most clear example of this is the term ‘the democratic deficit’ itself which is applied to the EU by a number of scholars despite the fact 29

Debating Democracy in the EU

that no unanimity exists as to what the deficit entails (Lord 2001: 642), leading at least one scholar to deem the term ‘meaningless’ (Hix 1998: 19).

2.2.1. The EU All of the above issues are related to another problem, namely the lack of a common perception of the ‘European Union’. There is no agreement as to what kind of entity the EU is and the union itself has never sought to define or describe its political character (Nugent 2003: 464). Despite the emergence of other examples of regional integration, scholars struggle to find an equivalent to the EU. This has led many to conclude that the EU is sui generis (Wessels 1996; Majone 1998; Dinan 1999; Green 1999; Nugent 2003). Consequently, any discussion about democracy in the EU is bound to be more complex than a discussion about democracy in a nation state. In a debate about democracy in France or Sweden, there would at least be an intersubjective understanding of ‘France’ and ‘Sweden’ because most scholars would consent to classifying the two as ‘states’ thereby providing a frame of reference for the object of the discussion.3 This is not the case if France or Sweden were to be replaced by the EU, which despite its actual existence has no conceptual ideal to measure itself against.4 And although no actual democracy ever measures up to the standards set for an ideal democracy (Dahl 1989: 84; Mény 2003: 9), in a state there would be a common idea of what the ideal should be, making it comparatively easier to identify and evaluate shortcomings. The EU democracy debate is more complex because it concerns the development of the polity type as well as the specific case of the EU.5 Furthermore, the EU is an ‘overlapping polity’ meaning that it is not there to replace the member states (as far as most commentators are concerned) but an additional layer of government. As a consequence, it is impossible to separate the EU democracy debate from the structural EU debate. After all, ‘arguments about Europe’s democratic deficit are really arguments about the nature and ultimate goals of the integration process’ (Majone 1998: 5). Hence, all proposals on how to increase democracy will partly be prescriptions for the design of the system. Of course, uncertainties about democracy are not specific to the EU (Mény 2003; Pharr, Putnam, and Dalton 2000). However, it would appear that public acceptance of the EU is suffering to a higher degree than that of the regimes of the European nation states at least as regards the ‘mainstream political public’, by which I mean the part 30

Debating Democracy in the EU

of public opinion which is not separatist or anti-system in one way or another.6 This may be related to the fact that the EU is a relatively new regime, which can only derive legitimacy from legal sources (efficiency and democracy). In contrast, most domestic systems derive part of their legitimacy from tradition, which is why certain particular characteristics of a system (such as the House of Lords in the UK, the banning of certain political parties in a number of European countries, or the illegality of Holocaust denial in Austria and elsewhere) are accepted by national publics although these arrangements may seem peculiar to outsiders.7 This is not to say that legal and traditional legitimacy are equal in national systems, only that there is an element of tradition present which allows for differences in domestic systems, an acceptance which does not translate to the EU polity.

2.2.2. Democracy Measuring democracy is difficult, most notably at the operational level. But this is precisely why attempting to do so is important. The inherent elasticity of ‘democracy’ as a concept necessitates evaluations because of the risk that the term can be stretched to cover a wide range of regimes and in a worst case scenario could result in democracy being diluted to the extent that it becomes completely meaningless. Or as Lord puts it— slightly more elegantly: ‘The notion that there is no universally appropriate approach to democratic politics should not, however, be taken as opening the door to an extreme relativism in which anything will do’ (Lord 1998: 15). On the other hand, the term democracy is not so ambiguous that scholars do not share at least some lowest common denominators as to what constitutes a democracy. In fact, it is assumed here that mainstream scholars would be able to agree on a set of lowest common denominators for a democratic system. This is a similar rationale to the one followed by Collier and Levitsky in their ‘minimal definitions’ of democracies where they ‘deliberately focus on the smallest possible number of attributes that are still seen as producing a viable definition’ (Collier and Levitsky 1996: 6). As has been demonstrated, the two central terms of the discussion (the EU and democracy) are both highly ambiguous. However, going back to the headline ‘democracy in the EU’, in the context of this particular discussion even the seemingly harmless preposition ‘in’ needs some clarification. The reason being that for an investigation into EU democracy to be complete, the term ‘democracy in the EU’ should cover 31

Debating Democracy in the EU

all aspects of democracy in the EU. Not only those pertaining specifically to EU institutions and decision-making (horizontal democracy) but also the former combined with domestic processes at national or regional level (vertical democracy), since these are all part of the multilevel system to which the EU populace is subjected. An account of these arguments will be provided in the description of the vertical democracy paradigm.

2.2.3. The Democratic Deficit—Analytical Considerations Whether the deficit is such that the EU can be labelled ‘not democratic’ depends on two things. First of all, it depends against which yardstick for a democracy we measure the EU (what are our lowest common denominators for a democracy) and secondly whether we consider the EU to be a system from which we would normally require these lowest common denominators to be satisfied. This explains why there are such great disagreements in the literature about the state of democracy in the EU. The term democratic deficit in its initial meaning denoted the gap that arises when power is transferred without being accompanied by adequate amounts of transparency and accountability (Marquand 1979: 64). The problem was that: In the Community system, no one is unambiguously answerable for anything. The buck is never to stop; it is hidden from view, in an endless scrimmage of consultation and bargaining. This may not matter much when the Community’s competencies are as restricted as they are at present. If they were extended sufficiently to overcome the challenges described above, it would matter a great deal. (Marquand 1979: 64)

According to Marquand, monetary union and qualified majority voting (QMV) in the Council would increase the powers of the centre without making it more accountable. The definition of the democratic deficit is no longer this clear-cut and in recent writings Marquand’s usage of the term has become known as the ‘standard version’ or ‘conventional definition’ of the democratic deficit (Weiler, Haltern, and Mayer 1995; Hix, 1998). At the present time, the phrase has numerous different meanings depending on who is applying it and to what part of the EU they are applying it. However, before looking at the numerous operational meanings of the term it is worth considering its analytical implications. 32

Debating Democracy in the EU

Obviously, the use of the word ‘deficit’ immediately gives a negative impression. But there are different ways in which this is interpreted, and the analytical connotations of the term will always affect the weight scholars give to its operational meanings. The literature offers a range of theoretical definitions that—each in its own way—are useful in providing frames of reference for analysing the EU. Habermas provides a general definition, not specific to the EU, according to which: ‘Deficits in democratic legitimisation arise whenever the set of those involved in making democratic decisions fails to coincide with the set of those affected by them’ (Habermas 2000: 52). Because most would consent to the impracticality of complete participation in collective decisionmaking, the implication here is that democracy must be ensured by means of a representative system. Thus, we can use this definition to ask the question of whether those who make the decisions are representative of the people who have to abide by them. This would be one example of a lowest common denominator of democracy. Lord (1998) presents two definitions specific to the EU: 1) . . . the gap between standards of democratic practice in national and Union politics, and arises when powers are transferred from ‘more democratic’ national institutions to ‘less democratic’ European ones. 2) . . . the gap between ideals and practices which may arise in any political system with the implication that domestic systems cannot be used as a privileged benchmark against which to judge the Union since this begs the question of whether those institutions are themselves democratic. (Lord 1998: 14)

The first definition is a result of comparing the EU to an actual democracy (such as that of a democratic state) and is very similar to Marquand’s original definition. The second is a result of comparing the EU to a conceptual democracy (a theoretical ideal, i.e. not an ideal type but a normative ideal). However, precisely because the EU is not a state and because the EU is not a theoretical construct, in order to analyse the EU one needs a slightly more analytical and less empirical definition. Wincott suggests two interpretations of the word deficit used in this context. According to one understanding, the term ‘immediately suggests that there is too little democracy’ (definition 1) whereas the other interpretation suggests ‘that Europe has overdrawn its democratic account, that it was “borrowing” to finance current projects, against future legitimacy’ (Wincott 1998: 414) (definition 2). It is implicit in definition 1 that the situation is unacceptable and needs to be addressed whereas definition 2, the idea that ‘the democracy 33

Debating Democracy in the EU

account is overdrawn for the sake of efficiency’ seems to carry an implicit acceptance of such an overdraft for the sake of ‘progress’. This raises the question of whether there is ‘room’ within the concept of democracy to temporarily suspend democratic legitimacy (input democracy) in order to achieve effectiveness (output democracy)? In other words, this line of argument gives associations to justifications voiced by non-democratic leaders who list a range of objectives they need to achieve before they restore (or introduce) democracy. Thus, to the onlooker faced with the notion that the EU has a democratic deficit, the outcome of any analysis will, to an extent, depend to which of these interpretations he or she intuitively subscribes.8 Before looking at the alleged shortcomings of the EU system, it is useful to establish some lowest common denominators of democracy. Democracy can be viewed in absolute as well as relative terms. And although democracy is a contested concept, it is possible to construct a definition based on these lowest common denominators. Lord (1998) offers a useful definition of common denominators of a democracy. These can be shown in a simple graphic. These are the tests that all democracies will have to satisfy: (1) Authorization: There has to be proper authorization of leader and power relations. This is twofold. First of all, there must be ‘consent to the overall structure of the power relations’ (Lord 1998: 16). This is equivalent to Weiler’s category ‘legitimacy as a category of formal/legal validity’ (Weiler 1998: 231) or in David Easton’s terminology the ‘regime’ as in the ‘rules of the game’ (Easton 1957: 392), that is the system whereby leaders get to become the leaders.

Equal representation

Identity

Authorization

Accountability

Figure 2.1. Elements of democratic legitimacy

34

Debating Democracy in the EU

Secondly, the leaders themselves must be accepted as legitimate (Lord 1998: 16). (2) Accountability: Rulers have to be accountable to the public, which in turn requires transparency (Lord 1998: 15). (3) Representation: Decisions will have to be made in a manner which is representative of public needs and values (Lord 1998: 15). (4) Identity: The latter presupposes a common identity because needs and values have to be shared (Lord 1998: 15). Scharpf describes this phenomenon as the ‘essential sameness . . . arising from pre-existing commonalties of history, language, culture and ethnicity’ (Scharpf 1999: 8). Whilst these tests may not be beyond dispute, they constitute a fair description of the lowest common denominators for a democracy. However, it would be reasonable to add the word ‘equal’ in front of ‘representation’ (see Dahl 1989: 109). In an article from 2001, Lord and Beetham do exactly that, stating: ‘Where there is political equality, the structural incentive is further honed towards tracking the needs and values of the median citizen, whose views have the unique quality of being the least average distance from those of everyone else where preferences are normally distributed. . . ’ (Lord and Beetham 2001: 454), and that ‘. . . the Union cannot escape the need for representative institutions if it is to deliver the core attribute of democratic governance, which we take to be public control with political equality’ (Lord and Beetham 2001: 458).9 Such a need for equality is not merely a matter of rhetoric but is an inherent part of democratic government: Every discussion of democracy basically revolves around three concepts: popular sovereignty, equality and self-government. These concepts are interrelated. The people are sovereign in that they are equally sovereign because unequal sovereignty would imply that some people are sovereign and other people are not. (Sartori 1987: 58)

2.2.4. The Democratic Deficit—Operational Definitions So, does the EU live up to these democratic standards? That, of course, depends on who we ask. The most common democratic deficits identified by the literature will be listed below. It is important to keep in mind that this list is a summary of a broad literature rather than one single source. This is made particularly evident by the fact that some of the deficits 35

Debating Democracy in the EU

are contradictory. Most notably, if one applies Wincott’s first definition (‘too little democracy’), the deficit which says that the EP has insufficient power, cannot coexist with a deficit which argues that the EP lacks legitimacy. It could, of course, be argued that a more powerful EP is an institutional measure designed partly to create an EU demos. Invariably, such a strategy would mean having to accept a ‘transition period’ of a deficit until the demos developed and thereby eradicated this sociopsychological deficit. Such an argument would be a result of adopting the second definition, that is, the overdraft analogy. This rationale, however, has one rather crucial weakness in that it intuitively uses ‘time’ as an intervening variable. But in time anything can happen. For instance, Europeans could turn against the EU because of a perceived lack of legitimacy. Thus, it is rather optimistic to assume that the fact that time goes by contributes to the developments that one perceives to be desirable. In the following paragraphs, the most commonly raised alleged deficits will be listed. Subsequently, these arguments and their counter-arguments are discussed and categorized according to their foci under each of the four paradigms. (1.a) The accountability argument—EU legislature versus EU executive: This position is particularly dominant amongst scholars who use nation state democracies as the yardstick by which they measure EU democratic performance. Basically, the position holds that because the EP does not form the ‘government’, that is, the executive, voters cannot hold the executive to account.10 (1.b) The accountability argument—national legislatures versus EU executive: This critique is directed primarily at the Council and is mainly associated with Daniel Wincott and Renaud Dehousse. In short, these critics argue that the Council escapes democratic control due to secret deliberations. This is an argument under the vertical democracy paradigm and will be addressed under Section 2.3.2. (2) The QMV-critique: This is a criticism of the way in which decisions are made in the Council, in an increasing number of policy areas (or rather the fact that the Council has a legal possibility to use QMV should achieving consensus be impossible). Due to the nature of QMV, outvoted governments are prevented from truly representing the views of their domestic electorates. Hence QMV is undemocratic to outvoted publics. This argument is associated with a range of scholars, most notably Fritz Scharpf who argues that

36

Debating Democracy in the EU . . . the citizens of countries whose governments are outvoted have no reason to consider such decisions as having democratic legitimisation. In fact, even the cautious expansions of qualified-majority voting in the Single European Act and in the Maastricht Treaty have triggered judicial responses and public debates in the member states which are so critical of the legitimacy of majority decisions in the Council that any further progress will need to be based on more solid foundations of legitimisation. (Scharpf 1996: 26)

The issue will be addressed below under the ‘myth of the democratic deficit’. (3) The agenda-setter argument: In short, the fact that the Commission is an unelected agenda setter is detrimental to democracy because agenda setting in itself is a powerful instrument. This issue will be discussed in detail under Section 2.3.3, the horizontal democracy paradigm. (4) The bureaucracy argument: This is the notion that the EU is a gigantic bureaucracy where decisions are largely made by technocrats rather than the peoples’ elected representatives. There are two main reasons why this can be seen as a problem: First of all: In the European Union the largely bureaucratic decision-making process of the experts in Brussels offers an example of the type of democratic deficit caused by the shift away from national decision-making bodies to interstate committees of government representatives. (Habermas 2000: 52)

Secondly, there is a moral argument that such decision-making is in fact quasi-guardianship because decisions about public matters . . . require (implicitly or explicitly) judgements both moral and instrumental. These decisions are not and cannot be strictly about ends, but neither are they nor can they be strictly about means. No intellectually defensible claim can be made that policy elites (actual or putative) possess superior moral knowledge or more specifically superior knowledge of what constitutes the public good. Indeed, we have some reason for thinking that specialization, which is the very ground for the influence of policy elites, may itself impair their capacity for moral judgement. Likewise, precisely because the knowledge of the policy elites is specialized, their expert knowledge ordinarily provides too narrow a base for the instrumental judgements that an intelligent policy would require. (Dahl 1989: 337)

(5) The expansive judiciary argument: It has been suggested that the European Court of Justice uses its rulings to promote integration thereby continuously expanding its own powers (Weiler 1981; Rasmussen 1986).

37

Debating Democracy in the EU

Hix refers to this phenomenon as ‘activism by the ECJ’ (Hix 2005: 136). This argument is particularly relevant in the discussion about the legitimacy of the integration process. (6) The lobbyist argument: Criticism has been raised of extensive EU lobbying. Regarding the policy process, Wallace argues that it ‘is significant that some 20 per cent of Commission staff are “irregulars”, far removed from the Weberian model, many of these irregulars being drawn from the private sector and with some functions of policy implementation now farmed out to the large army of consultants clustered around Brussels’ (Wallace 1993: 300). Similarly, Schmitter questions the legitimacy of such interest participation: Especially since the signing of the Single European Act in 1985, Brussels has been literally invaded by ‘Euro-lobbies’—not just formal interest organizations but also a considerable variety of social movements, large enterprises, and law firms. While all this pluralism (to use the American expression) is entirely appropriate in a modern democracy, its highly skewed nature does raise some questions about whether these channels for expression of particular intensities are freely and fairly available to all citizens of Europe. (Schmitter 2000a: 81)

(7) The identity/solidarity argument: This is the notion that the EU citizenry is merely a citizenry on paper rather than a real civic body and that consequently there is no basis for self-government. The details of this argument are clarified in Chapter 3. Most scholars agree that the EU does not have a particularly strong transnational identity. The disagreement arises when one attempts to establish whether this matters. Chapter 3 is dedicated to this issue. (8) The social critique: In short, the social critique argues that the EU has a neoliberal bias and that this in itself constitutes a democratic deficit. This is a political argument and therefore, not encompassed by the four paradigms. In other words, although it is recognized that output affects legitimacy, the specific political nature of output is not a constituent element of the lowest common denominator for a democracy. This critique has been raised by a number of scholars, perhaps most famously by Scharpf. Such ‘social’ arguments are not encompassed by the paradigms discussed in this book. For the sake of good order, the paragraphs below explain why. First of all, the EU is also frequently accused of having a socialist bias (see Moravcsik 2002: 617). Secondly, the notion that a neoliberal bias should constitute a democratic deficit is disputable if one defines 38

Debating Democracy in the EU

democracy as a system of government by the people (as does Scharpf). To that extent, Moravcsik’s counter-critique—the essence of which is that Scharpf does not know what the public interest is—is convincing because if the public interest is better served by liberalization then ‘. . . assuming for the moment that there is in fact a neoliberal tendency promoted by the EU, there is no neoliberal bias’ (Moravcsik and Sangiovanni 2003: 128). Furthermore, . . . member countries of the EU differ not just in their level of development but in their political ideology. The critical question is why the ‘public interest’ of more social democratic polities should prevail over the public interest of more liberal polities’. (Moravcsik and Sangiovanni 2003: 138)

Finally, from the no demos perspective the model of ‘social Europe’ could not contribute to EU legitimacy. This is because such a model requires precisely those levels of solidarity that are not present. In the words of Majone, Advocates of a ‘social Europe’ must face the fact that in several countries even national redistribution in favor of poorer regions is increasingly challenged in the name of fiscal federalism and regional autonomy. It is difficult to see how politically acceptable levels of income redistribution could be determined centrally in a polity like the EU, where economic, social, political, and legal conditions are still so different. We may conclude that the attempt to legitimate the Union by developing a ‘European social model’ is bound to fail because it goes against the clearly expressed preferences of the governments and the citizens of the member states. In fact, a large-scale development of welfare policies at European level would likely aggravate the legitimacy problem, reinforcing the popular image of a centralized and bureaucratized Union. (Majone 2005: 30)

Not only the EU, but also the EU member states themselves are accused of such bias. In an article from 1997, Annabel Kiernan argues that due to globalization (which in her interpretation equates ‘increased market competition’) European governments have felt forced to create ‘post welfarestate’ economies which have resulted in the removal of certain social rights and the acceleration of social exclusion to an extent that makes Kiernan ask the question of whether ‘we have seen the end of the social contract’ (Kiernan 1997: 323). In essence, Kiernan’s main point seems to be that—although the EU provides certain rights for workers—these rights are not extended to all citizens (because not all citizens work) and, therefore, the weakest are excluded from the benefits of the EU although not from the setbacks of the nation state. This, she argues, undermines EU’s democratic legitimacy. 39

Debating Democracy in the EU

Kiernan’s argument is based on two assumptions: (a) that states have ceased to protect citizens and (b) that social rights are an essential part of the social contract and that regimes, therefore, derive part of their legitimacy from social delivery. As regards the first assumption, it is far from reasonable to proclaim EU states post-welfare. Citing OECD figures, Majone states that ‘social expenditure in twenty-one advanced democracies between 1980 and 1997 shows that a race to the bottom in welfare provision did not take place. Average social spending as a percentage of GDP increased from 19.2 in 1980 to 23.8 in 1997. The increase in social expenditure was particularly significant in countries like Finland, Greece, Italy and Portugal, indicating that social expenditure dynamics have been driven by catch-up as well as by rising unemployment and population aging’. (Majone 2005: 184). As regards the second assumption, that social rights should be an essential part of the social contract, this depends on which group of contract theorists one is inclined to rely for definitions. To liberals, the social contract is mainly about protecting negative rights whereas the absence of certain positive rights does not necessarily mean that a system is not democratic. From a liberal perspective social rights could even make a system less democratic as many liberals would hold that the social rights of one individual would serve to violate the more important liberal rights of another individual thereby making the system in question less rather than more democratic (see Berlin 1958). Therefore, the importance subscribed to negative versus positive rights in a political system cannot be used as a yardstick to measure degrees of democracy or a distinction between a democracy and a non-democracy but only to determine which kind of democracy the system is, that is, an ideal type liberal democracy (Mill) or an ideal type social democracy (Rousseau). For this reason, arguments about social rights and social policy substance are not included in the ‘lowest common denominator definition’ as described above.

2.2.5. The ‘Myth’ of the Democratic Deficit A more general critique of the ‘democratic deficit literature’ is the argument that the EU is the victim of an unfair comparison because of the tendency to analyse its workings and institutions in ‘ideal and isolated terms’ (Moravcsik 2002). According to Moravcsik, most critics ‘compare the EU to an ideal plebiscitary or parliamentary democracy, standing alone, rather than to the actual functioning of national democracies adjusted for its multi-level context’ (Moravcsik 2002: 621). In a similar 40

Debating Democracy in the EU

vein, Mény argues that in the public debate, the political elites and the media have always behaved as if democracy was the ‘government of the people, by the people, for the people’, a definition which he describes at mythical (Mény 2003: 9). In more specific terms, the counter criticisms of the democratic deficits can be summarized as follows: Andrew Moravcsik holds that QMV in the Council is tolerable from a democracy point of view when all governments agree on a common policy goal but not when the parties face disagreement over fundamental objectives (Moravcsik 1998: 44), in other words, if the means to an end are disputed but the end itself is not. Generally speaking, this definition is rather elastic, particularly if, say, one considers integration to be the end, and everything else to be the means. However, when looking at concrete policymaking instruments, the argument is more convincing. For example, in the second pillar (the CFSP), common strategies are agreed via unanimity and subsequently, common positions and joint actions under a strategy ‘only’ need QMV to be adopted. If one accepts common positions and joint actions as means to an end (the end being achieving the common strategy) there is no situation where a government can find itself pursuing an objective that it has not previously supported. At this stage, one could possibly object that this argument does not take into account the fact that council members can be replaced. If there were to be a change of government in a member state in between the adoption of the common strategy and the subsequent common position/joint action, one could perhaps question whether it is ‘democratically justifiable’ that the new government were to be bound by such a common strategy. As regards the ‘Brussels bureaucracy’ Moravcsik also presents a counterargument stating that the EU bureaucracy is in fact rather modest and heavily constrained by institutional checks and balances (Moravcsik 2002: 607). Hence, the alleged threat of a European superstate is a myth (Moravcsik 2002: 606). If one compares statistics of EU bureaucratic staff (a popular comparison in many textbooks is that of Birmingham County Council, which employs more staff than the European Commission), this argument is clearly convincing. However, the key word in Moravcsik’s argument is myth. It is not necessarily the numbers or the institutional situation that determines whether the EU is perceived to be legitimate. If the myth exists that the EU is an impenetrable technocracy, such a myth is in itself enough to affect legitimacy. When it comes to matters of legitimacy, it is quite possible that a myth is as strong as a truth whether it is ‘factually’ warranted (perhaps one could even argue that 41

Debating Democracy in the EU

a myth is a ‘truth’ to those who believe it). This is possibly linked to the issue mentioned elsewhere that citizens see the EU as ‘foreign’. Hence, while ‘many of the member-states are equally bureaucratic, the fact that the Commission is “foreign” to everyone in all the member-states makes its administrative complexity far more striking and frustrating’ (Sbragia 2002: 5). Although it may be an exaggeration to say that the Commission is foreign to ‘everyone’, the fact that the EU is a new polity means that the almost habitual legitimacy that often is a product of familiarity does not apply to the Commission. The issues raised by the critics of the democratic deficit literature are largely valid. It is true that the EU is often analysed in ideal and isolated terms. This is indeed one of the great problems in the normative EU debate. As mentioned earlier, one of the problems that faces the EU is that it is being constructed at the same time as its blueprint. In other words, the ideal type is being constructed alongside the case. Hence, the EU is being analysed in isolated terms because, as explained earlier, so many scholars consider it sui generis, and in ideal terms because it is a polity whose structures change comparatively often, thus providing scope for continued improvement. Thus, the EU is constitutionally flexible. This applies to institutions as well as policy areas. Institutional changes (although some have been minor) have taken place four times since 1986, if one counts only the treaties, five if one counts the deliberations of the convention as an institutional change. As regards policy, there have been changes to EU policy involvement in all treaties since the the Single European Act (SEA). This again is something that makes the EU unique. Hence, unlike in a state, there is ongoing debate about what policy areas should involve the EU. (This, admittedly, is also true within nation states as there may be debates about what constitutes the private as opposed to the public responsibility.) For these reasons (amongst others discussed below) the conditions for the debate about democracy and legitimacy in the EU are different to the conditions for the debate about these matters within nation states. As stated elsewhere, there is no doubt that as far as legitimacy is concerned the EU has a much more difficult task than most other polities. When an alleged deficit is pointed out, the entire EU construct as opposed to only the deficit will be questioned. In case of fraud in the European Commission, sceptics call for the abolition of the EU whereas in similar circumstances within a state, critics will call for the resignation of a minister or even a government, but not the abolition of the state itself (see Sbragia 2002). Mény picks up on this

42

Debating Democracy in the EU

point, arguing that despite the fact the deficit is myth, there is a real problem because ‘many citizens feel frustrated by the way the European institutions work’ (Mény 2003: 11). What is needed are new paradigms, rules, and institutions, perhaps even including some elements of direct democracy, as in pan-European referendums (Mény 2003: 11). One could perhaps go even further and say that citizens may not even be frustrated by how the institutions work. This would be difficult to argue in a book that finds that so many people are unaware how the institutions work (see Chapter 6). According to the findings in Chapter 6, people are more ignorant than frustrated. One problem that applies to Moravcsik’s critique, however, is that democracy and legitimacy are used almost interchangeably. For instance, Moravcsik states that: ‘Concern about the EU’s “democratic deficit” is misplaced. Judged against existing advanced industrial democracies rather than an ideal plebiscitary or parliamentary democracy, the EU is legitimate’ (Moravcsik 2002: 603). This is perhaps why Moravcsik disregards the fact that the perception that there is a democratic deficit is sufficient to affect legitimacy, even if technically there is no democratic deficit. Mény, however, does make a distinction between democracy and legitimacy, arguing that in the EU perhaps the ‘problem might be more of a question of legitimacy than democracy’ (Mény 2003: 11). Also, in many contemporary democracies there is a tendency towards an overweight of the constitutional element over the popular element (Mény 2003: 5). This includes the delegation of powers to other unelected bodies such as central banks on the implicit grounds that economic efficiency has become more important than political accountability: Democratic citizens have the feeling that their votes matter less and less, that parties in power do not deliver what they promised while in opposition, that policies do not fit their needs and aspirations. . . . The legitimacy of the system is weakened when citizens perceive their representatives as incapable of acting according to the messages they have sent through their votes, protest or other form of mobilization. This feeling can be attenuated or exacerbated according to various circumstances; by the capacity of the opposition to propose credible alternatives; by the international environment; by rates of economic growth, etc. However, these variations in time and space cannot hide a paradigm shift in Europe over the years. One pillar of democracy—the popular one—has become weaker and weaker while the other has been strengthened, contributing to the progressive isolation of governing elites from people’s pressures. (Mény 2003: 4)

43

Debating Democracy in the EU

By this reasoning, however, the EU should be subjected to no more criticism that most other polities. Mény ascribes the ‘pervasiveness of the negative opinion’ about the EU’s democratic credentials to the following: — the democratic deficit is a powerful catchphrase which can easily be manipulated by Eurosceptics (who use it for populist purposes) and Europhiles (who are interested in improving EU efficiency) (Mény 2003: 8). — the tendency to compare the EU to mythical democracy, that is, an idealized theoretical democracy which does not exist empirically (Mény 2003: 8). — the tendency to compare a new democracy in the making and operating at the supranational level to ‘fully-fledged and highly sophisticated democracies that took years or even centuries to reach their present state of development’ (Mény 2003: 10). ‘Despite the fact that the word “democracy” is so broad that the same word is applied to ‘Athenian government, de Tocqueville’s America, to the British or continental parliamentary systems, to the new political systems emerging from the collapse of communism, etc.’ (Mény 2003: 10), the EU is continuously compared to the standards at the national level or rather ‘those mythical rules and principles which are supposed to prevail’ (Mény 2003: 10). As is evident, Mény is rather in agreement with Moravcsik although he introduces a new explanation (similar to that offered by Sbragia) that because the EU is ‘new’ and ‘unusual’ the polity is subjected to harder scrutiny than polities which have a longer history. The recommendation offered by Mény is that ‘we have to change or to adapt expectations, rules and institutions’ and that ‘the worst course would be to replicate through mimicry rules and practices at the national level’ (Mény 2003: 10). This would be ‘a recipe for disillusion as the experience of the EP shows. It is not enough’ (Mény 2003: 10). What is needed, according to Mény is innovative thinking and new paradigms. However, such ideas have yet to surface. His own recommendation is that democracy should not only rely on the power of the demos but also on liberal components such as checks and balances, division of powers, the rule of law, the social covenant, and so on (Mény 2003: 11). This, however, is a general comment. As regards the EU, Mény holds that the ‘constitutional’ element of democracy is 44

Debating Democracy in the EU

overdeveloped whilst the ‘popular’ element is ‘rather meagre’ (Mény 2003: 11). On the basis of the above discussion, it is possible to identify four paradigms within the literature, each with their own foci, discourse and research questions.11 As will hopefully become clear, the paradigms are not divisions of scholars into four groupings. Rather they are divisions of foci into four groupings meaning that scholars can sometimes work under more than one paradigm.12 Under each paradigm, there is scientific progress and proper communication between scholars (‘normal science’ in Kuhn’s own terminology). However, the paradigms themselves do not interact sufficiently.

2.3. Four Paradigms 2.3.1. The Efficiency Paradigm There are two objects of discussion under the efficiency paradigm. The first is the actual existence of the EU (or in other words European integration), the other the policymaking process in the EU. However, the arguments are based on a similar logic. And in both cases, the primary object of study is policy (including policymaking where the outcome is intended to serve partly as provider of momentum to integration). The efficiency paradigm derives from the second of the two analytical interpretations of the democratic deficit as outlined in the previous section, that is, the overdraft analogy. Under a given set of circumstances (permission granted from the bank), being overdrawn is acceptable. In other words, by using this analogy, commentators can diminish the significance of a perceived lack of democracy. This raises the question whether it is legitimate to suspend democracy temporarily in order to achieve effectiveness. The argument that efficiency can make up for such a (temporary?) loss of democratic legitimacy is partly founded in the assumption that European integration is a response to exogenous factors (such as globalization and interdependence) and that regardless of its shortcomings on the input-side, the EU benefits the European peoples because of the increased capabilities it offers those who have chosen to pool their sovereignty and resources. This way, democracy is enhanced because with the EU acting for them, at least the publics have a greater say in what happens in the world than they would have had without the EU even with ‘flawless’ national democracies. This is the part of the argument 45

Debating Democracy in the EU

that applies to the existence of the EU (and thus also the legitimacy of the integration process itself). But the argument also exists in the debate about the decision-making process in the sense that for instance QMV in the Council can be seen as a way out of the joint decision trap (as described by Scharpf 1988) and thus increases efficiency in decision-making. In both cases, the argument is based on the assumption that the relationship between democracy and efficiency is one of interrelation, and that consequently at certain points in time one will have to be sacrificed in favour of the other or in other words, that there is a ‘tradeoff between effectiveness and democratic participation’ (Wessels 1996: 58). As regards the specific case of the EU, Wessels argues that: As the institutional setting of the EU does not correspond entirely to national models and realities, i.e. its character can be regarded as ‘sui generis’, it should not surprise the observer that democratic legitimacy is not to be found in familiar terms. [. . .] Our mixed polity called the EU is not the ideal but a highly practicable and thus ‘optimal’ form of government. (Wessels 1996: 69)

Most defenders of democracy would probably agree that the Athenian model of direct democracy is not a realistic option and that representative democracy really is the only workable solution. This is certainly the case for a polity the size of the EU. Thus, it is true that to some degree, it must be assumed that there is a trade-off between efficiency and democracy. In fact, representation itself is a sign of this. However, since it is perfectly possible to think of both regimes and examples of policymaking that are neither efficient, nor democratic, the universality of this interrelationship must be questioned. Indeed, some theorists argue that there is no trade-off at all between efficiency and democracy (see Lipset 1969: 80–91; Lijphart 1999: 302).13 Finally, if a decision is made which is deemed efficient but not entirely democratic, the question can be raised whether the decision-makers are really acting as guardians rather than representatives because ‘carried to the extreme, the insistence that substantive results take precedence over process becomes a flatly antidemocratic justification for guardianship, and “substantive democracy” becomes a deceptive label for what is in fact a dictatorship’ (Dahl 1989: 163), or in other words a conversion of ‘functional arguments into normative ones’ (Habermas 1998a, 166).14 To sum up, the efficiency paradigm deals with classical dilemmas in democratic theory: how shall we place the balance between efficiency and democracy (assuming there is one)? Should this balance be the same 46

Debating Democracy in the EU

in any polity and at all times? And more specifically, should familiar democratic standards (as seen in nation states) apply to the EU? And finally, how do we ensure that efficient government does not become an elite-driven guardianship or even dictatorship?

2.3.2. The Vertical Democracy Paradigm The second paradigm focuses on the governing, that is, member state and EU institutions and the relationship between national and EU levels of government. This paradigm analyses the various levels of government to which the populations are subjected and the relationship between these levels, primarily the link between EU and domestic institutions (although other levels such as local, regional, and global are occasionally drawn in). The connection between EU and domestic democracies is interpreted in two ways. In the first interpretation, domestic decision-making processes are seen as contributors to the total input legitimacy of EU decisionmaking (vertical democracy). First of all, there is the crucially important argument made by intergovernmentalists that the EU is still mainly legitimized via the democratic accountability of national governments (Moravcsik 2002: 619). Secondly, although these systems operate far from perfectly, some member states have adopted checks and balances designed to ensure that their national executives do not exceed their mandate at EU level, more specifically, during negotiations in the Council. Such measures for parliamentary monitoring of the executive exist in Denmark, the United Kingdom, Holland, and France (Lord 1998: 54). These bodies are intended to contribute vertically to legitimizing EU decisions as far as the national publics are concerned.15 Hence, in this way these domestic processes arguably contribute to maintaining democratic governance for the publics in question. (Increasingly, scholars are also debating whether a model could be developed which formally involves national parliaments in EU decision-making). The second interpretation of the link between national and EU democracy is not quite so optimistic. According to this school of thought, European integration has affected national democratic processes to the extent that some have suggested that European integration perverts national democracy. This is because elements in the EU’s legislative process, particularly Council negotiations and Comitology, allow actors within national executives to escape democratic controls (Mancini 1998; Wincott 1998). This position is supported by Schmitter who argues that there is a risk 47

Debating Democracy in the EU

that the shift of functions to and the increase in supranational authority of the EU has contributed to a decline in legitimacy of domestic democracy because it calls into question whether national officials are still capable of responding to the demands of their citizenry. According to Schmitter, this ‘combined with the reputation of EU institutions as having a lack of accountability could mean that democracy in Europe as a whole is in crisis (it is clear that the strength of this argument will vary between different member states depending on national systems)’ (Pollack et al. 2000: 2). Although Moravcsik agrees that proof of a decline of parliaments exists in domestic European politics, he maintains that in most cases this has nothing to do with the EU and that the EU executive continues to derive its powers via the Council (Pollack et al. 2000: 8). Elsewhere, Moravcsik argues that transparency at EU level is insured due to the constant ‘scrutiny from 15 different governments’ which ‘renders the EU more transparent and less corrupt than almost any national government in Europe’ (Moravcsik 2001: 115). However, this argument does not address the potential loss of transparency suffered by many national parliaments and hence national demoi. The fact that governments scrutinize the EU will only benefit the publics if it is in the interest of the national executives to reveal to the publics what they found during their ‘scrutiny’. Thus, Dehousse, like Wincott, argues that European integration has favoured national executives and the judiciary and that parliaments have lost ground which cannot be compensated for via the granting of increased power to the EP (Dehousse 1997: 53).16 Like Wincott, Dehousse subscribes to the argument that executives have escaped national parliamentary control due to the complexity of decision-making at the European level. Not to mention the fact that many Council decisions are taken by COREPER and therefore never get to the democratically elected representatives. (Dehousse 1997: 43). Measures have been taken, however, to improve Council accountability. These include increased public access to Council documents as well as public access to the decision-making process (Hayes-Renshaw and Wallace 2006: 123–5). However, it is debatable how much real transparency these measures have provided. Briefings do not include nuances or political detail (Hayes-Renshaw and Wallace 2006: 125). From the perspective of this book, there is no crucial need to ‘settle’ whether the arguments associated with Wincott and Dehousse are the most convincing or whether Moravcsik’s point is the more persuasive. The point of including the arguments is to illustrate that there is an ongoing scholarly debate about vertical democracy in the union. This debate 48

Debating Democracy in the EU

is also related to the governance debate because both are concerned with the interplay and distribution of powers between different levels of authority.

2.3.3. The Horizontal Democracy Paradigm To those who credit the EP with sufficient legitimacy, the alleged loss of national parliamentary control can be compensated for by increasing the powers of this institution. Such a step then becomes a matter under the horizontal democracy paradigm because it invariably increases the powers of the EP relative to the other EU institutions. However, the traditional argument under the horizontal democracy paradigm, also known as the ‘conventional definition’ of the democratic deficit (Wincott 1998: 414), goes that the EP has insufficient powers, because it does not hold the executive to account. This argument is based on an analogy to the nation states. And the central question under the horizontal democracy paradigm is whether it is appropriate to apply this analogy to the EU. For those who think so, increasing the powers of the EP is one way of increasing the EU’s democratic legitimacy, not least if this would mean a decrease in the powers of the European Commission, which has often been seen as lacking legitimacy due to its formal monopoly on agenda setting. Since Commissioners are appointed rather than elected, the commision has been accused of having a large democratic deficit (Peters 1992: 99). Peters argues that if the EU is to become ‘a genuine political entity, better mechanisms of popular and political accountability need to be articulated for the Commission’ (1992: 99) and suggests that the president should appoint the Commissioners (1992: 100). In fact, this wish has now partly been fulfilled. Article 215 of the Treaty of Nice gave the president of the Commission formal involvement in choosing the College of Commissioners: The Council, acting by qualified majority and by common accord with the nominee for President, shall adopt the list of the other persons whom it intends to appoint as Members of the Commission, drawn up in accordance with the proposals made by each member state. (Treaty of Nice, article 215)

In the Constitution this role was upheld. In addition, it would be the prerogative of the president to appoint two vice-presidents from the members of the Commission (2004/C 310/01). Hix proposes a more drastic change, namely a transformation of the EU system into a partly presidential one with an electoral college of MPs and 49

Debating Democracy in the EU

MEPs electing the president of the Commission. This way the electorate will have somebody to ‘throw out’ (Hix 1998: 35). This proposal does address the problem that the EP cannot dismiss the executive. Normally, this has been regarded as a particular problem in relation to the Commission for reasons stated above, whereas many scholars have agreed that the Council was legitimized via the national mandates of its parts. However, Lord offers a range of reasons why the Council cannot be adequately authorized through its parts:

r r r

r

r

the practice of extensive policy delegation within the Council itself to a European level technocracy of permanent representatives and working parties of civil servants; bargaining norms that actually require member governments to show flexibility in their preferences, to justify their positions to one another, and to adjust them if found wanting; the substitution of problem-solving bargaining style—in which the aim is to reach a viable solution for the whole—for a strategic one in which the main goal is to decide on distribution between the parts; extensive opportunities for policy learning at the European level itself, including the probability that governments will only complete their preference formation once they have had the opportunity to articulate their positions and cope with tricky problems of preference interdependence (where it is impossible for any one actor to know what is rational until the wants of the others are also known) (Lord 1998: 32–3). Finally, underlying all of the above is the knowledge of each government that it is engaged in a ‘repeat game’ and would like to be seen as a reliable partner (Lord 1998: 33).

In recent treaties, amendments have been implemented that increase the powers of the EP. First of all, the areas in which the co-decision procedure applies have been expanded. In the Treaty of Amsterdam, the EP was given an effective veto over much of the new legislation in that the use of the co-decision procedure was expanded to cover thirty-seven areas (Nugent 2003: 347). This means that most EU legislation, with the exception of agriculture, justice and home affairs, trade, fiscal harmonization, and EMU issues, is now subject to co-decision (Nugent 2003: 347).17 In the subsequent Treaty of Nice which entered into force on 1 February 2003, co-decision was extended to another seven areas (Nugent 2003: 348) and 50

Debating Democracy in the EU

the EP was granted the right to challenge the legality of an act before the ECJ and was given equal status with the Council, Commission and member states in being able to obtain an opinion from the ECJ on the validity of international agreements. (Nugent 2003: 88).18 However, there may be reasons why increasing the EP’s powers does not solve the democracy problem:

r

r r r

Participation in European elections is relatively low. In the 1999 election, average turnout was 49.8 per cent ranging nationally from 91.0 in Belgium to 24.0 in the UK (Nugent 2000: 1; Decker 2002: 261). In 2004 the average turnout was 45.5 (European Parliament website).19 National electorates consider EU elections ‘second order’ and EU elections are often more about national issues (Anderson and Goodman 1999: 27; Nugent 2000: 1; Decker 2002: 261). The ratio of MEPs is not proportional to the size of the populations in the member states. The non-existence of a European demos also known as the sociopsychological dimension of the EU’s democratic deficit (Chryssochoou, Stavridis, and Tsinisizelis 1998: 114).

Thus, because the conventional definition is based on an analogy to nation state division of powers, it may be that the wrong analogy has been used. If this is the case, this could mean that increasing the EP’s powers would in fact reduce the democratic legitimacy of the union. Especially, if participation in EP elections remains low and if a European community fails to materialize. It is the latter of these arguments that is investigated by the fourth paradigm.

2.3.4. The Socio-Psychological Paradigm In the final paradigm, the primary focus is the governed rather than the governing. (The governed meaning the citizens of those nation states that are members, or about to become members, of the EU).20 Going back to the elements of democratic legitimacy graphic, what is essentially being questioned under this paradigm is whether EU citizens have a common identity combined with the element of the authority-part which deals with the ‘consent to the overall structure of the power relations’ otherwise known as ‘legitimacy as a category of formal/legal validity’ (see Figure 2.1: 34). In other words, it is not the operational legitimacy of EU institutions 51

Debating Democracy in the EU

and leaders that is being questioned but their roles or even the right to exist of those roles: For unless a coherent identity presides over the process of European integration, the process will, sooner or later, lead to disorder. The habits and attitudes required to sustain new European institutions depend, finally, upon some shared beliefs. (Siedentop 2000: 189)

The point of departure is that democracy cannot exist without a people, or put another way, if there is no collective self, there can be no collective self-determination.21 Thus, it is logically impossible to imagine a democracy without a demos. However, it is not logically impossible to imagine a polity without a demos. In fact, it could be argued that the EU is a polity with no demos but with a number of demoi (Chryssochoou 2001). This does not mean that the EU cannot be democratic, but it does limit the range of options held by those in charge of the institutional (re)design and it also limits the scope for widening the EU’s influence into other policy areas. In a recent monograph Majone explains why this is the case: Nonhomogeneous polities find it particularly difficult to pursue redistributive and other policies with clearly identified winners and losers. Redistribution of income and wealth is a zero-sum game since the gain of one group in society is the loss of another group. Public decisions concerning redistribution can only be taken by majority vote because any issue over which there is unavoidable conflict is defeated under more inclusive rules. It follows that in a transnational federation all such policies would have to be excluded from the public agenda as being too divisive. This is, in fact, the main reason why redistributive social policy plays such a small role even in the present EU. (Majone 2005: 190)

Moravcsik also contends that the EU ‘lacks the grounding in a common history, culture, discourse and symbolism on which most individual polities can draw’ (Moravcsik 2002: 604). However, Moravcsik does not see this a problem because of his insistence that the EU is legitimized via the domestic democratic nature of its member states which eliminates the need for a demos at the EU level. In a similar vein, Lord argues that no primordial European identity exists on which the EU can draw (Lord 1998: 114). However, the fact that there is no European identity as such does not mean that there is nothing to build on because ‘although it is most unlikely that the Union could function as an ethno-cultural unit, it could conceivably construct a demos around shared civic values’ (Lord 1998: 321). 52

Debating Democracy in the EU

Similarly, Scharpf argues that: Within established nation states, where the sociocultural preconditions of collective identity are more or less taken for granted, these considerations might seem more or less academic. For the European Union, however, they explain the concern over a ‘democratic deficit’ that persists, and even increases, though the competencies of the European Parliament have been significantly enlarged by the Single European Act and by the institutional changes adopted in Maastricht and Amsterdam. Given the historical, linguistic, cultural, ethnic, and institutional diversity of its member states, there is no question that the Union is very far from having achieved the ‘thick’ collective identity that we have come to take for granted in national democracies—and in its absence, institutional reforms will not greatly increase the input-oriented legitimacy of decisions taken by majority rule. (Scharpf 1999: 9)

Without these social structures, even fair and open elections (to the EP) will not be sufficient, as elections will only work as a legitimizing force if they are embedded in a wider context of societal and political structures and practices, including checks and balances among different branches and levels of government, enforceable guarantees of free communication and association, and the de facto existence of a wide range of intermediary associations, competitive political parties, and inquisitive and credible media of mass communication (Scharpf 1999: 14). Furthermore, even if all these ‘intermediary associations’ were to be put in place maybe in the shape of the formalization of European-wide parties (which is in itself a long shot), this may not even be sufficient. Even if, from an institutional (procedural) point of view, a political system has all the characteristics of a perfect democracy it can still lack legitimacy (Decker 2002: 263). Thomassen argues that the decline in power of national parliaments due to the transfer of powers from the member states to the EU has not been fully compensated for by a simultaneous increase in the powers of the EP. However, the question, from the perspective of the demos, rather than a weakened legislature, is whether such an increase in EP powers could compensate for the loss of power suffered by national parliaments. If there is no European demos an increase in the powers of the EP at the cost of the Council would, in fact, increase the democratic deficit rather than address it. As Thomassen rightly points out, rather than increasing the legitimacy of the EP the direct elections introduced in 1979 have revealed some ‘fundamental problems in the development of something like a representative democracy in the EU’ (Thomassen 2002: 15). 53

Debating Democracy in the EU

Hence, the debate about the democratic deficit ‘never raged as much at it did after the election of MEPs by universal suffrage’ (Mény 2003: 8). Obviously, the natural question to ask at this stage is whether it would be possible to create a demos, as surely, if an EU people came into being the socio-psychological problem would be solved. This issue is discussed at length in Chapter 3. To sum up, although there seems to be a consensus within the paradigm that there is no such thing as a European demos at present, the unsettled questions are: (a) whether this matters, (b) whether it would be possible to create an EU demos and (c) whether or not such a development would be desirable.

2.4. Conclusion The Table 2.1 below sums up the paradigms discussed in the previous sections. Rather than a division of scholars into sub-groupings, this table divides the main lenses of the debate into sub-groupings. The table must be thought of as an abstract illustration as in real life some scholars take part in more than one of the four debates. But although scholars can move between debates, the debates themselves do not communicate effectively with each other. For instance, it is pointless to discuss the potential democratic benefits of an increase in the powers of the EP (an argument under the horizontal paradigm) with somebody who believes that the EU has no demos (an argument under the socio-psychological paradigm). However, a discussion—say between Moravcsik and Wincott—as to the relationship between EU and member state governance is feasible because both are analysing the same relationship although they reach different conclusions. Using Kuhn’s terminology (1962), this is a situation of ‘normal science’ whereas the relationship between the four debates resembles that of a ‘pre-paradigmatic situation’. What the paradigms really show is an apparent communication breakdown which has arisen not only because of different approaches but is also due to the lack of shared definitions of the two objects of study, that is, the EU and democracy. Hence, this chapter has shown that rather than one debate about EU democracy, there are four different debates taking place alongside each other caused by the four paradigms (i.e. lenses). In a worst case scenario, scholars could continue exchanging ideas and producing knowledge within their own research community which—although relevant to 54

Table 2.1. Four concurrent paradigms analysing democracy in the EU The efficiency paradigm

The vertical democracy paradigm

The horizontal democracy paradigm

The socio-psychological paradigm

Main object of analysis Questions

Policy (including integration process) What is the relationship between democracy and efficiency in the EU? And to what extent, if any, are they related?

The governing

The governing

The governed

What institutional measures should be taken at EU level in order to fix imbalances in the distribution of powers?

What (if anything) does the present lack of a European people mean for the democratization of the EU?

Intra-paradigmatic debate

One argument goes that efficiency and democracy are interrelated so that a high degree of output will increase overall legitimacy despite affecting democracy negatively. Others deny this and argue that a high output necessitates a high input in order for governance to be democratic.

What is the relationship between EU and member states’ domestic political processes and how does this affect democracy? Some argue that the democratic legitimacy is ensured via the Council. Others argue that the EU has negatively affected democracy within member states.

Within the paradigm authors debate how institutional change at EU level can improve democracy (assuming that it can). The debate is essentially about the future design of the EU system and the division of powers between EU institutions.

One argument goes that the EU can never be democratic without a demos, the counter argument being that the EU should not attempt to develop a demos (this could even be risky)

55

Debating Democracy in the EU

Paradigm

Debating Democracy in the EU

all debates—never really reaches the other debates due to barriers caused by different foci or methods. Thus, used in its sociological sense the term ‘paradigm’ is useful in order to explain this situation and thus possibly also to resolve it. This is not to say that the term ‘paradigm’ equates ‘debate’. Rather it denotes the ‘matrix’ that determines what variables are brought into each debate and thus also the methodological barriers that prevent influences from other concurrent debates. For instance, even though most scholars probably subscribe to the lowest common denominators for a democracy as quoted in this chapter, they do not place equal emphasis on each element. But even if they did, they might not feel that these all apply to the EU. Thus, different scholars identify different problems and thus look for different, sometimes even contradictory, solutions. But no satisfactory outcome will be produced by analyses of selected parts in isolation. The position taken here is that the socio-psychological debate must take precedence over any other debate. The debate about efficiency and its relationship with democracy is important but for now it is more important to settle ‘who is to decide what is to be done’ than simply ‘what is to be done’. And this decision can only be made once the issue of the demos has been settled, that is, once it has been decided what limitations the lack of a European people put on the scope for institutional design that is available to the EU. Also, any analysis of horizontal and vertical democracy will have to take socio-psychological aspects into consideration in order to form a realistic picture of what can be regarded as sustainable. The discussion under the horizontal paradigm is after all a debate about the balance of power between the EU institutions. By having identified the four paradigms, it is hoped that this chapter can contribute to the cross-paradigmatic communication that is needed if scholars are to start looking at the whole picture. In the following chapter, the argument will be made that the socio-psychological paradigm asks the most fundamental of the questions about democracy in the EU.

Notes 1. The analogy was originally used to describe the study of the EU whereas Lord uses it to describe ‘the study of the so-called deficit’ (Lord 1998: 11). 2. Other reasons for the increasing concerns about democracy are the inclusion of new members with strong traditions of direct and popular government (such as Sweden), the planned deepening to include social, monetary, foreign, and regulatory policy reins (Moravcsik 1998: 42), and the increased use of majority voting in the Council (Decker 2002: 259).

56

Debating Democracy in the EU 3. Of course, classifying Sweden and France as states is by no means equivalent to having answered the question of what constitutes a state, only that whatever a state is, Sweden and France are both examples of one. 4. The EU is un object politique non-identifiè (Schmitter 2000b: 344). 5. This causes a problem because of the inevitable tendency to combine the definitional stages (what is democracy?) with concerns over what is realizable. However, in discussions on democracy these two stages should not be mixed because it opens the floodgates and the ‘task of defining democracy becomes ridiculously subjective’ (Saward 1994: 16). 6. Although there are indications that faith in national political systems has declined, it is still a minority of people who question the right to exist of the nation state. In fact, loss of faith in a political system does not have to correlate with loss of belief in the polity at all. However, because the EU is a relatively new polity, lack of faith in its political system causes a doubt in the right to exist of the actual polity. Especially because the institutions have not succeeded those of another system but are supposed to complement another system (Sbragia 2002: 5). Likewise, a doubt in the polity’s right to exist may increase the scrutiny of the political system which again can lead to additional criticism. 7. Weber distinguishes between three ‘pure types’ of legitimate authority: legal, traditional, and charismatic legitimacy: Legal legitimacy is based on rational grounds and rests on a belief in the ‘legality’ of patterns of normative rules and the right of those elevated to authority under such rules to issue commands. Traditional authority, however, rests on an established belief in the sanctity of immemorial traditions and the status of those exercising authority under them. Finally, charismatic authority rests on devotion to the specific and exceptional sanctity, heroism or exemplary character of an individual person and of the normative patterns or order revealed or ordained by him (Weber 1968: 212). 8. Admittedly, this is true only if the observer sets out with the aim of investigating an already reputed deficit. Had he or she never heard of the term but just decided to evaluate democracy in the EU, this kind of intuitive reaction to an already established term would obviously not have occurred. 9. By adding ‘equal’ to ‘representation’ we have now—substantially—changed the definition of democracy. An example of an operational consequence of this is that it would impact on how much legitimacy one subscribes to networks. Although these are channels of representation they do not provide equal representation. In fact, quite the contrary. ‘Networks are sectional interests with needs and motivations of their own. Access to them is governed by ownership of resources, not by possession of an opinion worthy of equal consideration’ (Lord and Beetham 2001: 455). In a similar vein, Hix argues that ‘. . . access of all interest groups to EC policy channels is far from equal,

57

Debating Democracy in the EU

10.

11.

12.

13.

14.

15.

16.

58

despite the Commission policy of subsidising non-economic interests’ (Hix 1994: 13). It is true that there is not absolute consensus in the literature as regards who exactly constitutes the EU Executive. However, the above critique applies both to the Council and the Commission. Kuhn, in his own admission, uses the term paradigm in a number of ways and has been on the receiving end of hard criticism. (See Ball 1976 for an account). However, used in the sociological sense, i.e. to mean ‘the entire constellation of beliefs, values, techniques, and so on shared by members of a given community’ (Kuhn 1996: 175), the term is helpful when it comes to describing the EU democracy debate. This is consistent with Kuhn’s usage of the word: ‘Usually, individual scientists, particularly the ablest, will belong to several such groups either simultaneously or in succession’ (Kuhn 1996: 178). For instance, Lijphart states that ‘. . . contrary to the conventional wisdom, there is no trade-off at all between governing effectiveness and high-quality democracy—and hence no difficult decisions to be made on giving priority to one or the other objective’. (Lijphart 1999: 302). Habermas provides the following as an example of turning a functional argument into a normative argument: ‘The governments of Malaysia, Taiwan, and China appeal to a “priority” of social and cultural basic rights in an effort to justify the violations against basic legal and political rights of which the West accuses them. These dictatorships consider themselves authorized by the “right to social development”—apparently understood as a collective right— to postpone the realization of liberal rights and rights of political participation until their countries have attained a level of economic development that allows them to satisfy the basic material needs of the population equally. For a population in misery, they claim, legal equality and freedom of opinion are not so relevant as the prospect of better living conditions.’ (Habermas 1998a: 166). For instance, in Denmark the European Affairs Committee consists of members of parliament who must be consulted by ministers before they attend Councils in Brussels. The minister must obtain a mandate for negotiation. This means that a majority of the Committee members must not be opposed to the minister’s position as regards a given proposal on the Council’s agenda. (The EU Information Centre of the Danish Folketing 2003: 2). It is important not to confuse the ‘perversion of national democracies’ argument with the argument that European integration has weakened the nation states. A ‘strengthened state’ and a ‘weakened democracy’ are not mutually exclusive. On the other hand, if a state is externally weakened to the extent that it has lost large amounts of sovereignty, even a flawless domestic democracy may not be able to secure the rule of the people. This would depend whether one argues that it is still the people who rule under the externally given

Debating Democracy in the EU

17.

18.

19. 20.

21.

circumstances or that the people are ruled by the externally given circumstances. In any case, one could raise the question of whether this situation is particularly new. Dahl argues that ‘. . . we need to remember that the autonomy of the city-state and the sovereignty of the national state were always less fact than fiction. International conflicts, rivalries, alliances, and wars have eternally demonstrated how much the autonomy of all states, democratic and nondemocratic, has been radically incomplete. Not just conflict but also trade, commerce and finance have always spilled over state boundaries. Democratic states, therefore, have never been able to act autonomously, in disregard of the actions of outside forces over which they had little or no control’ (Dahl 1989: 319). This view is backed by Krasner, who argues that: ‘There has never been some ideal time during which all, or even most, political entities conformed with all of the characteristics that have been associated with sovereignty— territory, control, recognition, and autonomy. Alternative principles—notably human and minority rights, fiscal responsibility and international security— have been used to challenge autonomy. In the absence of any well-established hierarchical structure of authority, coercion and imposition are always options that the strong can deploy against the weak’ (Krasner 1999: 238). The Amsterdam Treaty greatly increased the scope of co-decision by extending its coverage to most of what had so far been covered by the cooperation procedure. But it also revised the third reading stage so that should the Council and EP fail to agree on a text in the conciliation committee, the Council may no longer (as it could under the original version of co-decision) act unilaterally by going back to the common position it had agreed upon before the conciliation procedure and adopt the act by qualified majority (Dinan 1999: 286). The Constitution contained further empowerment of the EP, most notably the expansion of the use of co-decision (named the ‘ordinary legislative procedure’) which would become the default decision-making procedure for internal policies (excluding matters of external relations such as the common foreign and security policy). http://www.elections2004.eu.int/ep-election/sites/en/results1306/turnout_ep/ index.html The citizens of applicant countries can be seen as being at least semi-governed by the EU because of the conditions that they have to satisfy prior to membership, such as standards for economic performance or the Copenhagen criteria (see also Wæver 1997: 334). See Chapter 3 of this thesis and Weiler, Haltern, and Mayer (1995) for a thorough discussion on the ‘no-demos’ thesis.

59

This page intentionally left blank

3 The demos

3.1. Introduction In this chapter I elaborate on the conclusion of Chapter 2, that the most fundamental question regarding democracy in the EU is the one identified by the socio-psychological paradigm. Hence, the outcomes of the debates within the three other paradigms are reliant on the answers to the sociopsychological question. Should the EU be ruled by a group of national demoi or by one European demos or maybe even a combination of the two—if indeed such a combination is feasible? From this perspective, it is imperative to clarify from what sources it is intended, and from where it could realistically be expected, that the EU should derive its democratic legitimacy. Thus, the overall aim of the present chapter is to explain why the first step towards democratization of the EU must be to look at the issue of the demos. The chapter opens with a discussion of the term demos based on classical democratic theory as well as contemporary literature. This is followed by an evaluation of the claim that a demos is a prerequisite for democracy. And finally, the demos in the context of the EU is analysed from a theoretical perspective. The theoretical analysis has three different parts, these being the descriptive part (does an EU-wide demos exist?), the predictive part (could it be made to exist?), and the normative part (ought an EU demos to exist?). It will be argued that at present there is no EU demos, that creating a demos based on a common cultural identity is neither possible nor desirable, that we need to distinguish between the legal and the social dimensions of the demos in order to better analyse these difficult concepts, and that by developing a more meticulous definition of the demos it may be possible to construct a minimal demos sufficient to justify a certain amount of supranational governance. Having said that,

61

The Demos

the extent to which the EU requires a demos depends on the model of democracy which is eventually chosen for the polity. A detailed explanation for the latter of these conclusions is provided in Chapter 4. At this stage, the first step is to clarify what is to be understood by the term demos itself.

3.2. Rule by the People The Greek term ‘demokratia’ is composed of two words, demos (people) and kratia (rule).1 Consequently, demokratia—or democracy—translates into English to mean ‘rule by the people’.2 However, this only amounts to a word-word definition as opposed to a word-thing definition (see Sartori 1987: 7) in that it raises two central questions, namely what does it mean to rule? and who are the people (Dahl 1989: 3). In contemporary democratic terminology, to rule means to govern. The way to understand the word ‘people’ in this context is as in ‘a people for political purposes’ rather than a nation (ethnos). Because these issues (particularly the ‘who are the people’ question) are central to my argument, they are addressed in detail in the following paragraphs.

3.2.1. What Does It Mean to Rule? In contemporary terminology the verb ‘to rule’ has been replaced by the word ‘to govern’ which in the present tense (first person singular and plural) remains unchanged and in the past tense becomes ‘governed’.3 In this context, the corresponding noun (government) refers to the undertaking or activity that is government as opposed to the group of individuals that constitute an executive. In this interpretation, democracy in modern times means government by the people. However, in recent years another term has surfaced which so far does not have a corresponding verb of its own and hence ‘borrows’ its active use from government. This is the term ‘governance’—a term which despite being extremely popular is often used very imprecisely (Jordan 2001: 198) and with ‘too many meanings to be useful’ (Rhodes 1997: 15). Frequently used in business studies, governance has been incorporated into the language of political science to describe elements of policymaking that are not captured by the ‘government’ framework. There exists a rather substantial literature on the difference between government and governance and I address this issue relatively briefly. 62

The Demos

Modern policymaking has been described as ‘governance without government’ (Rosenau and Czempiel 1992; Rhodes 1997). This phenomenon describes both the decentralization of government to supranational levels (such as the EU) and to sub-national levels of authority (regional or local authorities) and also the involvement of non-governmental actors in policymaking. The latter applies mainly to the literature on networks, which is addressed in Chapter 4. According to Rosenau, governance is a more encompassing phenomenon than government because in addition to covering governmental institutions it ‘also subsumes informal, non-governmental mechanisms whereby those persons and organizations within its purview move ahead, satisfy their needs and fulfill their wants’ (Rosenau 1992: 4). Hence, governance covers ‘a change in the meaning of government, referring to a new process of governing’ (Rhodes 1997: 15) and to ‘self-organizing interorganizational networks characterised by interdependence, resource exchange, rules of the game and significant autonomy from the state’ (Rhodes 1997: 15). According to Webster’s Encyclopedic Dictionary, governance can be defined as the exercise of control whereas government means the exercise of authority over a political unit or ‘the making of policy as distinguished from the administration of policy’ (Webster’s New Encyclopedic Dictionary 1995: 433). Despite the unavoidable simplification and analytical limitations inherent in a dictionary entry, this distinction is rather useful because it explains why governance can be applied rather more broadly than government. For instance, in business studies there exists a literature on ‘corporate governance’ whereas there is no such literature on ‘corporate government’. Nevertheless, the word governance covers the involvement of actors outside the traditional governmental framework. Very often, governance will include involvement of interest groups and cover decision-making processes or elements of decision-making processes that include actors other than elected representatives.4 As a result, governance can be understood as an analysis of different levels of decision-making and—according to some schools of thought—the diminishing role of the state in policymaking resulting from the transfer of elements of government from the national to both the sub-national and the supranational levels (Rhodes 1997; Jordan 2001; Hooghe and Marks 2003). Essentially, this literature stresses that the ‘idea that national governments are the major actors in public policy and that they are able to influence the economy and society through their actions now appears to be in doubt’ (Peters and Pierre 1998: 223). Hence in recent decades the ‘traditional conceptualization of the public sector’ has come under 63

The Demos

increasing strain (Peters and Pierre 1998: 223). According to these authors the strain has two sources, that is, pressures from the international environment, including through supranational organizations as the EU and secondly from the changes in the relationship between government and the private sector (Peters and Pierre 1998: 223). Both these influences serve to diminish the influence of the state. Hence the authors argue that: In the governance arguments the State does not become totally impotent; rather, it loses the capacity for direct control and replaces that faculty with a capacity for influence. (Peters and Pierre 1998: 226)

Governance has become a widely used term and its framework is used to describe policymaking in a range of contexts, for example national, European, and global. However, in addition to functioning as a description of modern policymaking (or rather a range of descriptions) the term is also used normatively. For instance, Hooghe and Marks state that ‘Modern governance is—and according to many, should be, dispersed across multiple centers of authority’ (2003: 233). The reasons, according to these authors, are primarily pragmatic: Centralized government is not well suited to accommodate diversity. Ecological conditions may vary from area to area. Controlling smog in a low-lying flat area surrounded by hills (such as Los Angeles) poses a very different policy-problem than smog control in a high plateau such as Denver. Preferences of citizens may also vary sharply across regions within a state, and if one takes such heterogeneity into account, the optimal level of authority may be lower than economies of scale dictate. In short, multi-level governance allows decision makers to adjust the scale of governance to reflect heterogeneity. (Hooghe and Marks 2003: 236)5

The authors argue that it is possible to divide modern governance into two different types which they name simply Type I and Type II. Type I describes jurisdictions with a limited number of levels. What this means for the citizens is that each is located in a ‘Russian doll set of nested jurisdictions with only one relevant jurisdiction at any particular territorial scale. These territorial jurisdictions are intended to be, and usually are, stable for periods of several decades or more, though the allocation of policy competencies across jurisdictional levels is flexible’ (Hooghe and Marks 2003: 236). Type I jurisdictions are designed around human and usually territorial communities (Hooghe and Marks 2003: 241). Thus: Disputes about Type I jurisdictional boundaries usually cannot be settled by comparative evaluations of the efficiency of competing jurisdictional arrangements in

64

The Demos providing public goods but involve contending conceptions of community as well. Type I jurisdictions are often rooted in communal identity. (Hooghe and Marks 2003: 240)

Type II jurisdictions, on the other hand, are task driven and Type II governance is thus fragmented into functionally specific pieces that solve for instance a particularly local problem. The number of such jurisdictions is potentially huge and there is no great fixity of their existence (Hooghe and Marks 2003: 237). Both types, however, emphasize the diminishing role for central government in policymaking. Having studied this framework, it is tempting to conclude that the EU satisfies the criteria of Type I (except perhaps the community dimension). However, as the EU could be seen as primarily task- rather than community-driven, the truth is perhaps that the EU is a Type II jurisdiction moving towards becoming a Type I jurisdiction. This does not mean that the EU is becoming a state modelled on a nation state institutional architecture. Rather, it means that if political integration proceeds the EU will move towards being ‘general purpose’, that is, involved in an increasing number of policy areas rather than an instrument for handling specific tasks such as matters related to the single market or environmental issues. It may even be possible to argue that Type I is an input oriented regime whereas Type II governance—being task-driven—is mainly output oriented. Hence, Type I governance derives legitimacy from a territorial— perhaps even identity-driven—source of democratic input whereas Type II focuses on problem solving. This would mean that under Type II, legitimacy was sought in the delivery of ‘good policy’ whereas under Type I, the territorial or community dimension would provide legitimacy. There may be a difference when it comes to the legitimacy of national governance and European governance. Sbragia offers a useful distinction, arguing that governance in national states can be thought of as ‘government plus’ whereas governance in the EU is ‘government minus’: In national capitals, the term ‘governance’ can be thought of as ‘government plus’—government plus networks of experts, non-governmental groups, professional groups, business groups, labor unions, environmentalists, feminists, etc. The legitimacy of this ‘government plus’ is essentially unquestioned. In Brussels, however, ‘governance’ is ‘government minus’. Therefore, issues such as transparency and accountability become particularly salient (whereas in national capitals the existence of ‘government’ legitimates much that is opaque and secretive). (Sbragia 2002: 6)

65

The Demos

According to Sbragia most of the literature and debates about governance are based on certain core premises, none of which apply to the EU. Those basic assumptions include: (a) the existence of an elected central government with legitimate constitutional power recognized by both subordinate governments within its territory and the international system of which it forms part, (b) the incorporation of society through regular elections as well as a range of other mechanisms (political parties, interest groups, corporatist structures, etc.) which vary widely across nation states, and (c) ‘the existence of a national identity which may be contested by several smaller groups within the nation state but which is nonetheless accepted by the majority’ (Sbragia 2002: 2). With respect to governance in the EU, Majone (2005) makes the following statement: Scholars may try to describe the EU as a system of ‘governance without government’, but to citizens and political leaders alike this is only a combination of contradictory ideas, an oxymoron. (Majone 2005: 211)

It is certainly true that in the literature on the EU, the occurrence of the term governance has become rather more frequent than that of the term government although there is one notable exception to this in that Hix uses the term government in his book about the EU as a political system (see Hix 1999, 2005). Having said that, this label is applied to Part I, which describes the institutions. In other parts of this work, Hix uses the term governance frequently. For the purposes of my argument making a choice between governance and government is not important. This is because my focus is on the governed no matter whether they are governed as a result of government or governance. It is, however, recognized that if one were to argue that government and governance were legitimized differently, such a choice would have to be made. For instance, if one could argue that government required a demos whereas governance did not, and that what takes place at EU level is pure governance, and not government, then there might not be a need for a demos. However, the reality is not this simple. Whilst it is obviously recognized that there is no EU government (as in a specific group of individuals who constitute the cabinet with a direct or indirect mandate), there are institutions who undertake the action of government such as agenda-setting, mediating between interests, budgeting, and setting overall policy goals and objectives. In other words, they ‘rule’ even if only in a limited number of policy areas. 66

The Demos

3.2.2. Who Are the People? Traditionally, this discussion has centred on the internal question, that is, which individuals—or groups of individuals—within a given polity were to constitute the demos. In Athenian direct democracy this would mean who would be included—or have the opportunity to be included—in the assembly. In contemporary representative democracies, the operationalization tends to be who has the vote and the right to run for office: From classical Athens to modern times some persons have invariably been excluded as unqualified, and until this century, when women gained the suffrage, the number of persons excluded has exceeded—sometimes as in Athens by a wide margin—the number included. Such was the case in the first modern ‘democracy’, the United States, which excluded not only women and, of course, children, but most blacks and native Americans as well. (Dahl 1989: 4)6

In this sense of the word, ‘demos’ and people are not the same thing because the people could be either all the people living in Athens or all the people living in the United States, whereas the demos would be those people within Athens or the United States who were entitled to take part in the democratic process: Within ‘a people’ only a limited subset of persons is entitled to participate in governing. These constitute the people in another sense. More properly, they are the citizens or citizen body, or as I shall often say here, the demos. (Dahl 1989: 4)

To sum up, in the internal discussion about the demos, the question is, which inhabitants within a given polity are qualified (by right or merit) to take part in the democratic process. In Figure 3.1, this is illustrated by the inner circle. The outer circle, however, illustrates the second way in which we ask questions about the demos. Here, the focus is not on those who have the opportunity to take part in the democratic process. Instead, the circle illustrates the actual group of people from where those individuals are chosen. So rather than asking which individuals amongst a number of persons living within a given space are to take part in the democratic process, we now ask who constitutes the people. Why should Americans constitute ‘a people’ and the neighbors the Canadians and the Mexicans separate peoples? Why should there be a political boundary between, say, Norway and Sweden, or Belgium and Holland, or French-speaking Swiss and French-speaking French? Or put the question another way: are people in local communities within a nation state entitled to a measure of self-government? (Dahl 1989: 3)

67

The Demos

Demos (the People)

Demos (participants)

Figure 3.1. Two questions about the demos Source: This figure and the accompanying analysis was first published in the article ‘A Demos for the European Union’ (2005) Politics, 25(1): pp. 12–18 and has been reprinted with permission from the editors.

The above should serve to illustrate that there are two angles from which we can discuss the demos, the first being who within a given polity should take part in the democratic process, and the second being when exactly can we say that there exists a people. In the first case, we could say that X demos included all men over 25 years of age (inner circle) but we would still have to decide on a group of people from which to select these men over 25 (outer circle). On the basis of the above discussion, it would appear that one could conclude that the people are the group of individuals from amongst whom the demos is chosen. But this is not the definition used in the no-demos thesis. The demos and the people are the same thing, although the word people is used far more generally, both to describe a group of individuals (‘people’) and also to describe a group of individuals who are united due to a common identity (a people). It is when the word people means ‘a people’ that it is the same as a demos—at least in contemporary terminology. (As a consequence, the title of this book could give the false impression that its author believes the EU has a people. This is not the case, ‘the people’ in this context simply refers to the people that live within the union and are governed by its political system rather than referring to a bond between them.) Unfortunately, having established that, under the above circumstances, the words demos and people can be used interchangeably is not the same as having established what exactly constitutes a demos/people. But it is evident that there has to be some kind of ‘glue’ between a group of 68

The Demos

individuals that makes them ‘a’ people rather than just ‘some’ people. One such type of glue could be a feeling of common identity or what Fritz Scharpf describes as ‘thick collective identity’ bound in the feeling of ‘essential sameness arising from pre-existing commonalties of history, language, culture, and ethnicity’ (Scharpf 1999: 8). Similarly, Smith defines collective cultural identity as ‘a sense of shared continuity on the part of successive generations of a given unit of population, and to shared memories or earlier periods, events and personages in the history of the unit’ as well as a ‘shared destiny of that unit and its culture’ (Smith 1992: 58). Although this is often the case, a people does not have to be of the same ethnicity. An ethnic group can be said to exist when ‘a core group holds some of the following attributes: a collective proper name, a myth of common ancestry, shared historical memories, one or more differentiating elements of common culture, an association with a special homeland, a sense of solidarity for significant sectors of the population (Wæver et al. 1993: 31). This ethnicity may or may not coincide with nationhood—although some argue that ethnicity is a prerequisite for nationhood. Whether or not one agrees with this, it is clear that there is often overlapping between a nation, its ethnicity, and a feeling of common identity.7 However, technically, a demos can consist of more than one nation or of a nation plus various individuals or groups of individuals who are not members of that particular nation but who are nevertheless part of the demos. Thus, the demos and the nation will often overlap, although in many polities, the demos will include members from more than one nation.8 Because I use the words demos and people (as in a demos and a people) interchangeably, this also means that a people can consist of a range of nations. An example of such a people would be the British people which includes English, Scottish, and Welsh nationals, albeit sometimes with mixed degrees of enthusiasm.9

WHEN DOES A PEOPLE EXIST? Democratic theory tends to presuppose the existence of a people which is why this literature is not particularly helpful in answering the question of what constitutes a people for political purposes (Dahl 1989: 209). In fact, both the pre-existence of the polity and its inhabitants tend to be assumed although the realities behind such assumptions are often questionable (Dahl 1989: 3). Again, this stems from the fact that we are used to looking at the polity first before discussing or designing its political system, whereas recently, tendencies towards integration (and 69

The Demos

disintegration) have caused a transfer of the focus from the system to its prerequisites, that is geographical borders, ethnicity, identity, etc. Furthermore, when it comes to government, our conceptions of a democratic polity tend to be that of a nation state (Weiler et al. 1995: 28; Decker 2002: 263). Claus Offe is even blunter about this, arguing that: ‘The only precondition worth keeping in mind is that in order to have a democracy, one needs a state first’ (Offe 2000: 4). One of the reasons that the EU is so complicated is precisely that it is not state and as such does not have an ideal to measure itself against. However, this gap in the literature is slowly being filled by new writings on democracy and in recent years, a number of scholars have addressed the issue of the demos in the context of the EU.10 Weiler et al. summarize the essence of the demos by stating that: ‘The subjective manifestations of peoplehood, of the demos, are to be found in a sense of social cohesion, shared destiny and collective self-identity which, in turn, result in and deserve loyalty’ (Weiler, Haltern, and Mayer 1995: 10). Thus, in order to earn the label demos (a people) the inhabitants of a polity must to some extent share a common identity.

WHY THE DEMOS IS A PREREQUISITE FOR DEMOCRACY The short answer is that because democracy means rule by the people it cannot exist without a people. If there is no people who can rule, there is no basis for democracy, or put another way, if there is no self, there can be no self-determination. Thus, it is logically impossible to imagine a democracy without a demos. However, as mentioned elsewhere, it is not logically impossible to imagine a polity without a demos. And it can be difficult to distinguish between a group of people and a people, which means that scholars may not always agree whether a demos exists in a particular context. This is especially true because the concept has traditionally been linked so closely to that of identity, which, like so many others in the social sciences, is extremely difficult to measure empirically. The axiom that a democracy cannot exist without a people holds true regardless of one’s chosen definition of democracy. However, there are also a number of normative reasons why democracy cannot exist without a demos, most of them related to the inevitable unfairness of majoritarian governance suffered by the minority in a demos-less polity: There is a very simple reason why a sense of shared political identity is so important to the formation of a democratic political system: members of a democracy

70

The Demos have to accept the decision of some kind of majority as collectively binding on all. (Lord 1998: 107)

Similarly, Chryssochoou argues that there exists a wider assumption that . . . democracy, in the form of representative and responsible government, presupposes the existence of a demos, as the necessary popular infrastructure upon which its basic properties, i.e., adherence to and acceptance of majority rule— would apply in a given political community. (Chryssochoou 2001: 251)

This argument is backed by Weiler: Citizens constitute the demos of the polity—citizenship is frequently, though not necessarily, conflated with nationality. This, then, is the other, collective side, of the citizenship coin. Demos provides another way of expressing the link between citizenship and democracy. Democracy does not exist in a vacuum. It is premised on the existence of a polity with members—the demos—by whom and for whom democratic discourse with its many variants takes place. The authority and legitimacy of a majority to compel a minority exists only within political boundaries defined by a demos. Simply put, if there is not demos, there can be no democracy. (Weiler 1998: 237)11

If a political system has no demos, the decision-makers are not equal to those who have to abide by the decisions (ref. Habermas’s definition of the democratic deficit, i.e. ‘whenever the set of those involved in making democratic decisions fails to coincide with the set of those affected by them’ as described in Chapter 2) and this contributes to the lack of legitimacy, which may eventually cause the downfall of the system. Without such a sense of identity ‘the legitimacy of the unit will be contested, however impeccable its procedures’ (Lord and Beetham 2001: 444). Having said that, the necessity of a demos in a political system would depend on what kind of decisions the political system makes (i.e. different types of legitimacy for an international organization and state). But there may also be variations between for instance a parliamentary and a consociational system. This is investigated in Chapter 4. For now, it is sufficient to conclude that any type of majoritarian-based government would need a demos for such a system to be democratically legitimate. Thus, although a demos is a prerequisite for a democracy, the extent to which a demos is a prerequisite for democratic decision-making will depend on how it is intended that the making of certain decisions is going to be legitimized. In other words, a system such as the EU could—up to a point—be legitimized via national demoi. Majone (2005) makes a clear suggestion as to how policy responsibilities could be divided between the 71

The Demos

national and EU levels of government. Building on game theory, Majone distinguishes between zero-sum and positive-sum games (Majone 2005). Policies which are zero-sum games (redistributive policies) require levels of solidarity which are not present at EU level because such policies produce winners and losers. Hence, due to the absent demos, such policies must remain within the realm of the nation states. Positive-sum game policies, however, benefit all and thus do not require the same levels of solidarity and identity: Unlike redistribution—a zero-sum game—efficiency issues may be thought of as positive-sum games where everybody can gain, provided the right policy is adopted. Hence, efficiency-enhancing policies do not need a strong normative foundation: output legitimacy (accountability by results) is generally sufficient. In contrast, redistributive polices can only be legitimated by majority decisions and hence place too heavy a burden on the fragile normative foundations of a transnational polity. (Majone 2005: 191)

THE ABSENT EU DEMOS So, what specifically would be characteristic of a European demos? A detailed answer is provided by Chryssochoou: A European demos can be said to exist insofar as the constituent publics see themselves as parts of a democratic whole, and are given the institutional means to mark their impact on the governance of the composite polity, thus creating the basis for a self-directed and self-determining collectivity. It is therefore possible to identify the following prerequisites for the emergence of a European demos: democratic self-consciousness of the collective citizen body; adherence to shared democratic values; public awareness of the transnational process; and a desire to shape democratically the future of a plurality of interrelated peoples, without endangering the existence of that plurality. (Chryssochoou 2001: 252)

Presently, most scholars agree that no EU demos exists (Lord 1998: 114; Weiler 1998: 237; Scharpf 1999: 9; Dahrendorf 2003: 104) although there is some disagreement as to whether a demos is developing (Verdun and Stavridis 2001: 218).12 There are indications that a civic community may be developing at elite level, or rather perhaps, interest group level, but as Smith argues, it is ‘one thing for elites in Brussels, Strasbourg and some European capitals to identify with and work for a united Europe, quite another to attribute such sentiments and beliefs to the great mass of middle and working classes . . . ’ (Smith 1992: 72). A similar conclusion is reached by Chryssochoou, who argues that: 72

The Demos . . . although the EU has given rise to the formation of a large-scale civil society— composed of policy communities, structures of functional representation, patterns of interest intermediation, networking activities and the like—the development of a shared polity identity among its constituent demoi has not yet grown into an extended political public sphere. (Chryssochoou 2001: 261)

This argument, in slightly more empirical terms, is supported by Schmitter: . . . the almost complete disinterest in the deliberations of the Convention on the part of the public and the press offers clear evidence that there does not yet exist a European demos actively concerned with ‘its’ supranational governance. Even the multiple European demoi that do exist seem serenely unconcerned with ‘their’ supranational governance. (Schmitter 2003: 73)

Thus, even if there exists a bond between transnational elites, this is not sufficient to legitimize the polity.

3.3. The no-demos thesis The no-demos thesis has its roots in a verdict by the German Constitutional Court which ended a court case in which a German citizen argued that ratifying the Maastricht Treaty would be in violation of German constitutional law. Although less publicized than the well-known troubles faced by the British, French, and Danish governments in connection with the ratification of the Maastricht Treaty, this verdict could have important implications for the German role in continued European integration. The German Constitutional Court tried a case put forward by a German national (and former civil servant in the European Commission) who argued that the Maastricht Treaty would deprive the German people of its constitutionally guaranteed power (The German Basic Law) to govern itself due to the transfer of powers from the national to the EU level of government.13 In its verdict the court ruled that democracy in Europe is structured in such a way that each member state has a people (ein Volk) who are organized in their respective states (Staatenverband) and that something similar does not exist at EU level (Rasmussen 1996: 70). In its final ruling, the court stated that in order to guarantee the German people continued democratic governance, the Maastricht Treaty could be ratified only if the competencies transferred to the EU were clearly defined (Rasmussen 1996: 70) and continued to depend on the agreement of democratically accountable national governments (Scharpf 1999: 10). 73

The Demos

It is the premise rather than the verdict itself, which has given name to the no demos thesis: One of the most trenchant critiques of authority to emerge recently has come from a certain strand of German constitutional theory and can be entitled the NoDemos thesis. Interestingly, it found powerful expression in the so-called Maastricht decision of the German Federal Constitutional Court. (Weiler, Haltern, and Mayer 1995: 10).

The consequences of this verdict are summed up by Maurer: . . . the German Constitutional Court’s Maastricht ruling lead to a more general critique of the EU’s parliamentary model. The basic assumption of the Court and later of its protagonist commentators was that a polity presupposes a demos in ethno-national or ethno-cultural terms (the ‘Volk’ instead of the ‘Gesellschaft’ or ‘Gemeinschaft’), and that without a single European people sharing heritage, language, culture and ethnic background, that without a European public space of communication that could shape the wills and opinion of the population, no European statehood could be founded. (Maurer 2001: 101)

In other words, the court concluded that Europe does not have a people and thus does not have the social structures in place on which to build a democratically governed polity. This claim is to a very high degree supported by EU scholars preoccupied with democracy. As mentioned in Chapter 2, Lord argues that no primordial European identity exists on which the EU can draw (Lord 1998: 114). Chryssochoou states that ‘. . . the EU suffers from a severe case of deficient ‘civic we-ness’ (Chryssochoou 2001: 251) and Scharpf argues that in the absence of a thick collective European identity, institutional reforms cannot greatly increase the inputoriented legitimacy of decisions taken by majority rule (Scharpf 1999: 9). The natural question to ask at this stage is whether it would be possible to create such a demos, and if so, whether this is really desirable.

3.3.1. Can an EU Demos Be Constructed? If a European demos, or rather an EU demos, were a reality, the EU could be governed as a parliamentary democratic system based on majoritarian principles known to varying degrees from the domestic systems of the member states. From this perspective, creating a European demos appears to be an attractive solution to the democratic deficit of the EU. Whilst the demos can be defined as a people for political purposes, that is, a group of people who can use the word ‘self’ about their group, hence making it capable of self-determination and suitable for self-government, for the 74

The Demos

group to reach this stage, there is agreement that an amount of ‘sameness’ is necessary, a sameness based on a feeling of common identity. This feeling of sameness then produces the ability amongst the individuals to feel ‘social solidarity’ with one another. So, would it be possible to artificially construct such a European demos? The literature is sceptical, primarily for some of the reasons mentioned in Chapter 1, that is, if a demos necessitates a common identity (which many scholars argue) it means that a common EU identity must be created. And whilst accepting the necessity of a deeper transnational civic relationship, the majority of scholars agree that seeking to construct a European identity would be a risky project. For instance, Weiler concludes that ‘Europe is “not yet” a demos in the organic natural-cultural sense and should never become one’ (Weiler, Haltern, and Mayer 1995: 21; Weiler 1998: 246). The reason being that: It would be more than ironic if a polity set up as a means to counter the excesses of statism ended up coming round full circle and transforming itself into a (super) state. It would be equally ironic if the ethos, which rejected the boundary abuse of the nation-state, gave birth to a polity with the same potential for abuse. (Weiler 1998: 241)

Instead, scholars suggest (to varying degrees) that the EU seeks to develop a demos without the ethno-cultural elements (Weiler, Haltern, and Mayer 1995: 20) and construct the demos around shared civic values (Weiler, Haltern, and Mayer 1995: 21; Lord 1998: 312; Green 1999: 33). Although Habermas does not talk about a full-blown European demos in the ethnocultural sense, he goes further than most: Peoples come into being only with their state constitutions. Democracy itself is a juridically mediated form of political integration. Of course, democracy depends, in its turn, on the existence of a political culture shared by all citizens. But there is no call for defeatism, if one bears in mind that in the 19th-century European states, national consciousness and social solidarity were only gradually produced, with the help of national historiography, mass communications and universal conscription. If that artificial ‘solidarity among strangers’ came about thanks to a historically momentous effort of abstraction from local, dynastic consciousness to a consciousness that was national and democratic, then why should it be impossible to extend this learning process beyond national borders? (Habermas 2000: 57)

This view is echoed by Decker who states that contrary to popular opinion national consciousness and solidarity did not precede the nation states (Decker 2002: 264). Hence, if Europe is built properly from above, then 75

The Demos

a community will develop from below (Decker 2002: 264). Decker does, however, consent to the fact that building an EU identity will be harder than the nation state formation for a number of reasons, including language problems (Decker 2002: 259–64.) Again, Habermas is more explicit than most others: It is undisputed that there can be no Europe-wide democratic will-formation capable of enacting and legitimating positively coordinated and effective redistributive policies without an expanded basis of solidarity. Civic solidarity, long limited to nation-states, will have to be appropriated by citizens of the Union such that Swedes and Portuguese, for example, are prepared to stand up for each other. Only then could they be expected to support a roughly equivalent minimum wage, or the general equality of conditions for pursuing individual life projects, even as they remain shaped by national belonging. (Habermas 2003: 97)

Habermas suggests that the long-term aim for the European Union will have to be the ‘gradual elimination of the social divisions and stratification of world society without prejudice to cultural specificity’. This is related to the fact that Habermas is worried that the creation of European Unity may lead to the formation of a fortress Europe. (Habermas 2000: 57) But others dispute that the European integration process is another case of nation-building (see Schmitter 2000c: 45). Certainly, the techniques applied in Westphalian state/nation building could not be adapted to the EU level (e.g. coercion). In practical terms, therefore, the debate is about whether a demos can be created via an elite-driven top-down approach (as argued by Decker) or whether the social structures need to be there in the first place in order for such institutions to be taken seriously. As often, it may be a bit of both. However, the notion that European national identities and to an extent European domestic demoi are socially constructed is widely accepted although still disputed. Wæver creates a hybrid between the modernists or constructivists, that is, writers such as Hobsbawm14 and the primordialists or perennialists (such as the German Court) arguing that although national identity is a discursive construction it still needs to work on raw material, that is, there has to be a . . . reservoir of myths, stories, old battles and historic figures. This past is neither determining nor trivial for contemporary national movements and identifications. The discursive-articulate approach is an in-between position, avoiding the voluntarism and instability of the ‘invented traditions’ approach. The discursively constructed investments of meaning in historical and ethnic fragments are inert but powerful social realities. (Wæver et al. 1993: 30)

76

The Demos

According to Wæver, a European identity is necessary in order to back up the institutions required in the new Europe. In other words, there needs to be a common identity in order for the asymmetry between political integration and civic spirit to be reduced. Wæver agrees with Weiler that such an identity should not be an ethno-cultural one (Wæver et al. 1993: 194). Instead the EU should strive for a political identity of Franco-American style civic republicanism (Wæver et al. 1993: 195). It can perhaps be seen as slightly paradoxical that many of the exact same authors who argue that the EU cannot become a real, transnational democracy without a demos are often adamant that the EU should not strive to develop a demos. As mentioned, often this argument is based on an inherent fear of the forces of nationalism.15 I return to this debate in Chapter 7. At this stage, however, it is relevant to bring in Claus Offe who—although writing about democratization of Eastern European and Latin American states—provides some reflections which can be useful in highlighting this issue: While democratic institutions and economic resources can be ‘transplanted’ from the outside world (or their introduction facilitated and their durability protected by a host of positive and negative sanctions designed to support and strengthen new democratic regimes), the civic ‘spirit’ or ‘mental software’ that is needed to drive the hardware of the new institutions is less easily influenced by external intervention. The rise of a robust ‘civil society’ cannot be initiated from the outside. To be sure, the creation of political and cultural patterns of self-recognition and civic confidence within a mixed-nationality state can be furthered to some extent by the legal framework, but legal provisions alone cannot, in the end, create such a pattern where its cultural and ultimately moral ingredients are not present. (Offe 2000: 4) (Italics added)

The analogy may be stretched in the sense that the ‘states’ here would be ‘the EU’ and the ‘outside intervention’ would be elite-driven integration and institution-building; however, the comparison is relevant to the EU debate because in essence it touches on a crucially important aspect of the debate, namely how far institutional and legalistic measures can serve to create a stable political system and possibly even a demos. By highlighting the concept of legality, Offe introduces a useful distinction in the debate about the demos, that is, the distinction between a top-down-created legal demos (such as the material kind, as in a citizenship which provides rights and obligations for individuals within a polity) and an organic, social demos where people identify themselves as ‘same’ (whether this is a result of a myth or even an illusion). For example, by giving citizens 77

The Demos

of EU member states an EU citizenship the rulers can increase sameness in the sense that if nothing else then at least all those citizens are equal in that they share rights and obligations as well as in their relationships with EU institutions. However, the question is whether that is sufficient if the social elements are still not in place. Because citizenship ‘is not only about the politics of public authority. It is also about the social reality of peoplehood and the identity of the polity’ (Weiler 1998: 238). Furthermore, for a European demos to exist as such, its members should recognize their collective existence rather than merely be granted common citizenship rights (Chryssochoou 2001: 251).16 And the assumption that the granting of citizenship or other formalized rights and obligations will automatically lead to the formation of a common identity may be flawed, just like the assumption that institutional structures may fill any gap left by such a lack of common of identity may be flawed: Although one might presume that in a democracy any horizontal trust among citizens is unnecessary as there are institutions (rights, courts, power-limiting checks and balances) which—if need be—protect ‘me’ against evil and unreasonable actions of my fellow citizens. But this would be an overly optimistic misconception. Because these institutions themselves are only as reliable to the extent that they have clearly become a respectable certainty for ‘everyone else’. If they enjoy this respectability, they create the trust that binds people together who are otherwise strangers. (Offe 2000: 10)

In a similar vein, Weiler argues that . . . one of the great fallacies in the art of ‘federation building’, as in nation building, is to confuse the juridical presupposition of a constitutional demos with political and social reality. In many instances, constitutional doctrine presupposes the existence of that which it creates: the demos which is called upon to accept the constitution is constituted, legally, by that very constitution, and often that act of acceptance is among the first steps towards a thicker social and political notion of constitutional demos. Thus, the empirical legitimacy of the constitution may lag behind its formal authority—and it may take generations and civil wars to be fully internalized—as the history of the US testifies. (Weiler 2001: 3)

As regards the analogy to US nation-building, Majone offers the following additional critique, arguing that although the notion of the American people appeared a myth to some, it did have ‘some plausibility: a common language; legal systems derived from, and still very much influenced by, English common law; similar political and administrative systems at state level; a fairly homogeneous population, largely of English, Scottish, or Irish stock; above all, a war fought together for eight years against 78

The Demos

the former colonial power. None of these conditions is even remotely approached in contemporary Europe, so that the Madisonial solution to the problem of legitimating a significant centralization of powers is simply unavailable to the political leaders of the Union. (Majone 2005: 27). Hence, so far we can conclude that a demos can be perceived as the legal and institutional framework, in other words, people are a demos in a technical, legal sense if they have rights and obligations granted by a political authority. But without the common identity there is still no social demos, in other words, there still is no ‘we’. A legal demos, on the other hand, can more easily be created because it is nothing but a ‘people on paper’. As regards the social side of the demos, this consists of two types of properties. Weiler names the first group the ethno-cultural conditions (Weiler, Haltern, and Mayer 1995). These are matters such as language, history, political habits, ethnic origin, and religion. The second dimension consists of the subjective emotional elements such as solidarity and identity.17 To some extent, the two dimensions are interrelated in the sense that high ethno-cultural homogeneity would probably tend to increase the intensity of the subjective emotional identification. This assumption has normally led to the hypothesis that a shortage of the former leads to the absence of the latter, although this hypothesis has not been systematically proven, only frequently repeated. Hence, if a high level of subjective emotional identification could be achieved without the ethno-cultural elements the EU would have a greater chance of acquiring a demos: We suggest that European integration is a matter of identity, but that European integration does not necessarily demand close integration of peoples, shared culture and homogeneity. It is first of all a political body that is needed, a political Europe. [. . . ] What is needed is a development of political identity. But a new European political identity should not, and cannot, substitute for national identity. (Wæver et al. 1993: 76)

Keeping the distinction between the ethno-cultural and the subjectiveemotional elements (identity and solidarity) of the social demos in mind, the question then becomes to what extent it is possible to talk about a political identity and perhaps also a political solidarity which are not preceded by ethno-cultural elements. In other words, is it possible to create a political community where there is no—or only a weak—cultural community? 79

The Demos

Easton distinguishes between a ‘political community’ and ‘the sense or feelings of community’ (Easton 1964: 183): It is important to bear in mind this distinction now being made between the political community and the sense of community. The latter is a dimension of the former, the affective aspect. It may or may not be present; and when present, it may appear in different degrees. It will reflect the varying cohesiveness of the political community. (Easton 1964: 184) (Italics added)

According to Easton, support for a political system may be derived from sources other than ‘mutual feelings of belonging together’ (Easton 1964: 196): Not only is this conceivable; there is ample evidence to demonstrate the relative independence for considerable periods of time of political communities from the feelings of community among the members. It is possible for a political structure to bind a group together before feelings of mutual identification have emerged. (Easton 1964: 186)

As will be clear, this reading of the political community takes issue with the no-demos thesis—particularly with the strong, organic version expressed by the German Court. However, the crucial difference is that Easton separates the social from the political—a move that may not be acceptable to all scholars of political science or sociology. In that logic it follows that: ‘The members of a political system who are participating in a common political community may well have different cultures and traditions and they may be entirely separate nationalities’ (Easton 1964: 177). From this position, Easton goes on to claim that: ‘A political community may precede and become a condition for the growth of a sense of community’ (Easton 1964: 186). One could then ask why this has not happened at EU level. But Easton does not categorically claim that it will, only that it can, and he does consent that if a sense of community does not develop over time it ‘may leave the system extremely vulnerable to stress’ (Easton 1964: 187). Nevertheless, in Easton’s view the fact that a political system exists is in itself a factor that binds the people together: . . . regardless of the degree of cohesion among the members of the system, as long as they are part of the same political system, they cannot escape sharing in or being linked by a common division of labour. This forms the structural connection that gives minimal linkage to political activities that might otherwise be isolated or independent. (Easton 1964: 178)

80

The Demos

Hence, Easton’s framework is particularly useful in helping us understand the original intentions of European integration because it was precisely such ideas that lay behind the unification project. As regards the notion of the ‘division of labour’, this is also a helpful concept to those searching for ways to legitimize the union. This must be understood in a rather abstract sense and is best explained by the author himself: At the one extreme, we might conceivably find a system in which each member has an explicit and highly active role in the regulation of conflicting demands or the authoritative allocation of values. Every adult may be expected to show some interest in politics, participate at least to the extent of casting a vote and discussing the issues. At the other pole, it will be enough if a person sees his role as one of complete passivity and acceptance of the absolute authority of others over him. In this case, the member’s part in the division of labour is entirely that of owing a duty to accept his subordinate status in political life. (Easton 1964: 187)

In order to determine who is and who is not a member of this order, or division of labour, political systems establish a range of criteria for inclusion, and to that extent the division of labour can serve as a link between the notion of the concept of legitimacy and the presence of the demos: But if a person could not locate himself anywhere in the political division of labor in a system, was not able to relate himself to anything else that occurred in the system, even to the extent of feeling it was his obligation to obey a superior or abide by some vaguely apprehended tradition, one of two conditions might exist. Either he would not be part of the system or his ties to the political community would be in imminent danger of being severed. To avoid any ambiguity as to who is or is not part of this division of labor, each system provides criteria of membership through territorial presence, legal definition, blood, subjection, kinship, or the like. (Easton 1964: 179)

The difficult task in these circumstances is whether it is possible to tell whether an individual is accepting his or her role in the division of labour or considers him or herself completely outside the system. There are two consequences of using this framework: First of all, it can help us understand why a democratic deficit is more serious in some polities than in others because it explains why an individual may not oppose a political system—despite not feeling very involved in it—if he or she feels included in the polity due to other circumstances. Secondly, although it is perfectly possible to imagine an individual having different roles (in the division of labour) in the different systems of which he or she is a member, this does not mean that overlapping systems can draw on the same sources 81

The Demos

of legitimacy. This applies particularly to citizenship and rights, because rights that the citizens already have cannot be ‘used’ as a legitimizing force for another polity. After all, at how many levels does it make sense to guarantee an individual his or her right to enjoy the respect for the rule of law? Either an individual is guaranteed respect for the rule of law, or he is not. Or in more abstract terms, it really only makes sense to guarantee a citizen the rights to ‘life, liberty, and estate’ once. What this means is that the EU will have to look for other sources of legitimacy unless the EU citizenship replaces the national citizenship, which is not likely to happen. Now, some would probably point out that a national protection of rights does not stretch to cover individuals who are living outside the state of which they are a citizen. In that sense, the new ‘European’ rights can be seen as a natural consequence of the internal market and the free movement of labour. But the fact remains that the majority of people remain living and working within their nation state throughout their lives. And this practical and utilitarian view of rights is limited in understanding the impact of rights because it does not take into account the potential ‘cost’ to the individual, that is, the fact that others gain these rights too. Finally, the utilitarian ‘travelling rights’ analysis does not address the issue of rights as legitimizing factors. Hence, it is possible that, for instance, regional political systems could be legitimized in a cultural community bound in history, the national level could be legitimized in strong national identities coupled with the acting ability in a range of policy areas and the EU level could be legitimized in the shared belief amongst citizens that the EU could have more muscle as an actor on the international scene or in dealing with externalities (for instance, pollution). Weiler proposes a multiple-demoi scenario in which the treaties are seen ‘not only as an agreement among states (a Union of States) but as a “social contract” among the nationals of those states—ratified in accordance with the constitutional requirements in all Member States—that they will in the areas covered by the Treaty regard themselves as associating as citizens in a broader society’ (Weiler 1998: 245). In a similar vein, Schild argues: If the populations of member states are to accept the fact that their national government is more and more often overruled on specific topics, there must be a minimum sense of European identity in the populations of the Member States. If the European polity is to be accepted by its citizens as legitimate, then the national frame of reference should not be the only one to which citizens refer when they evaluate the costs and benefits of European policies. (Schild 2001: 335)

82

The Demos

The idea that the national demoi could exist alongside an EU demos is based on the notion that it is possible for an individual to have multiple identities, or indeed the fact that individuals all have a range of identities. Thus, an individual may have identities bound in their age, gender, class, or religion, but also professional, civic, and ethnic allegiances (Smith 1992: 58). However, above all of these, national identification has become the ‘cultural and political norm, transcending other loyalties in scope and power’ (Smith 1992: 58). Under normal circumstances, most people live quite happily with their multiple identities. Hence, the argument that it may be perfectly reasonable to imagine an individual living with an identity as a Parisian, a Frenchman, and a European (Easton 1964: 181). Indeed—as is shown in Chapter 6—it is very often the case that individuals, when given the choice, define themselves as both European and in terms of their nationality and if this is the case, why should the concept of a dual allegiance in terms of the two polities be in any way controversial? Smith offers a good explanation, arguing that one must make a distinction between individual and collective identification: For the individual, or at any rate for most individuals, identity is usually ‘situational’, if not always optional. That is to say, individuals identify themselves and are identified by others in different ways according to the situations in which they find themselves; as when one goes abroad, one tends to classify oneself (and be classified by others) differently from one’s categorization at home. (Smith 1992: 59)

In contrast, collective identities are more pervasive and persistent. They are ‘less subject to rapid changes and tend to be more intense and durable, even when quite large numbers of individuals no longer feel their power’ (Smith 1992: 59). Hence: In this respect, national identifications possess distinct advantages over the idea of a unified European identity. They are vivid, accessible, well established, long popularized, and still widely believed, in broad outline at least. In each of these respects, ‘Europe’ is deficient both as idea and as process. Above all, it lacks a premodern past—a prehistory which can provide it with emotional sustenance and historic depth. (Smith 1992: 62)

Similarly, Wæver argues that the strength of the national identity—as opposed to other identities—lies in the fact that it . . . ties together past, present and future, and implies the possibility of feeling obligations towards the continued unhindered flow through time. The national

83

The Demos identity is about a special way of being alike to a degree where time-crossing identification is particularly strongly felt. (Wæver et al. 1993: 40)

But even if the union were to cultivate a political as opposed to a cultural ‘European’ identity there would be a need for values and unity around things other than institutions. Again, going back to Lipset’s definition of legitimacy, the union would as a minimum need, if not quite an American dream, a value system that could tie the people together. And as argued earlier, thus far the EU’s values seem far too general. The problem with choosing more specific values is that whenever a specific value is selected another specific value is unselected. This is why for instance Christianity cannot be promoted as a great unifying force in Europe. If one religion is chosen as European, all other religions are automatically un-European. But, if all religions are chosen, then none is distinctly European. As a result, it is difficult for the EU to define what is European because if something is defined as European, its adverse is automatically defined as not European. Hence, creating a meaningful framework for a European identity often ends up including everything and therefore nothing. Or alternatively, the values chosen are more or less universal but proclaimed European with an appropriate European reference attached. An example of the latter can be found in one of the leaflets published by the Commission (and available on the Web) which among other things states that the EU ‘promotes tolerance, respect and mutual understanding—values that Europe’s long history has taught us’ (European Commission 2003). In a similar vein (although with a far more sophisticated argument) professor Ash Amin—in search for a foundation for an all inclusive European identity—proposes that: A possible starting point, which happens to dig deep into European philosophical thought, is publicity for empathy and engagement with the stranger as the essence of what it is to be European. This signals an ethos of being and belonging that builds on the Socratic definition of freedom as the product of dialogue and engagement, rather than pre-given judgements of worth. (The Guardian, 14 July 2003)

From this, Amin derives two principles for a ‘new idea of Europe’. First the principle of hospitality which suits a Europe . . . in which we will all be strangers one day as we routinely move—virtually or physically—from one space to another, the principle of refuge will become crucial for many more than the minorities that at present need protection from persecution and hardship. (The Guardian, 14 July 2003)

84

The Demos

The second principle is an acceptance of mutuality as the source of identity and affiliation: To be European thus becomes a matter of publicity for, and the terms of engagement with the stranger, not only because without the stranger constituted as other the self cannot be defined, but also because the stranger—whether we like it or not—always disrupts the certitude of stable or pure identities. (The Guardian, 14 July 2003)

The problem with Amin’s proposals is that despite the Socratic interpretation (i.e. something classically European), such values are not exclusively European. To the extent that they are European, they are no more European than say American or African. Secondly, if one goes on the current European mood as regards the stranger, Amin’s idea sounds more like a normative agenda—not of the sort that looks at reality and makes suggestions that combines what is realistic with what is desirable—but of the kind that only takes into account what is desirable. However, the article does offer a valuable contribution because unlike most of the literature on European identities it addresses the fact that many current Europeans are not historically European. Thus, the geographical area of Europe is becoming increasingly multicultural and Europeans will have to cope with new values developing within their communities (Kerkhofs 1991: 378). This means that not only is there a deficit in the transnational bond between Europeans but there are also new identities inside EU member states. As a consequence, we cannot even define ‘a European’ as a person historically rooted in the geographical area of Europe (whether that includes the Ukraine, Israel, and Turkey is another debate) because that would exclude all those citizens originating from outside this geographical area. And European leaders are ‘understandably wary of ethnic- or religion-based definitions of Europe’ (The Economist, 13 March 2004). Thus, whilst the need for a definition of what it means to be European increases, producing such a definition is getting harder. The problem in trying to manipulate identities is that they tend to defend themselves. This brings me back to Weiler’s argument that European integration which in fact was intended to save the Europeans from themselves may in itself be perceived as a threat.18 For European identity formation this is likely to lead to the very paradoxical situation that this identity will refuse to develop precisely because rather than weakening national identities the top down approach may strengthen these. Wæver explains this process in terms of relating the concept of culture to the 85

The Demos

concepts of integration versus fragmentation. Hence, he argues that if cultures feel threatened they defend themselves with more culture: ‘If you feel your identity is threatened by internationalisation or Europeanisation, you have to strengthen the expression of existing identities’ (Wæver et al. 1993: 70). In a similar vein, Lehning (whose main interest is citizenship vs. identity) argues that ‘. . . promoting implementation of the objective conditions necessary for Union citizenship, may have the unintended consequence of strengthening national senses of community, of “belonging” to a specific nation-state . . . ’ (Lehning 2001: 276). In summary, without the ethno-cultural dimension which cannot be constructed in the short-term (and is undesirable because of its inherent alienation of ‘the other’) we are left with the possibility that it may be possible to create identity and loyalty (the subjective-emotional elements of the social demos) from other sources. However, as has been argued, this promises to be a very long-term project and one that will be very difficult because the competition from the national identity is so strong at the collective level. In line with other authors, Wæver asks whether it may be possible to decouple the cultural identity and the political identity so that it will be possible to build up a European political identity whilst leaving the cultural elements in the national sphere. Furthermore, whilst the lack of a demos may entail some limitations as regards what kind of decisions can legitimately be made by use of majoritarian supranational decision-making, it need not be a disadvantage as far as integration is concerned. If on the one hand we argue that there is no distinct European identity, we cannot at the same time argue that inclusion of new members is difficult because these countries are too different, hence ‘. . . since European identity is not based on a common historical memory, it is difficult to argue that Turkey should be excluded on those grounds’ (Mayer and Palmowski 2004: 593).

3.4. Conclusion It is interesting that such a large number of writers look unfavourably on the ‘creation of an ethno-cultural transnational demos’, seeing that ethnic and cultural patterns—even if allowing for the occasional historical success of top-down approaches—would be almost impossible to change. Particularly because democratic means alone are not very effective for this purpose. Changing a person’s identity surely is a very long-term process to the extent that it is even possible at all. And although the ‘problems’ 86

The Demos

with multiple identities, ethnic groups or nations have been overcome in other large-scale polities, most frequently this has been achieved by means of imposing a new religion or another structure that the people could unite around. Historically, such approaches have often been combined with an ‘expansive foreign policy’ partly intended to increase social integration by the unifying force facilitated by the identification of an external enemy. Such means are not available to EU integrationists, and instead most suggestions involve an adherence to a value system that could perhaps unite the Europeans. Although many express such views in a rather muddled language, Habermas is quite clear about the fact that he believes that the development of a strong transnational solidarity is a condition for the creation of a European democracy. However, the trouble with the Habermasian argument is that it is so affected by the political agenda at its roots. The solution offered by Habermas in view of the EU’s legitimacy problem does not relate so much to democracy as it does to distribution. Furthermore, Habermas attacks what he sees as the emerging ‘fortress Europe’ and predictably ties this to the neoliberal agenda that in his view has hijacked the EU project. Naturally, Habermas, like any other human being, is entitled to his political opinion. However, when this becomes the foundation of his analysis, it prevents the establishment of solutions to existing problems and instead produces ideas of a European utopia based on a constitutional arrangement, which not only unites the European people but also is expected to make amends on human nature. Add to this the fact that the cornerstones of Habermas’s utopian Europe are mutually exclusive. With the means allowed by Habermas, it is not possible to create a European transnational solidarity without creating a global transnational solidarity, because it is not desirable to define oneself against the other, and consequently that solidarity becomes solidarity with other human beings, that is, humanity. But if a global transnational solidarity were to be created, the risk would be that a distinct European solidarity would still not exist. Rather than be embedded in nation states, solidarity would now be embedded in the global sphere, thus leaving Europeans once again with no distinct social base for their polity—or perhaps more realistically, a national solidarity and a solidarity with other humans would exist alongside each other. In fact, as is shown in Chapter 6, there are indications that this is already the case. Essentially, this point is exactly the same critique that was applied to Amin’s proposals for the European value system. Having said that, Habermas is right that a European polity with supranational characteristics needs a base in civic society whilst at the same 87

The Demos

time being adamant that such an identity should not be modelled on the kind of national identities that we see in most European countries today, mainly because attempts at creating such identities are very likely to backfire or have such long-term prospects that they offer no solutions. However, it should be kept in mind that the extent to which a demos is necessary will depend on the model of democracy that the EU will eventually try to follow. Hence, because the institutional frameworks can be created by individuals and put into existence by the signing (and subsequent ratification) of a treaty, whereas creating or even experimenting with identities is far less achievable, it may make more sense to turn the problem on its head. Instead of imagining what may or may not be possible at the mass psychological level it makes sense to look at the social material which is available and then create a model of democracy for the EU that takes into account the limitations of the demos. Thus, on the basis of the theoretical discussion, it can be established that:

r r r

there is consensus in the literature that there is no EU demos, although; it may be possible for the EU to create a legal demos via a top-down approach for instance by introducing further citizenship benefits and rights; such a demos could not be based on ethno-cultural identities and it is doubtful whether a demos at social or civic level will necessarily follow automatically, and in fact there are two dangers associated with such a move which together constitute what could be described as a ‘double alienation danger’: (1) the creation of a European identity may serve to alienate Europeans from each other because their identities will defend themselves and hence become stronger rather than weaker, and; (2) legal and institutional initiatives aimed at creating a demos may serve the opposite purpose and alienate inhabitants of EU member states from third parties and create a new European nationalism which according to some scholars could be a destabilizing factor.

This problem is enhanced by the fact that other than democratic institutions it is hard to find the necessary values that people can unite around. This means that the ethos of the polity always remains at the macro level (i.e. avoid war and poverty but respect rule of law and democracy) and does not reach the individual level. 88

The Demos

The above conclusions will be taken into consideration in the following analysis of eight ideal types of models of democracy that have been suggested for the EU.

Notes 1. The word democracy came into the English language from French (democratie) in the sixteenth century but originates from Greek (Held 1987: 2). 2. This is the translation provided in the Penguin issue of Aristotle’s The Politics (The Politics, p. 492 and p. 499), Held’s Models of Democracy (1987: 2), and Dahl’s Democracy and its Critics (1989: 3). Although ‘kratia’ can also mean authority (Dahl 1989: 3), this alternative translation would not change anything for the purposes of this book as ‘demokratia’ would then mean ‘the people’s authority’ which would also necessitate the presence of a demos. 3. The actual word ‘rule’ more often suggests the exercise of arbitrary or nondemocratic power (see Sartori 1987: 132; Webster’s New Encyclopedic Dictionary 1995). 4. Of course, interest groups and other segmental actors are representatives of groups. However, the crucial difference is that elected representatives are elected on general principles to be responsible for all government rather than a specific policy agenda. 5. Note that in this quotation, Hooghe and Marks add the prefix ‘multi-level’ to ‘governance’. The multi-level label illustrates that decisions are literally made at different levels which in some definitions is already encompassed by the ‘governance’ term. Hence, authors are not always rigorous in their use of these terms, a fact that is illustrated by this particular article where Hooghe and Marks use the terms ‘governance’ and ‘multi-level governance’ interchangeably. This is particularly evident on pages 236–7 (see Hooghe and Marks 2003). 6. Although excluded people may have access to the decision-making process via the use of other channels such as interest groups or the media to make their positions known, they are not able to take part in the formalized democratic process, i.e. they cannot vote, nor run for office. 7. Habermas offers a useful distinction: ‘In political usage the concepts “nation” and “people” have the same extension. But, in addition to its legal definition, the term nation also has the connotation of a political community shaped by common descent, at the minimum by a common language, culture, and history’ (Habermas 1998b: 399). 8. The consequence of this interpretation is, of course, that the nation and the people are not always the same thing. 9. Technically, the citizens of Northern Ireland are UK citizens without being included in the British people.

89

The Demos 10. The literature referred to is the ‘end-of-the-cold-war’ literature on democratization in Eastern Europe (see for instance Offe 1995, 2000) as well as literature related to the debate on EU democracy (see for instance Chryssochoou 1994, 2001; Weiler et al. 1995; Weiler 1998; Scharpf 1999; Decker 2002). 11. It is important to note that Weiler does not advocate the creation of an EU-wide demos. On the contrary. The details of Weiler’s argument will be described at a later stage. 12. It should be noted that only a minor part of the literature on democracy in the EU deals with the demos problem (although this literature is growing). Therefore, it is possible that scholars who do not address this issue refrain from doing so precisely because they believe that there is a demos and therefore see no reason to treat that problem in their analyses, or that they simply do not think that this is a particularly relevant or important issue. Thus, my claim that the literature agrees that there is no European demos must be read with this reservation in mind. 13. First of all, the complaint was over article F., 3. of the treaty, which according to the complainant granted the EU institutions powers to expand their own powers. This is what is known as the kompetenz-sur-kompetenz problem (an argument put forward in a similar case at the High Court in Denmark a few years later). The complainant further argued that the EU had a notorious democratic deficit because the EP did not have sufficient powers. These two first claims were dismissed by the Court. The third claim was that the Maastricht Treaty was unconstitutional to German voters because it violated the article in the German constitution which states that the German people has the right to govern itself (Rasmussen 1996). 14. In essence, Hobsbawn’s argument is that the idea of the nation is a modern phenomenon and that traditions and nations are socially constructed and should not be treated as material structures (see Hobsbawm 1992). 15. For instance, Joseph Weiler draws a parallel between the circumstances and alienation that helped fascism gain ground to the alienation that could be caused by European integration: ‘At an epistemological level, modernity was premised on and experienced in an attempt to group the world into intelligible concepts which had to be understood through reason and science— abstract and universal categories. On this reading, fascism was a response to, and an exploitation of, the angst generated by these practical and cognitive challenges.’ In the same article he compares this situation with the one facing Europe: ‘. . . . eerily, at the end of the century, the European Union can be seen as replicating, in reality or in the subjective perception of individuals and societies, some of these very same features’ (Weiler 1998: 229). 16. In any case, there is a limit as to how far rights are able to substitute—or even supplement—democratic input. In the words of Joseph Weiler: ‘the right to go to court does not emancipate you’ (Weiler 1998: 237). If an individual feels that his or her rights have been violated it is up to that individual to

90

The Demos mobilize the resources necessary to take the perpetrator to court. But this can only happen after the damage has already been done. Hence, rights provide reactive individual influence rather than active collective participation (see Weiler 1998 for a thorough discussion of this issue). In a similar, albeit more subtle vein, Mény argues that although a European Charter of Fundamental Rights is an excellent thing in itself ‘its concrete impact is as simple as it is brutal: the constitutional pillar is more powerful than ever while nothing has been done to strengthen the popular one’ (Mény 2003: 12). 17. A way of distinguishing the two categories is by looking at ways in which one could determine the values of their variables. Where the ethno-cultural components can most often be identified without the actual involvement of the objects of study (the individuals), establishing the values of the subjective– emotional variables necessitates the cooperation of these individuals. 18. The concern expressed by Weiler is that the alienation of the people can produce a crisis of confidence: ‘Whereas in its founding period Europe was positioned as a response to a crisis of confidence, fifty years later it has shifted to become one of the causes of that crisis’ (Weiler 1998: 230).

91

This page intentionally left blank

4 Models of democracy for the EU

4.1. Introduction Where the previous chapter focused on the governed, this chapter will concentrate on government and will evaluate the possible roads to democratization of the EU that have been suggested in recent debates. The models that will be presented are predominantly based on the academic literature, the starting point being a recent article by Dimitris Chryssochoou who identifies four ‘competing models’ of government in the EU, these being the parliamentary model, the consociational model, the federal model, and the confederal model (Chryssochoou 2001). Based on recurrent arguments made in the literature, I have added the presidential model, the two-tier model, the network model and the neo-medieval model.1 These models are not neatly separated by visible demarcation lines; rather, they often have many features in common.2 However, each displays one or more characteristics that distinguish it from the others to the extent that its existence can be justified. Each model will be evaluated for its descriptive strengths, that is, does it contribute to our understanding of the EU, and from a normative perspective, that is, to what extent does the model address the socio-psychological deficit of the EU? This distinction between yardsticks for evaluation is based on the following rationale. In the vast literature that makes recommendations for the future design of the EU there is a tendency amongst authors to recommend the very model that they see as a fit description. Authors do not always make it clear whether they are mapping out a scenario for the future or analysing the workings of the EU system from a descriptive or explanatory angle. Very often these boundaries are blurred and in many cases an author will cross the border into normative territory without explicitly making

93

Models of Democracy for the EU

it known that he or she is doing so, often only to revert into descriptive mode later. However, the fact that a model perhaps serves as an insightful description cannot in itself plausibly lead to the conclusion that it also represents the most favourable scenario. There is no logical reason why the EU should not descriptively fit one particular model but be better off striving towards another. The reverse is also true. Some models may have valuable input as to how the EU can deal with its socio-psychological deficit but be of no or little value as descriptions of the status quo. The advantage of separating the normative from the descriptive also makes it possible to compare like with like, that is, ideal with ideal and reality with reality, hence to be fairer to both, so that when comparing the suitability of a range of models of democracy one uses the ideal type model in all cases rather than a mixture of descriptions and ideal scenarios.3

4.2. The Models 4.2.1. The Confederal Model A confederation is a loose inter-state cooperation based on intergovernmental bargaining. This system, one of governments rather than government, ‘provides the states with a variety of opportunities to achieve mutually advantageous co-operation without resigning their individual sovereignty to a single government’ (Chryssochoou 2001: 256). Similarly, Bauböck defines a confederation as: . . . a model of voluntary association the individual members of which are states. They may leave when they so decide and new ones may be admitted if they are accepted by the present members of the confederation. This model presupposes that states do not abandon their sovereignty and do not merge into some larger unit, just as individual persons are assumed to retain their basic autonomy when joining a voluntary association. (Bauböck 1994: 13)

Another assumption is mentioned by Majone who argues that in general, ‘confederation is government by agreement, and this form of collective governance is possible only if the confederation is not required to act in fields where true agreement cannot be achieved’ (Majone 2005: 213). Crucially, the constituent units remain sovereign and the federal level is relatively weak because it exists solely at the discretion of the units and can do only what they allow it to do (McCormick 1999: 4). Nevertheless, confederalism is often seen as a ‘phase of integration’ implying that the units may not retain this autonomy forever. 94

Models of Democracy for the EU

It is evident that the EU has now crossed the line from confederation into a supranational arrangement. The increasing use of (or scope for use of) qualified majority voting as a decision-making mechanism in the Council underlines this as does the expansion into new areas of both the co-decision and assent procedures.4 However, intergovernmentalism is still a relatively fitting description as regards the CFSP and some areas under Justice and Home Affairs. The fact that there is yet no direct EU tax but instead transfer payments (contributions) to the EU is also a confederal element. However, in order to fit the confederal description perfectly, the EU would have to give up some of the supranational arrangements, an idea which—despite being seen as a setback to the integration process by some—is considered by others to be the only way to preserve the Union, for instance: ‘A transfer of authority downwards, including the handing back of some powers to nations or localities should surely be basic to the future evolution of the EU’ (Giddens 2000: 75). As a confederation, the EU would be legitimized via its segments, the states, and not via a politically active European demos (Chryssochoou 2001: 256), and thus, would be an ideal scenario seen from the no-demos perspective. Thus, the EU would be a gathering of democracies rather than a democracy in itself. As Decker argues: If the unification process were to be carried out without exception along intergovernmental lines, the extent of integration as well as the issues on which decisions are made would still be subject to the consent of the member states. Control over this process by the national parliaments and the public in general would in principle be guaranteed. (Decker 2002: 259)

However, because confederal arrangements are essentially based on consensus they may not be particularly efficient. Thus, with twenty-five members, or more, the EU would be certain to lose some of its acting ability.5 Admittedly, a loss which will not be mourned by all, but which would nevertheless be a large setback to the integration process and hence would be met with resistance from the European institutions, particularly the Commission and the European Parliament, who would see their influence diminished under a confederal arrangement. That a confederal system should be embraced by the European Council is also highly unlikely because it would almost undoubtedly mean a strengthened legislature in most member states. Thus, the confederal scenario is appropriate for describing limited policy areas at EU level; however, as a model it is not fit to describe the EU because it does not, in its nature, encapsulate the extent of the current 95

Models of Democracy for the EU

supranational structures. However, at the normative level such a solution could reduce the democratic deficit from a socio-psychological perspective. Hence, Majone (2005) writes that it ‘is interesting to note that from the end of World War II to this day, the significance of confederation as a possible goal of the European integration process has been systematically overlooked’ (Majone 2005: 209). The reason may be that many (particularly federalists) would see this as a setback to the integration process because of the implications, that is, that some policy areas must remain within the realm of the nation states.

4.2.2. The Federal Model A federation is often seen as the ‘next step on’ from a confederation because it involves the units surrendering part of their sovereignty to a new and permanent supranational level of authority.6 Federalism is normally found in countries with large populations and territories as well as geographically based linguistic fragmentation (Stepan 1999: 19). According to Riker, a system is federal, if

r r r

two levels of government rule the same land and people each level has at least one area of action in which it is autonomous there is some guarantee (even if merely a statement in the constitution) of the autonomy of each government in its own sphere (Riker 1964: 11).

This can be contrasted with a more recent definition offered by McCormick who defines a federation as a system which has an elected national government with sole power over foreign and security policy, a single currency and a joint defence force, national as well as local bodies of law, and a minimum of two levels of government, bureaucracy, and taxation (McCormick 1999: 7). Thus, under Riker’s definition the EU would be a federation, whereas under McCormick’s definition, this would not be the case because the EU has no real common foreign policy, no direct tax, and no directly elected national (here EU-level) government. However, there is no question that Riker also considers a common foreign policy part of a federation, only in Riker’s terminology, tax and foreign policy are the premises for the construction of the federation rather than part of the definition of the system, and thus the creation of a federation is seen as a bargain between ‘prospective national leaders and officials of constituent governments for 96

Models of Democracy for the EU

the purpose of aggregating territory, the better to lay taxes and raise armies’ (Riker 1964: 11). In fact, Riker considers the wish to meet an external military threat or to prepare for aggression an integral part of the motivation for leaders to join forces in a federal system. In return for giving up some independence ‘for the sake of union’ leaders gain protection from an external threat or the possibility to participate in potential aggression (Riker 1964: 12–13). Although Riker is specific in talking about ‘military’ threats or aggression, in modern security terminology (particularly from the constructivist school) economic or environmental threats are part of the vocabulary. Seen in this light, the European case fits Riker’s description. As regards the military threat, European integration was undoubtedly partly given momentum by the perceived Soviet threat during the Cold War and perhaps more recently in new security threats such as global terrorism. But initially European integration was motivated by the internal threat rather than a threat from the outside in the sense that integration was—to a large extent—recommended as a means whereby Europeans could be saved from themselves. Seen from a ‘demos focus’ what characterizes a federal system would be that the authority granted from the people is—at least partly—granted to the local unit of government rather than the central government. Thus, federalism has often been said to be the best way of creating a democratic EU because legitimacy could be provided at EU level whilst maintaining sources of legitimacy from the national systems at the same time. This is a major difference to the unitary system where local units are merely creations of the national parliament, fully subject, in constitutional principle at least, to its control. The national government delegates authority to local governments; it does not alienate its authority. Thus, the constitutional arrangements would permit the national demos to exercise full control over the agenda of political life. A national majority may, if it chooses, overrule the decisions of local units, for example, by removing the matters in question from the agenda of local governments (Dahl 1989: 198).

FEDERALISM AND THE DEMOS Unsurprisingly, scholars are divided on this issue. Siedentop—who in principle supports a federal model—is concerned that Europeans are not yet ready for this solution and is worried about the repercussions premature federalism could have for Europe (Siedentop 2000). This concern is shared by Stepan: 97

Models of Democracy for the EU The European Union is composed of independent states, most of which are nationstates. These states are indeed increasingly becoming ‘functionally federal’. Were there to be a prolonged recession (or a depression), however, and were some EU member states to experience very high unemployment rates in comparison to others, member states could vote to dismantle some of the economic federal structures of the federation that were perceived as being ‘politically dysfunctional’. (Stepan 1999: 33)

Another obstacle in the context of the EU has arisen because the word ‘federal’ in itself tends to raise controversy to the extent that it is frequently referred to as the f-word (see for instance Nicolaïdis 2003: 6; Schmitter 2003: 75). The most common misunderstandings are: (a) that a federation equates to a so-called super-state and (b) that on a continuum of integration, federalism is the most integrationist position (Katz 2001: 75).7 However, contrary to popular perception, federalism ‘by formally delimiting the power of the central government, may actually imply far less centralization of power than the current European emphasis simply on “completing the single market” coupled with a half-hearted bow to an ill-defined idea like “subsidiarity” ’ (Katz 2001: 75).8 In fact, federalism offers a legal guarantee to a subnational tier of government (Bogdanor 1997: 288; McCormick 1999: 4).9 This is in contrast to a unitary system where local units are ‘merely the creation of the national parliament, fully subject, in constitutional principle at least, to its control’ (Dahl 1989: 198). Thus, there is no dependent interrelationship between depth of integration and degree of federalism, and hence, the opposite of federalism is not less integration but a unitary system. Thus, paradoxically, to those who are not enthusiastic about European integration, federalism may actually offer the most attractive solution (perhaps with the exception of total withdrawal from the union) but at the same time, it is the sceptics who object the most to the actual use of the term ‘federal’. Having said that, although federalists tend to think of a federation as protecting cultural heterogeneity and guaranteeing the rights of constituent units, antifederalists tend to downplay the role that subsidiarity assumes in federalist thinking and emphasize the loss of freedom that a once-sovereign state suffers upon joining a federal union that by definition circumscribes its powers (Sbragia 1992: 260). Hence, there exists an ongoing controversy over the term and its suitability for the EU. It has also been argued that a federal system presupposes the civic spirit that it seeks to create. For instance, Weiler (quoted in Chapter 3) argues that in many instances the constitutional doctrine presupposes the 98

Models of Democracy for the EU

existence of the demos, which it seeks to create. According to Weiler, the difficulty in Europe is that the constituting demos is not present: Put simply, Europe’s constitutional architecture has never been validated by a process of constitutional adoption by a European constitutional demos and, hence, as a matter of both normative political principles and empirical social observation, the European constitutional discipline does not enjoy the same kind of authority that may be found in federal states where their federalism is rooted in classic constitutional order. (Weiler 2002: 567)

The explanation lies in the fact that the primary concern of the European integrationists was never European democracy but European stability: Federalism was deliberately not based on popular democratic aspirations, it was more about saving European citizens from themselves. It is more difficult to see why European federalism should be seen as an answer to problems of democratic legitimacy today. With Euroscepticism on the rise across the EU, a crisis of confidence in its institutions, and a revival of regional and local identities, people seem more worried about the intrusive nature of European integration than the lack of political infrastructure to support it. (Leonard 2000: 89)

Furthermore, Weiler may be expecting too much in arguing that the EU needs a constituent demos to support a federation. If there were one demos, the need for a federation would probably not exist. However, what the EU does need if it wants to pursue a federal model is support from its constituent demoi to do so. And the federal model avoids the question of whether democratic loyalty and representation can automatically be transferred to a higher level (Hix 1994: 11). Majone does not believe that it can: Those who assume that economic integration sooner or later must lead to full political integration tend to apply to the European institutions, seen as the kernel of a future federation, the same standards of legitimacy that prevail at the national level. Full-fledge federalism, however, is an improbable trajectory of the integration process: it never enjoyed popular support, and even elite support seems to be eroding at present. (Majone 2005: 41)

And further: . . . the federalist project was doomed from the start, because—absent a European demos—a European federation would lack the material and normative resources to provide the public goods people have come to expect from the state, whether unitary or federal. As was noted at the beginning of this chapter, and in the Preface,

99

Models of Democracy for the EU federalists have always been reluctant to face squarely this problem, indulging instead in the much easier task of designing alternative institutional architectures for the would-be federation. In the current atmosphere of Euro-skepticism such academic exercises are no longer sufficient. (Majone 2005: 220)

To sum up, the federal model has some descriptive strength as regards the EU, primarily because the federal lens far better captures the structures of the European system than a unitary lens. And to the extent that one considers the Treaties constitutional documents, it could be argued that the EU is in fact a federation with a division of powers between segments and the federal level. However, whether the federal model will be a good way to democratize is a far harder question. If, as argued by Weiler, federalism requires a constituent demos, then a federal EU is a problem rather than a solution to the EU democracy problem.

4.2.3. The Parliamentary Model A parliamentary system of government, in its purest form, can be defined as a system of mutual independence where: (1) The chief executive power must be supported by a majority in the legislature and can fail if it receives a vote of no confidence; and (2) The executive power (normally in conjunction with the head of state) has the capacity to dissolve the legislature and call for elections’ (Stepan and Skach 1993: 3). Hence, in parliamentary regimes, the chief executive (i.e. the head of government) derives legitimacy indirectly via the legislature. The system of proportional representation used in the Scandinavian countries combined with a low minimum threshold for acquiring seats in parliament tends to produce a far greater range of competing parties whereas the first-pastthe-post system seen in the UK tends to serve as a barrier to smaller, newer parties because the electorates may be put off voting for them in order not to ‘waste’ their votes. In both cases, however, the principle underlying these models is that of majoritarianism, a principle which requires the existence of a demos, or a ‘shared identity’ to ensure legitimacy (Dahl 1989: 161; Lord 1998: 107; Weiler 1998: 238; Chryssochoou 2001: 251). THE PARLIAMENTARY MODEL AND THE EU If measuring the EU against the pure definition cited above, it is evident that the EU does not resemble a parliamentary system regardless of how 100

Models of Democracy for the EU

one decides to define its institutions. The Commission is appointed and the Council is indirectly elected via the domestic processes in the member states. It is true that the European Parliament can dismiss the Commission (this requires two-thirds of the votes); however, the EP has no such power when it comes to the Council. Thus, when looking at the two conditions mapped out by Stepan and Skach it is clear that the EU cannot be regarded as a parliamentary regime. The question is how it holds up if one uses a ‘softer’ definition of a parliamentary system? Chryssochoou lists four elements of a parliamentary model:

r r r r

the centrality of majority rule in central decision-making; the role of parliamentary institutions in law-making; the role of the parliamentary institutions in controlling functions at the larger level; and, the role of the opposition in governance (Chryssochoou 2001: 253).

In the EU, the majoritarian principle is used in the Council in a number of areas. In the absence of a consensus, areas that fall under the co-decision procedure can be settled via qualified majority voting. In the EP, the majoritarian principle is applied under most procedures. Under the codecision procedure the EP gives opinion by simple majority under the first reading of a proposal. For a proposal to be rejected under the second reading an absolute majority is needed. However, the EP despite having been given increasing powers in the Maastricht Treaty is not involved in all decision-making. In fact, the EP is effectively kept out of influence in a range of policy areas such as foreign policy and agricultural policy. And if the ‘larger level’ is defined as the executive, the EP has no real powers when it comes to controlling the Council, certainly not the European Council (which surely is the ‘larger level’). Finally, the opposition does not really exist in the EU because there is no particular colour of government, that is, the EP does not have a two-bloc system consisting of government and opposition (Bogdanor 1995: 59) and it remains impossible to translate a majority produced by the European electorate into an effective and predictable change of government or policy (Schmitter 2000a: 7).10 Although there are political groupings within the European Parliament, there are no European parties as such. Furthermore, despite the Commission’s attempts at creating one, there is no transnational European political discourse. Politics of the European Union are seen by most citizens as a matter of either for or against European integration 101

Models of Democracy for the EU

(perhaps by proxy, e.g. the Euro or Schengen) rather than pan-European competitive policymaking. Having said this, one could perhaps ask whether developments are moving in a parliamentary direction. As mentioned in the discussion on the horizontal democracy paradigm (see Chapter 2), many have argued that an increase in the powers of the EP could compensate for the relative loss made by national parliaments vis-à-vis their executives. For instance, whilst recognizing that the EU at present cannot be considered a parliamentary system, Chryssochoou argues that . . . in order to compensate for the loss of national democratic autonomy on the part of the constituent representative assemblies, it must build on its existing parliamentary structure and develop effective systems of indirect demos control at the European level. [ . . . ] For parliamentary institutions have become a prerequisite for democratic government, wherever they are located and whatever their formal powers may be, as they represent the place where public preferences are converted into political decisions. (Chryssochoou, Stavridis, and Tsinisizelis 1998: 125)

So, according to Chryssochoou, the system that people recognize from the domestic sphere should be copied at EU level because people have now become so familiar with parliamentary structures that they will never accept anything else. This view is backed by recent opinion polls which reveal that EU citizens see their channels of input mainly in terms of electoral participation. (see for instance Special Eurobarometer Report 251, 2006: 47) However, the counter argument is summarized by Decker: The term ‘democratic deficit’ downplays the difficulties, which are bound to be faced when the parliamentary model is transferred to the European level. It implies that democracy in the Europe of the future could take on a form similar to democracy in today’s nation-states. Seen in retrospect, however, the history of democracy is inseparable from that of the nation-state. (Decker 2002: 257)

The problem with the EU—and one that Chryssochoou recognizes (see Chryssochoou 2001: 253)—is whether the majoritarian principle known from domestic parliamentary systems can be justified at the EU level.11 The majoritarian principle is not an ideal but a practical way of getting out of a decision-making deadlock by saying that if all the people cannot get the result they want, then it must be preferable that most of them succeed. If a minority could always veto the decisions made by a majority, then the political effect would be minority rule (Dahl 1989: 137) which— all things being equal—is surely worse. Whether it works or not will, to a large extent, depend on the extent to which the representatives can really 102

Models of Democracy for the EU

be said to represent the people, because majoritarian decision-making can only be successfully applied where there exists a certain homogeneity of the people (see Chapter 3). Otherwise, those in the minority are going to find the system illegitimate. Dahl argues that: First, the more homogeneous the people of a country are, particularly in characteristics that are strongly associated with political attitudes, the less likely it is that a majority will support policies harmful to a minority and therefore the more likely it is that a broad consensus on the desirability of a majority rule will exist. At the limit, a country’s people would be so homogeneous that no majority could ever harm a minority without simultaneously harming its own members, an assumption of Rousseau’s, I believe, that allowed him to entrust collective decisions about the general good so confidently to the majority. (Dahl 1989: 161)

In other words, without a demos, a parliamentary system is not feasible: Second, the stronger the expectations among the members of a political minority that they will enter into tomorrow’s majority, the more acceptable majority rule will be to them, the less they will feel a need for such special guarantees as a minority veto, and the more likely they are to see these as impediments to their own future prospects as participants in a majority government. Finally, whether as a consequence of the first two or for other reasons, majority rule is likely to gain greater support among members of a minority if they are confident that collective decisions will never fundamentally endanger the basic elements of their way of life, whether in matters of religion, language, economic security or others. (Dahl 1989: 161)

At the present time, it would be difficult to argue that the EU satisfies these requirements. In fact, it has been argued that direct elections to the European Parliament have made the situation worse rather than better: It is a paradox then, that direct elections, intended to help create the political will for European integration might actually have served to increase popular alienation from European institutions, since the European party system is unable to act as a vehicle for genuine choice at the electoral level. (Bogdanor 1995: 59)

Thus, the dilemma as regards the parliamentary model has two potential solutions. One, we can accept that the EU does not have the social infrastructure required to justify a parliamentary system simply because the EU is not a nation state with a demos, and try to create a different, more suitable model. The second solution would be to attempt to change the social infrastructure of the EU so that it becomes appropriate for a parliamentary system. This solution, on the other hand, has its own downside: 103

Models of Democracy for the EU To attempt to change those cultures too rapidly—subjecting them to rule from a single centre in fact even if not in theory—would risk destroying the different forms of civic spirit which exist in Europe today. (Siedentop 2000: 231)

Furthermore, it has been argued that a system which had a true bicameral legislature, with EP assent required for all EU legislation; an EP power to propose legislation; and systematic majority voting in the Council of Ministers could potentially lead to a worsened situation. Under such a system, the EP would debate, formulate, and decide all EU policy. Since the EP is the only directly elected EU institution, this reform would provide a direct link between citizens and EU policymakers. Such a reform might well reduce the democratic deficit and advance the federalist agenda, but its effect on the EU’s stability is potentially catastrophic. The reform eliminates several institutions that, according to consociational theory, are necessary for stable governance in severely segmented societies: government by grand coalition, deliberations in secret, and a minority veto. Furthermore, the reformed EU would govern through democratic institutions that have proven unstable in segmented societies. (Gabel 1998a: 471)

Furthermore, such developments could eventually lead to a situation where ‘alienated segments of the polity may reject the political authority and turn to extrasystem protest’ (Gabel 1998a: 471). In a similar vein, although directed at reform more generally Schmitter argues that: Nothing could be more dangerous for the future of Euro-democracy than to have it thrust upon a citizenry that is not prepared to exercise it and that continues to believe its interest and passions are best defended by the national not supranational democracy (Schmitter 2000a: 115).

As regards the European Parliament, Majone argues that the . . . EP is seriously deficient as a system of representation of individual interests. These interests are still largely rooted at the national level and hence find their natural expression in national parliaments and political parties. Above all, the EP cannot represent a (nonexistent) European people in the same sense in which national democratic institutions represent a historically defined demos. (Majone 2005: 25)

In short, an imitation of a nation state parliamentary democracy is not an appropriate solution for the EU due to the polity’s absent demos. Thus, seeing that this dilemma will not be solved in the immediate future, a number of scholars have suggested an alternative model whereby Europeans would be given the opportunity to vote directly for the president of the Commission. These arguments will be presented in Section 4.2.4. 104

Models of Democracy for the EU

4.2.4. The Presidential Model As opposed to a pure, parliamentary system, a pure presidential system is characterized by the following: (1) The legislative power has a fixed electoral mandate that is its own source of legitimacy; and (2) The chief executive power has a fixed electoral mandate that is its own source of legitimacy (Stepan and Skach 1993: 4). In its pure form (as an ideal type), it is easy to see how the presidential system is significantly different from the parliamentary system. However, although there are examples of pure, presidential systems (Stepan and Skach mention the United States), empirically, a number of systems could be classified as semi-presidential. In other words, the demarcation line between a parliamentary and a presidential system is unclear and scholars do not always agree when differences in degree become differences in kind (Hazan 1996: 30). PRESIDENTIALISM IN THE EUROPEAN UNION? Today the term ‘president’ exists at a number of levels in the main institutions of the EU system. Perhaps most prominently, there is a president of the European Commission who is appointed for a period of five years. Secondly, there is a president of the European Council. The latter has this title because he or she is the head of government of the member state currently holding the Council presidency. Under the current treaty, states hold the presidency on a six-month rotation basis, which means that a Council president holds this title for a period of six months. Also, the European Parliament has a president who is responsible for overseeing the work of the institution. In the Draft Constitutional Treaty no provisions for a directly elected Commission president were added; however, the procedure for appointing the President of the Council was changed so that the members of the European Council were to elect the European Council president for a period of two-and-a-half years, renewable once. This means that the Council president’s role would assume an increasingly supranational character as he or she will not be permitted to hold national office (see article I -22 of the constitution). Having said that, this does not make the EU system presidential, as neither the Council president nor the Commission president would have a direct electoral manadate. Furthermore, as the prospects of ratifying the constitution are rather dire, to say the least, it is 105

Models of Democracy for the EU

likely that the current institutional design will remain unchanged in the near future. However, despite the use of the terms president and presidency (confusingly used about different high-profile roles) neither resemble a chief executive as defined above and neither is directly elected. Despite the undoubted agenda-setting powers and relatively high profiles of these posts, neither has the powers of a French or an American president. Many observers of European integration have recommended the direct election of a president of the EU as a way of increasing the EU’s legitimacy. The logic here is that if Europeans were to elect their own president directly, they would no longer feel so remote from the EU. Hix argues that the direct election of the Commission president may be the only way ‘to establish genuine “European” elections’ (Hix 1998: 35), maintaining that the aim should be to design institutions that do facilitate genuine European elections on the one hand, and cohesive but not collusive Europarties on the other (Hix 1998: 45). He suggests a partially presidential model based on the direct election of the Commission president, which would have a number of advantages. First of all, EP elections would no longer be second rate ‘as long as candidates are required to be nominated by at least one party in a large majority (or even all) member states’ (Hix 1998: 46). Also, the European electorate would have someone to ‘throw out’, and national parties that supported the winning candidate would be accountable to national electorates for his or her actions. Decker supports a presidential system for yet another reason, namely that in a parliamentary system, some countries would be underrepresented due to size because the prospect of MEPs ridding themselves of their national perspectives is quite unrealistic (Decker 2002: 267). For anybody who has followed the European debate about integration, there is an obvious problem with the presidential model in that for all its good intentions it seems to add to the problem that it intends to solve. The idea that Europeans would really be prepared to vote for a candidate outside their own primary identity framework (although they may find it appealing in theory) may look attractive on paper but whether it is realistic is another matter. It seems to rest on an almost romantic belief that as long as the electorate gets something that unites them, that is, the ability to vote in one election and for the same candidates (as opposed to the EP elections), the perceived lack of legitimacy of the Union will be reduced. But judging from history, it is doubtful whether such a process would have much civic impact. In fact, it may even have an alienating 106

Models of Democracy for the EU

rather than an integrating effect. Creating more institutions (and the creation of a presidency would be the creation of another institution) cannot be expected to solve the EU’s legitimacy problem. And although Hix is right in arguing that the direct election of a president would give the electorate somebody they could ‘throw out’, this may only legitimize the person in the ‘seat’, not the role itself, particularly if people would prefer representation via their domestic channels. Also, if we follow Lord, Weiler, and Easton in distinguishing between two types of authority, it is clear that a directly elected president would always satisfy legitimacy requirements if elections took place in accordance with the democratic rules that Europeans have come to accept and even require. However, if the other element of authority—the consent to the overall structure of power relations—were not present, or if this was based on consensus from an uninformed public, direct election may not be sufficient. Finally, if one really believes that a European demos does not exist, it is impossible to see how that nonexistent entity could be represented by one person. Hence, the question—at a deeper level—is whether it is really possible for one individual to represent the European peoples. Direct election does not necessarily create a representative (see Sartori 1987: 30). Having said that, this does depend on the prerogatives subscribed to that role. If the role were not particularly powerful, then it would not require such substantial legitimacy. But if that were the case, direct election might be pointless and motivating citizens to vote very difficult. If the ‘role’ were to be legitimized it might be wise to ask the Europeans in a referendum whether they would want a directly elected president, reminding them that the probability that such a person would be one of their fellow nationals would be 1/27 (assuming each member state had equal opportunity to put forward a candidate). It will not be denied that technically it might look good on paper to be able to say that the EU’s top figure had been elected by the people, however from the no-demos perspective this would not contribute to increasing democracy. Finally, it is doubtful whether all member states would support this. An EU-wide elected president would— despite his or her weak legitimacy caused by the absence of a demos— nevertheless be able to claim a stronger procedural legitimacy than most national heads of government (as many of these are not directly elected) and as such be seen as threatening by some national governments. It is questionable whether national governments would be prepared to accept such a dramatic institutional change. Even the Convention’s President Giscard d’Estaing did not believe it realistic (BBC News, 16 May 2003). The comment below may serve to highlight this problem: 107

Models of Democracy for the EU At some point Commission President Romany Prodi even played with the idea of a popularly elected head of the European executive. By the summer 2002, however, it was recognized that if would be impossible for a single candidate to fight a meaningful election across twenty-five or more member states. Under a later proposal the president would be nominated by the EP and by representatives of the national parliaments, and approved by EU leaders, thus attempting to tap all sources of democratic legitimacy in the EU. But precisely the need to appeal to the legitimacy of the national parliaments is the clearest admission of the continuing absence of a distinctive European political identity—and of the difficulty of legitimating an institution with such varied objectives and extensive political ambitions. (Majone 2005: 40)

In conclusion, the presidential model does not seriously address the sociopsychological deficit of the union. In fact, it would probably be wise to learn from the experience of the European Parliament, which has never gained ground as a major political player in the minds of the EU electorates. To have two major EU institutions whose members were chosen by a small part of the electorate might increase rather than decrease the democratic deficit of the union.12

4.2.5. The Consociational Model One way whereby polities could overcome the problems with majoritarianism is by adopting a consociational model of democracy where decisions are made by way of consensus seeking. This particular type of government has been known to work in divided societies where the majoritarian principle could not be applied because certain groups would consistently find themselves in the minority. In a consociational system, however, the groups (also known as segments) are all represented by their leaders in the form of an elite cartel where leaders negotiate policy and each maintain a veto over matters of particularly sensitivity. Consociationalism is mainly associated with Arend Lijphart who initially developed it as a theory that could explain stability in societies where stability might not have been expected due to strong segmental divisions (Hix 1994: 19; Weiler, Haltern, and Mayer 1995: 28; Bogaards 1998: 475).13 The main features of a consociational system are grand coalition governments that include representatives of all major linguistic and religious groups, cultural autonomy for the groups, proportionality in political representation and civil service appointments, as well as a veto with regard to vital minority rights and autonomy (Lijphart 1996: 258). Thus, 108

Models of Democracy for the EU

the above-mentioned features provide the polity with stability because its citizens perceive the system as legitimate.14 Critics point to a number of problems with the consociational model. First of all, it is likely to keep reproducing the divisions between fragments that it was designed to overcome partly because it is in the interest of the elites to maintain the segments (Weiler, Haltern, and Mayer 1995: 31). On the other hand, it has been suggested that by giving the EP increased agenda-setting power it may be possible to maintain the consensual politics whilst a demos developed: Only recently did the EP gain some conditional agenda-setting power. A more comprehensive power to initiate legislation might promote transnational coalitions of EU citizens in order to lobby the EP agenda. If these pan-European concerns successfully attract public allegiances, then citizens may begin to define their interests transnationally on EU issues. In this way, decision making could remain consensual while transnational affiliations emerge and reduce the segmental nature of the EU polity. (Gabel 1998a: 472)

Secondly, consociational polities are ‘governed by political elites whereby the individual citizen-voter loses the status of being the major determinant of politics’ (Chryssochoou 1994: 10). The system, therefore, relies heavily on the discretion of the segmental elites to act in accordance with the wishes and interests of their segments. Thus, in essence the criticism of consociationalism can be summed up in the question whether such a model is really democratic. CONSOCIATIONALISM AND THE EUROPEAN UNION When reading Lijphart’s description of the type of society suited for consociationalism, it is easy to see how it can be argued that the EU fits this format: Especially in plural societies—societies that are sharply divided along religious, ideological, linguistic, cultural, ethnic, or racial lines into virtually separate subsocieties with their own political parties, interest groups, and media of communication— the flexibility necessary for majoritarian democracy is likely to be absent. (Lijphart 1999: 32)

Hence, it is no coincidence that many scholars have argued that the EU resembles a consociational model (see Taylor 1996; Gabel 1998a; Lijphart 1999; Chryssochoou 2001; Schmidt 2002; Schmitter 2003). Grand coalition, proportionality, segmental autonomy, and mutual veto can all be found in the Euro-polity (Chryssochoou 2001: 253). However, in addition 109

Models of Democracy for the EU

to these four characteristics, Lijphart lists ten detailed criteria that define a consociational democracy. In 1999, Lijphart evaluated the EU according to these criteria. In the following these will be summarized as briefly as possible.15 (1) Executive power-sharing in broad coalition cabinets: The Commission is a broad coalition because ‘all fifteen nations that belong to the EU are represented on it’ (Lijphart 1999: 42).16 (2) Executive–legislative balance of power: Despite the EP’s increased powers, its role appears to be subordinate: ‘This judgement of executive-legislative relationships changes, however, when we add the Council of the European Union . . . .’ (Lijphart 1999: 43).17 The problem here is that the Council is as much an executive as a legislature so Lijphart may not necessarily be justified in thinking of the Council in this way. (3) Multiparty system: The party political groups in the EP are regarded as ‘parties’ in Lijphart’s analysis. With more than eight such groups and no group even close to a majority the EP can be described as multiparty. As regards the ‘upper house’ which in Lijphart’s terminology means the Council, members are replaced when the cabinets of member states change and changes take place depending which is the subject matter discussed. In practice the Council is a multiparty body (Lijphart 1999: 43). (4) Proportional representation: The EP is elected by PR even in the UK where this was not always the case (Lijphart 1999: 44). (5) Interest group corporatism: Of all the ten characteristics, this is the one where Lijphart struggles the most to fit the EU. ‘The EU has not yet developed a full-fledged corporatism, largely because the most important socio-economic decisions are still made at the national level . . . ’ (Lijphart 1999: 44). The use of the word ‘yet’ can be taken to indicate that the EU will in future be more corporatist and Lijphart ends this section by concluding that this is so. However, he does not account for this prediction. (6) Federal and decentralized government: Here the lack of a conceptual framework for the EU becomes apparent. Lijphart concludes that compared with other international organizations the EU is highly unified and centralized but compared to nation states the EU is ‘obviously still more “confederal” than federal as well as extremely decentralized’ (Lijphart 1999: 45). Here the word ‘still’ serves the 110

Models of Democracy for the EU

same purpose as did ‘yet’ above, namely to indicate that the situation is temporary and will change in future. (Although, of course, Lijphart cannot know this for sure.) (7) Strong bicameralism: According to Lijphart the ‘two criteria of strong bicameralism are that the two houses of a legislature be equal in strength and different in composition’ (1999: 45). Where the EU fits the second criterion, he accepts that because the Council is more powerful than the EP, the EU does not quite live up to this criterion. However, he does conclude that it is still more consensus model than majoritarian.18 (8) Constitutional Rigidity: Because Lijphart sees the treaties as the EU’s constitution and because the treaties require unanimity in order to be changed, the EU has extremely rigid constitutionality (Lijphart 1999: 45). (9) Judicial Review: Because the ECJ has the right of judicial review and can overrule EU and national law if they violate the treaties, this is justified. (However, only in those policy areas that come under EU jurisdiction.) (10) Central bank independence: The Eurozone has an independent central bank. The weakest points are 2, 5, and 7. The weakness in 2 is that Lijphart sees the executive as being the Commission, whereas it could be argued that the Council, in addition to being part of the legislature, is also part of the dual executive (see Hix 1999: 21). As regards 5 (the corporatist element), Lijphart does not explain why he is anticipating a development in this direction. Finally, in point 7 (strong bicameralism), Lijphart admits that the description does not fit the EU. As for 9 (judicial review), Lijphart is, of course, right that the ECJ has supremacy over national courts, albeit only in those areas where the EU has jurisdiction at all. Although Lijphart agrees that the institutions of the EU do not fit the classifications known in many nation states, he argues that there are some similarities. This is primarily true for the European Council. Of the other institutions, Lijphart argues that: . . . the European Commission serves as the executive of the EU and can be compared to a cabinet; the European Parliament is the lower house of the legislature; and the Council of the European Union can be regarded as the upper house. The responsibilities of the European Court of Justice and the European Central Bank are clear from their names. (Lijphart 1999: 42)

111

Models of Democracy for the EU

Finally, one could add that the fact that the consociational model relies heavily on elites is another feature that it shares with the EU, which has always been an elite-driven project.19 For instance, despite the introduction of qualified majority voting in the Council, by far the largest number of decisions is still reached via consensus. There is no doubt that the EU has a number of consociational features. However, as Lijphart’s ten criteria are mainly institutional, it makes sense to ask whether the underlying structures found in the empirical cases of consociational democracy on which the model builds can also be found at EU level. Consociational scholars (and their critics) are divided on the importance of a range of factors for the durability and stability of the consociational polity. Thus, it is not possible to establish consensus on a neat division between what constitutes a favourable factor or what in fact is a condition for sustaining the system. Regardless of the importance subscribed to each, the most common are listed below: segmental isolation (geographical or otherwise), the presence of external threats, a balance of power between the segments or no majority segment, and segments of equal size, small country or population size or a reduced decision load (see Weiler, Haltern, and Mayer 1995: 30; Bogaards 1998; Lijphart 1999). As mentioned, there is no agreement as to whether these conditions are essential for the stability of the system or merely determine the degree to which the system is stable. Having said that, some of these factors are perhaps more important than others. Hence, if using the above list as a yardstick to measure the European Union, it will make sense to subscribe the maximum importance to those elements that both proponents and critics of the consociational model find important. Thus, the main problems pointed out as regards the preconditions or favourable factors are discussed below:

r

112

One of the major problems is the size of the EU polity. There is agreement in the literature that consociationalism works best in small polities and it should not require a meticulous definitional process to establish that the EU is not a small polity. Having said this, in a recent article Lijphart—conveniently—argues that India can be interpreted as a consociational democracy (Lijphart 1996). If one accepts this argument, the problem is, of course, solved. An analysis of the consociational interpretation of India falls beyond the scope of this book; however, the size of the polity is perhaps not the greatest problem as regards the EU polity;

Models of Democracy for the EU

r

r

an arguably more serious issue is the requirement for equal or relatively equal size of the segments. The EU member states vary significantly in size (whether one chooses to measure this in terms of square kilometres or inhabitants). This is a potential problem because if the consociational mechanism is to work there may come a time when larger segments find their comparative lack of power provocative. (Indeed, this is a great debate in the EU, particularly as regards the allocation of seats in the EP, in the controversies over voting weights in the Council and to a smaller degree the number of Commissioners per member state.) Thirdly, even if the segments were of an equal size, equating segments with member states is problematic: Unfortunately, the assumptions of similarity and comparability underlying the (implicit) domestic analogies in consociational interpretations of the EU have not been examined critically and systematically. The equivalence of national segments and member states, for example, has invariably been taken as a starting point, not an object, of analysis. (Bogaards 2002: 369)

The main problem with interpreting states as segments is that the segments in a national consociational democracy were never sovereign, whereas EU member states are, or at least were (Bogaards 2002: 501). One point which enjoys less overall support in the literature—and which can serve as an illustration of the unclear division between a condition and a favourable factor—is the potential significance of an established tradition of elite accommodation. To some scholars this is a precondition for a sustainable consociation because citizens are familiar with, and trust, this form of government. Lijphart himself has included this element on his list of favourable factors (in response to criticism raised by Daalder and others), however, ‘denies its status as a prerequisite for consociational democracy’ (Bogaards 1998: 480). Another central question is whether the EU is really a divided society. Lijphart (2002) partly addresses this issue saying that he and other consociational scholars have so far been unsuccessful in measuring the degree to which a society is plural or deeply divided (Lijphart 2002: 171). This has led him to establish four criteria for a divided society:

r r

Can the segments into which the society is divided be identified exactly? Can the size of each segment be specified? 113

Models of Democracy for the EU

r r

Do the segmental boundaries and the boundaries between political, social and economic organization coincide; and Do the segmental parties receive the same level of voting support from election to election? (Lijphart 2002: 174).

Although the fourth is problematic when applied to the EU, the real problem here is that Lijphart only deals with the first half of the ‘divided society’ term, that is, ‘divided’. But would it not be reasonable to ask whether there really is an EU society? In other words, when is a society so divided that it does not constitute a society? This question remains unanswered. Finally, because, as mentioned above, the consociational model relies so heavily on the prudence of the national elites, it would require the introduction—at least in certain countries—of increased checks on the executive. This is due to the fact that the democratic structures within the ‘segments’ differ substantially, thus giving nationals varying degrees of democratic government: It will often be discovered that some elites, within the consociational cartel of elites, have very deficient internal democratic structures of control and accountability. Even a facile comparison among the structures which exist within the various member states to control their governments is sufficient to illustrate this point. (Weiler, Haltern, and Mayer 1995: 31)

To sum up, the fact that the EU displays some consociational features does not necessarily mean that the underlying preconditions for the durability of a consociational system are in place. On the positive side, the elitedriven consensus model may possibly lead to the creation of a society sufficient to secure political stability in the EU because the consensual institutions may have ‘the potential of making an initially adversarial culture less adversarial and more consensual’ (Lijphart 1999: 307). Thus, it is easy to see how—if one accepts (a) that the EU has no demos and (b) that the EU should continue to exist as a political unit, the consociational model can look like an attractive alternative to the parliamentary model. However, because the system is likely to keep reproducing the fragments, it could potentially hinder the future development of a demos (Chryssochoou 2001: 262). Finally, because the system relies heavily on elites, it may be unsuitable for the EU. Increased elitism is hardly what the EU is seeking at the present time. 114

Models of Democracy for the EU

4.2.6. The Two-Tier Model In his now (in)famous speech given at Humboldt University in May 2000, then German Foreign Minister Joschka Fischer laid out his vision of a future European Union. The basis for his vision was a federal and parliamentary system (in fact, a system not unlike the German federal system): The completion of European integration can only be successfully conceived if it is done on the basis of a division of sovereignty between Europe and the nation state (Fischer 2000: 14). In essence, the model proposed by Fischer looks as follows:

r r r r

transition from a union of states to full parliamentarization as a European federation, which means; ‘nothing less than a European Parliament and a European government which really do exercise legislative and executive power within the federation’; the federation should be based on a constitutional treaty (Fischer 2000: 14); and, a bicameral parliament. ‘One will be for elected members who are also members of their national parliaments. Thus there will be no clash between national parliaments and the European Parliament, between the nation-state and Europe. For the second chamber a decision will have to be made between the Senate model, with directly elected senators from the member states, and a chamber of states along the lines of Germany’s Bundesrat’ (Fischer 2000: 15).

In addition to these federal elements, Fischer added one particular proposal, namely the idea that the EU does not necessarily have to integrate at one speed. Instead, the EU could be centred around a treaty-bound core of countries interested in faster, deeper integration: On the basis of this treaty, the federation would develop its own institutions, establish a government which within the EU should speak with one voice on behalf of the members of the group on as many issues as possible, a strong parliament and directly elected president. Such a centre of gravity would have to be the avant-garde, the driving force for the completion of political integration and should from the start comprise all the elements of the future federation. (Fischer 2000: 18)

As Fischer quite rightly points out, there is already a tendency towards flexibility in the EU as regards participation in various types of 115

Models of Democracy for the EU

cooperation. Scope for opt-outs has always existed but became particularly associated with the Treaty of the European Union (Wallace 1993: 295). As far back as 1993, there was talk of a ‘fast track Union for the really committed’ or a ‘two-tier’ Europe (Wallace 1993: 296). However, the idea of a hard core goes further than this because of Fischer’s suggestion that the core should not only decide policy but also constitutional matters and institutional design. As regards the question of who would be members of such a core, Fischer is undecided although he suggests possibly the founding members (of the ECSC) or all the members of the single currency. However, he maintains that membership of such a core should be ‘open to all member states and candidate countries’. In the same paragraph, however, Fischer adds that should those who wish to participate not ‘fulfil the requirements’, there must be a ‘possibility to be drawn closer in’ (Fischer 2000: 19). The key word here, is, of course, ‘requirements’ because Fischer does not clarify whether he is referring to criteria of a more technical nature, possible economic criteria (such as for instance inflation targets) or political criteria (such as the Copenhagen criteria, mainly referring to human rights issues or possibly national political structures, i.e. democracy), or thirdly, a promise to adopt certain policies under for instance the Justice and Home Affairs umbrella or under the CFSP. The latter would be harder to justify because if membership of a ‘core’ would necessitate a specific foreign policy, the core would in effect have been reduced—or upgraded (depending on one’s preferences) to a political bully.20 Whilst such a system can be seen as one of enhanced cooperation (by those who prefer to look at it that way) it can also be seen as a system of detached cooperation. This is perhaps why it has such a broad appeal. But whether it becomes a system of enhanced or detached cooperation remains to be seen (provided, of course, that this model is chosen at all). The outcome will depend on the weight between the distribution of those who do and those who do not take part. Some problems with this model are easily spotted. It would institutionalize—perhaps at some stage constitutionalize—inequalities between member states, and could create suspicion towards the motives of the core in the periphery countries—and vice versa. Other issues one could raise are the wish to avoid clashes between national parliaments and the EP because, although undoubtedly smoothening the ride for acting politicians, such clashes may provide positive contributions to the democratic process; in fact, one could argue that they could be an integral part of it. The risk of a clash (effectively understood as a disagreement between chambers) may increase legitimacy because such potential clashes could 116

Models of Democracy for the EU

be seen as checks and balances of the system. Furthermore, clashes tend to lead to more media scrutiny and thus contribute to transparency. It would seem that Fischer’s main concern is with efficiency in the decision-making process rather than democracy.

4.2.7. The Network Model The network model is closely related to the multi-level governance literature discussed in Chapter 3.21 Where, at least initially, the network model was developed as a method of understanding the policymaking process, the network model has recently also been promoted as a favourable scenario for the European Union and hence constitutes a descriptive and normative typology as well as an analytical approach.22 As a tool, the network approach offers a lens that captures processes and actors not visible through traditional institutional ‘lenses’ in that it points to different relationships and participants in the policy process as well as to possible non-institutionalized relationships between actors. For these reasons it has been argued that the network is well suited to grasp the essence of governance in the EU (Jachtenfuchs 2001). The literature on networks in political science offers a range of definitions of the concept. In fact, authors provide not only different, but conflicting definitions and not only that, ‘sometimes even the same author provides several different definitions’ (Thatcher 1998: 396). However, the term ‘network’ is used in almost every academic discipline, be it social sciences, natural sciences, computer sciences, or business studies, and at the basic level it makes sense to look at the definitions provided in other disciplines. In IT-terminology, a network means a number of units (in this case, computers) that are connected to each other with the aim of sharing information and resources, and in most other disciplines the network has a very similar meaning. Networks can be large or small, but the essential characteristic of a network is that (a) it has no centre and (b) access—even in wide networks—is limited. To that extent the IT definition is nearly identical to the political science definition by Benson (quoted in Kassim 1994: 18) as ‘a cluster or complex of organisations connected to each other by resource dependencies and distinguished from other clusters of complexes by breaks in the structure of resource dependencies.’ In the social sciences the term was initially used in the 1940s and 1950s to analyse personal relationships, inter-connectedness, and dependencies (Parsons 1995: 185; Rhodes 1997: 32). To political scientists, the network became attractive in the 1970s ‘because it allowed a more fluid 117

Models of Democracy for the EU

and realistic analysis of how people interact at different levels’ (Parsons 1995: 185) and the approach was a critique of, or an alternative to, the pluralist model of interest group intermediation (Kassim 1994: 17; Rhodes 1997: 33). POLICY COMMUNITIES AND ISSUE NETWORKS Despite the problems in arriving at a clear definition, there is some consensus in the literature that one must distinguish between two different types of networks. This discussion is relevant when studying the EU particularly because an evaluation of their impact on democracy and legitimacy may have different results depending on which definition one applies. These two types are named ‘policy networks’ (occasionally ‘policy communities’) and ‘issue networks’. Policy communities are highly integrated within the policymaking process (Parsons 1995: 189) and constitute ‘analyses of civil servants from government departments and interest groups which are built on exchange relationships with clearly defined boundaries’ (Thatcher 1998: 391). Issue networks, on the other hand, ‘represent a much looser set of interests and are less stable and non-exclusive and have much weaker points of entry into actual policymaking’ (Parsons 1995: 189). These two types represent either end of a continuum of types of network developed by Rhodes (1997), where the policy community (also known as the policy network) is characterized by stability, highly restricted membership, vertical interdependence, and limited horizontal articulation (Rhodes 1997: 38). The issue networks, on the other hand, are unstable, have a large number of members and limited vertical interdependence (Rhodes 1997: 38). NETWORKS AS DESCRIPTIVE OR EXPLANATORY TOOLS The network model (or models) captures some of the characteristics of the EU system that other models described in this chapter do not cover. In the EU the existence, the types, and the influence of networks vary considerably across the political spectrum. Networks of a policy community type are often found in areas where EU policy is well established, where an organised ‘clientele’ exists, and where decision-makers benefit from the cooperation of interests. Examples of such policy areas include agriculture and research and development. In contrast, issue networks are more common where EU policy is not well developed, where the policy debate is fluid and shifting, and where such organised interests as do exist have few resources to ‘exchange’ with decision-makers. Consumer protection

118

Models of Democracy for the EU policy and much of environmental and social policy are examples of policy areas where issue networks are commonly found (Nugent 2003: 491).

According to Nugent Amongst factors identified as being conductive to policy networks are: the informal nature of much EU policy-making; the multiplicity of interests at EU level that are anxious to have access to policy-makers; the highly technical—almost nonpolitical—nature of much EU policy content; the powerful policy positions held by senior officials, especially in the Commission and especially in the early stages of policy-making; and the heavy reliance of officials on outside interest formation and advice about policy content and policy implementation (Nugent 2003, 491).

Hence, the network model is a valuable tool for analysing EU policymaking and also implementation, particularly considering the complicated comitology processes that include national experts, Commission civil servants as well as interest groups and are relatively short of transparency despite recent concessions to the EP, who must now have access to information as regards implementation of legislation subject to codecision. Unlike the other models discussed in this chapter, the network encompasses those actors which are more or less overlooked by traditional approaches. THE NETWORK AS A NORMATIVE MODEL Like many other descriptive and explanatory tools, the network model has recently acquired a new status as a recommended system. In this sense it is first and foremost a reaction to the parliamentary and federal models that have been suggested for the EU. For instance, Leonard argues that: . . . we need a new model to understand the EU: the network. Though most people don’t realise it, or talk in these terms, this is already the model which is closest to the EU we have and it is very different from the two alternative models. Networks do not figure prominently in political theory, but we all benefit from them in our everyday lives—in friendships, clubs, churches, trade unions, political parties or the internet. Instead of being planned by a central authority, they come about from the interactions of decentralised actors. The rules and order that govern networks are not imposed from above but depend on reciprocal relationships, shared values and a common identity. (Leonard 2000: 91)

The great problem here is that Leonard uses the same model to ‘understand’ the EU as he also recommends for the future, which means that there is no clear separation between the descriptive and the normative level. 119

Models of Democracy for the EU

Leonard also dismisses any talk of a European government, particularly a president: Because the EU will never have a single government or president that citizens can vote out, we should consider supplementing representative politics with forms of direct democracy. In the long term, we could explore Simon Hix’s idea of holding Europe-wide referendums giving citizens the chance to overturn an existing piece of EU legislation, or to put a new legislative issue on the agenda in policy areas of EU competence. We should also investigate the idea of a European People’s Panel which policy-makers in the EU’s institutions or in national governments could draw upon to test public attitudes to what the EU’s priorities should be, and how service delivery could be improved from the point of view of the user. (Leonard 2000: 96)

However, the problem with the referendum is that it only addresses the issue at hand, not the greater issue at stake, that is, whether people want a particular matter settled at EU level: . . . the referendum, while it can supplement representative democracy by remedying some of its deficiencies, cannot replace it. Even if the referendum were introduced into the Union as an instrument to validate constitutional change, that would hardly suffice to secure popular accountability for the institutions of the Union. (Bogdanor 1995: 59)

The first critique one could raise of the normative network model is whether—rather than being an ideal—it is just a picture of the status quo as seen by Leonard and thus is without much vision. But Leonard recognises this and sees it as an advantage: . . . our ‘Network Europe’ has not come about as a result of a conscious plan—it is a fluke. It is the product of an uneasy truce between the traditional visions of Europe as a free trade area and a federal state in which everybody sees Network Europe as a transitional phase—to be replaced by one of the other visions of Europe when conditions allow. (Leonard 2000: 91).23 [ . . . ] ‘Network Europe’ should now been seen as a ‘desirable goal’ rather than an unfortunate staging post on the road to a federal state or free trade area. We can then start to reform it rationally. (Leonard 2000: 91)24

In an article from 2003 Sørensen and Torfing offer a coherent and clear framework for evaluation of the network model from a normative, democratic perspective. They argue that it is evident that the networks represent a threat to representative democracy because they challenge the ‘oneman-one-vote principle’ (Sørensen and Torfing 2003: 610). However, the truth is slightly more complex because democracy can be defined in 120

Models of Democracy for the EU

many ways, thus one’s definition will determine to what extent and how one evaluates the effects that networks have on democracy. In short, the authors distinguish between aggregative and integrative democracy. In an aggregative democracy, individuals are seen as rational with pregiven interests who attempt to maximize their gains in a political process which ‘aggregates’ the diversity of interests to create overall policies that define and distribute authoritative values within society. An integrative perspective, on the other hand, sees society as consisting of groups whose interests are not pre-fixed, but determined by the social, political, and economic context. In this system the political process integrates the individual and collective wills with the attitudes and traditions of the democratic community, which are themselves changed as part of the process (Sørensen and Torfing 2003: 615). The authors conclude that in the first instance networks are essentially detrimental to democracy. When emphasis is on the individual any divergence from the one-man-one-vote principle is obviously a minus. In the second instance, that is, in an integrative democracy, networks, on the other hand, may tend to improve the quality of democracy. The authors do not focus specifically on the EU, but they do consider the demos-issue. In this light they argue that: The incorporation of affected citizens can be an expression of a restricted view of the democratic demos that can easily lead to elitism as only the active and resourceful actors are included in the network. (Sørensen and Tofing 2003: 616)

One way to settle the question of the legitimacy of networks in the EU context would be to distinguish between the two types. From a normative perspective, or more precisely from the demos-perspective, the democratic contribution, or indeed democratic deficit inherent in a network model, would be different depending on whether the network in question was a policy network or an issue network. The looser and less stable the network, the greater chance that a higher number of interests would be represented. However, this position can also be criticized: Networks, another buzzword of the debate on globalization and democracy, cannot compensate for the equalizing function of commonalities since they are too fluid, and lack temporal and territorial continuity as well as serious mutual commitments among their members. (Thaa 2001: 516)

Hence, it would appear that even with the distinction between the two rather different types of networks it is still difficult to achieve a consensus as regards their legitimacy. 121

Models of Democracy for the EU

David Marsh offers a convincing explanation as regards this difficulty by linking the conceptions of the network model to theories of the state. Such theories should address four questions: (1) Who rules or makes policy? (2) How do they rule or make policy? (3) Why are certain actors in a privileged position in the policymaking process? (4) In whose interest do they rule and how does their rule result in that interest being served? (Marsh 1995).25 Network theory addresses mainly (1) and (2) whereas (3) and (4) are the key concerns of political sociologists and state theorists (Marsh 1995).26 Nevertheless, in his analysis of different schools of thought (on the state) and their interpretations of the network, we can perhaps find reasons for the different normative reactions. As theories of the state and the policymaking process, Marsh looks at three overall paradigms: Elitism, Pluralism, and Marxism.27 Within this framework, the Elitist angle and the Marxist angle would tend to see policy networks as a restriction of democracy because ‘the tighter networks reflect a consistent pattern of structured political inequality and are characterised by the exclusion of most interests’ (Marsh 1995: 11 and 20). Pluralists, on the other hand, have the opposite interpretation: ‘Policy networks are not a restriction on democracy because they are numerous, relatively open, being usually issue networks, and not all dominated by the same interests. Plurality equates with democracy’ (Marsh 1995: 15). This framework, is, however, linked to the way that the different traditions see the networks. Marxists and Elitists tend to identify networks as communities, whereas Pluralists see issue networks. It is therefore likely that the different reactions are caused by a different empirical description of the networks. In summary, the network is a useful descriptive tool when it comes to understanding policy processes in the European Union. From a normative perspective, however, the model is more problematic, particularly if one is concerned with equal representation. Even if informal networks allow a wider range of stakeholders access to the policy process (and in that sense provide a forum for deliberative democracy), this does not constitute equal representation because there are other actors in the polity than those who are stakeholders in a certain area. Hence, from a normative perspective Leonard’s promotion of the network model as an alternative to the institutionally secured channels of equal representation would 122

Models of Democracy for the EU

not solve the democracy problem (seen from a no-demos angle). The framework offered by Marsh is useful because it helps explain divergence of scholarly interpretations of the network by showing that these are likely to be affected by the paradigm to which the scholar belongs. From the no-demos perspective, the issue networks would be more welcome than the policy communities. However, since the EU legitimacy crisis is not only related to the demos but also to the relationship between EU citizens and institutions, the networks may very well be part of the problem rather than part of the solution. Whereas it is likely that networks contribute to efficiency of government, it is highly questionable whether they can contribute to democracy in the EU. In a polity which is struggling with popularity, has no demos and a reputation for lack of openness, systems with limited access and low transparency would be likely to contribute to the legitimacy crisis rather than solve it.

4.2.8. The Neo-Medieval Empire Model The neo-medieval empire metaphor has been developed by international relations scholars. This school of thought is particularly associated with Barry Buzan and Ole Wæver who focus on security identities and societal insecurity (see for instance Wæver et al. 1993). Although the work has now inspired a normative model for EU government (see below) the IR theorists initially developed it as an analytical framework. (The normative neo-medieval model is associated with Jan Zielonka and will be discussed later in this section.) However, in order to understand the normative argument, it is necessary to take a closer look at the concept of the neomedieval empire. The concept is based on an analogy to the Middle Ages when the European continent had no strict and stable sovereign state system (Wæver 1997: 321). Because European integration has removed some of the sovereignty from the nation states without having affected the national affiliations, the new system arguably looks like the pre-1648, that is, preWestphalian system. Hence, in essence, the medieval metaphor underlines the alternative to sovereignty and the ‘surprising presence of overlapping authorities’ (Wæver 1997: 322), whereas the empire metaphor refers to ‘a centeredness which is not that of a sovereign state, not sovereign equality, but diffuse patterns of centers with power fading off’ (Wæver 1997: 322). Further, Wæver argues that: 123

Models of Democracy for the EU The emerging political structures can be seen as ‘neo-Medieval’ in the sense that in Western Europe political organization no longer fits into the format of territorial sovereignty and exclusivity. The nation-states are not sovereign, but neither is the EU a sovereign state. [ . . . ] A post-sovereign system can find analogies generally from suzerain systems, and the integration/fragmentation argument points to the importance of a centred region. The medieval metaphor could thus be the best metaphor for provoking a rethinking of the concept of sovereignty, but less appropriate with regard to centeredness. (Wæver 1997: 356)

In other words, an empire can be seen as the ‘direct administration of different communities from an imperial centre’ (Watson 1992: 16). This, of course, sounds very hierarchical; however, in practice, the empire is restricted in its freedom of action by the constraints which involvement with other communities imposes (Watson 1992: 16). Although the national idea is not dying out, the organizing of the political space is being modified and in the EU, there is no longer any one level which is clearly the most important to refer to but . . . there is a set of overlapping authorities, and thereby even those nations most closely approaching the ideal type of the nation-state, lose the option of always referring to ‘their’ state. The close link between state and nation is broken’. (Wæver et al. 1993: 69)28

As regards the people of the polity, Wæver argues as follows: It can even be argued that European unification is exactly impossible if attempted in the format of the sovereign state. A strong reason for this is found in the area of societal security: nations defend themselves against a perceived threat to their identity from Europeanization. The nations will strike back—as the Eurocrats possibly learned during 1992—and the European construct has to be of a kind that does not confront the nations on their most sensitive issues; not a nation state, but a unit with more limited political identity. This creates a pressure in the direction of a complexity that could be labelled neo-medieval. (Wæver 1997: 356)

The model has been developed further by Jan Zielonka, first in two articles in which this author made his counter-proposal to Fischer’s Two-Tier federation (see Zielonka 2000, 2001), and more recently in a monograph in which the argument has been expanded (see Zielonka 2006). By building on a distinction between two scenarios, the ‘Westphalian type of state’ and the ‘neo-medieval empire’, Jan Zielonka has suggested an alternative vision to that introduced by Fischer in the Humboldt speech.

124

Models of Democracy for the EU

According to Zielonka, Fischer’s model is both state-centric and unrealistic (Zielonka 2000: 101) mainly because three of its features create more problems than they solve: (1) the state-centred argument emphasizes the state and its institutions rather than issues such as nation, politics, or markets; for instance, the democratic deficit of the EU is to be tackled by reorganizing the parliamentary system rather than by creating a European demos: Of course, it is easier to engineer or manipulate state institutions than culture, politics or economic markets. One can probably envisage the creation of a European state, but not a European nation. However, institutional engineering would be unable to enhance European integration on its own if there were only little progress in terms of cultural identity, political solidarity and economic interdependence. One does not need to endorse cultural or economic determinism to see that all institutions have their social, economic, and cultural prerequisites, which are usually ignored or underplayed by statists. (Zielonka 2000: 103)

(2) the state-centric argument suggests handing over more and more responsibilities to the EU. Even though the EU government may be federal rather than unitary, it will ‘basically resemble a modern version of the sovereign, territorial state that emerged in Europe following the 1648 Treaty of Westphalia’ (Zielonka 2000: 103). (3) the state-centric argument would like the Union to provide an overlap between its functions and geographical borders (Zielonka 2000: 103). First, Zielonka argues that enlargement will make it harder to obtain a European demos because of increased cultural divergence (Zielonka 2001: 314). To the extent that one agrees with the assumption that there were no EU demos before the 2004 enlargement, this argument is convincing. Secondly, because the Westphalian-type state relies on ‘hard borders’, the EU will never fit this description because of its history and future of continued enlargement. Regardless of increased EU cooperation with respect to external borders, increased police cooperation and even sporadic attempts at an EU military force, Zielonka does not believe that achieving such a level of integration is realistic: . . . the Union is far from possessing any effective military forces or relatively hard external borders. [ . . . ] The problem is that Europe’s defence identity is being built upon Europe’s foreign policy failures, which have repeatedly manifested

125

Models of Democracy for the EU themselves in the Balkans. One cannot help but ask whether a common EU defense policy is possible without a workable common foreign policy. After all, it is important to know how, when and for what purposes the Union is going to use its soldiers. If the Union’s common foreign policy is often in a state of paralysis, especially when faced with a crisis, how can a common defense policy ever work? (Zielonka 2000: 105)

Zielonka sees the EU system as neomedieval because The existing body of evidence leaves little doubt that the enlarged EU cannot become a Westphalian superstate. The Union is on its way to becoming a neomedieval empire with soft borders in flux, cascading socio-economic discrepancies, multiple cultural identities, and a polycentric government. Some features of this model were present before the fall of the Berlin Wall, but enlargement represents a decisive step towards medievalism that prevents other possible solutions from emerging. (Zielonka 2006: 167)

As regards the hard borders, Zielonka argues that: Prospects for acquiring relatively fixed and hard external borders as envisaged by the Westphalian state model are not bright either. [ . . . ] This is because increased global economic pressures and cascading interdependence make it difficult for any political unit to install a hard border regime . . . (Zielonka 2000: 106).

Zielonka’s overview of all the differences between the two models is provided below (Table 4.1). Common throughout the table is the simplicity of the statements in the first column and the far less clear definitions in the second column almost giving the impression they are two different levels of abstraction (which is probably the case). For instance, surely it is far easier to operationalize a statement such as ‘A clear hierarchical structure with one centre of authority’ than its counterpart in the neo-medieval model ‘Interpenetration of various types of political units and loyalties’. Secondly, at the normative level, it is hard to see how ‘Distinction between the European centre and periphery is most crucial, but blurred’ can possibly be superior to ‘Distinction between EU members and non-members is sharp and it is most crucial’. Surely, if something is that crucial, it is important that it is not blurred? However, Zielonka’s concern is not simplification of the system; in fact his central argument is that the relatively well-defined state model is not applicable to the EU. As regards the model’s descriptive value, Zielonka argues as follows: That said, the current plurality of different forms of governance, legal structures, economic zones of transactions, and cultural identities is striking and bears a

126

Models of Democracy for the EU Table 4.1. Two contrasting models of the future EU-system Westphalian superstate

Neo-medieval empire

Hard and fixed external border lines Relatively high socio-economic homogeneity

Soft border zones in flux Socio-economic discrepancies persist without consistent patterns Multiple cultural identities coexist Disassociation between authoritative allocations, functional competencies, and territorial constituencies Interpenetration of various types of political units and loyalties Distinction between the European centre and periphery is most crucial, but blurred Redistribution based on different types of solidarity between various transnational networks Diversified types of citizenship with different sets of rights and duties Multiplicity of various overlapping military and police institutions Divided sovereignty along different functional and territorial lines

A pan-European cultural identity prevails Overlap between legal, administrative, economic, and military regimes A clear hierarchical structure with one centre of authority Distinction between EU members and non-members is sharp and important Redistribution centrally regulated within a closed EU system One single type of citizenship A single European army and police force Absolute sovereignty regained

Source: Zielonka (2006).

remarkable resemblance to the situation in medieval Europe. There can hardly be a greater contrast than the one between the Union of today and the classical Westphalian state, which is characterized by a low degree of economic, legal, and cultural divergence and a high degree of overlap between functional and territorial borders. (Zielonka 2006: 165)

This is a convincing argument and at this point it is relevant to reiterate the title of the book ‘Europe as Empire’, not the European Union as empire. Hence, this could also serve to illustrate the fact that the structures governing Europe are not only nation states and the EU but also other supranational, or at least, international regimes, such as the OECD and the European Convention of Human Rights (see Majone 2005: ch. 1, for a discussion). At the normative level, Zielonka states the following: I have no hesitation in concluding that Europe should embrace a neo-medieval EU. A neo-medieval empire would be in a good position to stabilize the external environment, manage the presence of overlapping structures of power, and facilitate the flow of goods, services, and people throughout the EU’s borders. (Zielonka 2006: 166)

127

Models of Democracy for the EU

Zielonka is as worried as many other scholars about the Union’s lack of legitimacy and democracy and offers a variety of ideas as to how these issues should be addressed within the empire framework. In the field of democracy, enlargement also reinforces the neo-medieval pattern. The structure of democratic governance is likely to become increasingly multilevel and multicentred. The system will lack a clear allocation of competences and straightforward hierarchy. A single powerful centre of authority as exists in all Westphalian states is not going to emerge in the enlarged Union. Nor can we expect the European Parliament to be the main vehicle for democratic representation. Instead we are likely to see various non-majoritarian institutions dominating the weak European Parliament (but also the increasingly weak national parliaments). The public space in the enlarged EU is likely to remain fragmented and weak preventing the rise of a single and dominant European identity, ethos, and demos. Enlargement has further increased the plurality of civic cultures in the Union making it difficult to arrive at the fairly homogenous cultural pattern required by the Westphalian model of democratic governance. (Zielonka 2006: 169)

In some respects, the neo-medieval model (as an analytical or descriptive tool) resembles the network model. Both are non-state-centric and point—amongst other things—to the overlapping power structures of a system of government. Also, at the normative level, Zielonka suggests that networks may contribute to democracy and legitimacy in the EU. He is, however, much more aware of the pitfalls of such solutions than mainstream network theorists. For instance, he argues that: Even noble plans to involve the institutions of civil society in European decisionmaking may in the end constrain rather than enhance democratic participation. Civil society representatives are not elected by the people (and usually not even by their members), they do not offer equal access to decision-making, and they pursue parochial agendas. (Zielonka 2006: 186)

Zielonka recognizes that the most ‘difficult challenges will probably come from the field of democracy and identity’ (Zielonka 2006: 166) and states that some ‘democratic quality will surely be compromised, but there are also some obvious gains for democracy in the new Europe’ (Zielonka 2006: 183). Amongst the solutions are strengthening representation at the local and national levels, network governance (although he recognizes that the nature of the networks would have to change in order to enhance democracy), as well as the enforcement of civic and political rights. There is nothing in the argument that would not be welcomed from the no-demos perspective. Indeed, all the weaknesses of the state-centric 128

Models of Democracy for the EU

model are addressed and solutions offered. The problem, however, is that the solutions are rather abstract in their wording and that the system is inevitably extremely complex. At the moment, increased complexity is perhaps not what the EU citizenry needs. On the other hand, there is no doubt that the neo-medieval empire metaphor (the term preferred by Wæver) is a useful analytical tool as opposed to a normative recommendation. It provides an empirically strong criticism of both the federal, the parliamentary and the two-tier models, and it is particularly successful in breaking with the nation state analogy trap which is so hard for theorists to overcome. In particular, it questions that a strengthened EP is the solution and from the no-demos perspective this is a key point. Indeed, applied to the EU, the state-centric model has many weaknesses and most of Zielonka’s criticism of using that framework is welcomed. But it is difficult to see how the medieval model becomes the only alternative. Are there really only two options? It may very well be that the EU resembles a medieval empire. However, in a union where lack of transparency and clarity is causing legitimacy problems it is hard to see how the above suggestions would contribute to a solution at least in the short term.

4.3. Summary and Preliminary Conclusions The models are summarized as follows (Table 4.2). The notion that the EU can be understood and designed in terms of one particular model of democracy is simplistic. No real system of government or governance is ever equal to an ideal type, precisely because ideal types are abstractions, and they all underline particular aspects in an analysis of the empirical world. Hence, the conclusion of this chapter will not contain a dismissal of seven of the models only to end up with a magic formula provided by the one particular recommendation. However, in the following paragraphs I summarize the analytical and normative values of evaluating prospects for EU reform in these abstract terms.

4.3.1. Descriptive Strengths of the Models If the EU ever really was a confederation, it has surely crossed the line where this was the case. The supranational elements, particularly those of 129

Model

Confederal

Federal

Parliamentary

Keywords

Institutionalized intergovernmental cooperation. No supranational elements. Division of powers between federal level and segments Judicial review.

Parliament prominent role Chief executive supported by majority in legislature Executive power able to dissolve legislature.

Descriptive Model

Normative model

Strengths

Weaknesses

Strengths

Weaknesses

Appropriate for understanding intergovernmental processes.

Inappropriate for supranational areas of EU decision-making.

Ensures legitimacy via national domestic processes.

Joint decision-trap and loss of acting-ability. Effectively, a reversal of integration.

Takes account of multi-level governance. Also, the constitution clarified principles for division of powers between levels Can be useful but mainly for comparative purposes (i.e. when comparing EU institutions to parliamentary democracies).

Not applicable to intergovernmental policy areas. Member states can withdraw from the EU.

Could possibly create clarity of grey areas of divisions of powers and combine different sources of legitimacy.

Does not seriously address problem of the absent demos as federalizing may rely on constituent demos.

EP cannot dismiss the Council. Executive cannot call untimely EP election.

Looks like nation-state model that publics recognize, providing familiarity.

Based on inappropriate analogy. Too focused on state-EU comparisons. Does not take account of the lack of an ‘EU people’. This is a problem due to the majoritarian principle.

Models of Democracy for the EU

130 Table 4.2. Models of democracy for the EU

Table 4.2. Models of democracy for the EU (Cont.) Legislature has a fixed electoral mandate, own source of legitimacy Executive has fixed electoral mandate, own source of legitimacy.

More suitable than the parliamentary model. The EP does have a fixed electoral mandate The executive, however, is not directly linked to a collective electorate.

Although the EU has a president of the Commission and a presidency of the Council there is presently no direct election of a president, nor has such a process been incorporated into the new treaty.

Direct election of a president could perhaps unite the European citizens and give the executive legitimacy.

Consociational

Consensus (elite-driven) democracy within the EU political system.

Analogy between state and segments may not be appropriate.

Could quite possibly be a way of overcoming the no-demos problem, but . . . .

Network

Multi-level system. Access to political process for interest groups and other actors. Federal statist model with optional enhanced cooperation segment inequality. Concentric circles, core-periphery.

Captures many features of the EU system, i.e. veto, grand coalition, elite-driven project, and elite bargaining. Takes account of actors and processes not captured by more ‘traditional’ models.

An additional framework rather than full model of democracy.

Offers alternative and new ideas as regards democracy and legitimacy, but . . . .

Takes account of opt-outs.

Only applicable in some policy areas such as EMU and Schengen.

Describes the European situation well.

Too complex for the present situation.

Will give the states flexibility to participate and not participate in a range of policy areas . . . Gives valid criticism of the statist ambitions, including by addressing the no demos-issue.

Two-tier

Neo-medieval

On the other hand it could be argued that direct election necessitates the existence of a demos and that under the current circumstances it would be impossible for one individual to represent the EU’s citizens. . . . may also prevent development of demos.

. . . does not focus on equal representation. Access to networks may favour some groups. . . . but may lead to inequality of citizens because of inequality of segments. Channels of democratic representation not very clearly defined.

131

Models of Democracy for the EU

Presidential

Models of Democracy for the EU

the community pillar, are beyond what could be described as confederal. Thus, when looking at its structures—at first glance—the EU seems to have many federal features, particularly because of the division of powers between the two main levels of government, that is, state and federal level. However, the fact that there is no common security and defence policy (as in no supranationally decided security policy) and more importantly the fact that it is possible for states to opt-out of certain areas of the system mean that the federal label does not fit completely. Furthermore, the unwritten (although stated in the rejected constitution) freedom of exit clause is a provision that does not exist in other political federations. Hence, the relationship between state and federal level differs from that in other federations. This is also the case regarding the parliamentary model. Although the EP has been given an increasing amount of power, the fact that the chief executive is not supported by a majority in the legislature, that is, that the EP does not form the government, means that the parliamentary label does not fit either, particularly since the executive does not have the power to dissolve the legislature and call a new election. In fact, when it comes to the sources of legitimacy of the legislature the EU looks more like a presidential model, because the EP does in fact have a fixed electoral mandate and because the executive cannot (as in a parliamentary system) dismiss the legislature and call for a premature election. The executive, however, is not directly linked to the electorate because the Commissioners (and thus the Commission president) are appointed and the Council only enjoys direct legitimacy at member-state level. This means that the Council has no direct, collective democratic legitimacy. (This need not be a problem from the no-demos perspective.) The network model, the two-tier model, and the neo-medieval model all have important contributions to make in that they capture structures that are not encapsulated by other models. Extensive lobbying and the strong involvement of unelected professionals both as regards policy formation and implementation mean that the network model has much to contribute when it comes to understanding where the power lies. The two-tier model (in its broadest sense) encapsulates the fact that not all member states are involved in all policy areas, and the neo-medieval model highlights the problem that has arisen as a result of the perceived reduction in state-acting ability coupled with a segmentation of the nations. Hence, although no single model is sufficient for understanding the EU, taken together these models all have important contributions to make. The neomedieval model is particularly valuable because better than any other, it 132

Models of Democracy for the EU

summarizes the difficulties that arise when governance is taken beyond the nation state. Despite the increased use of qualified majority voting in the Council, the consociational model has important contributions to make as regards policymaking in areas under pillars 2 and 3. This argument takes account of the critique raised by Bogaards—that the states may not be equivalent to Lijphart’s ‘segments’—because this is not an institutional but more so a structural critique. The greatest weakness at the descriptive level is the size of the polity because the traditional view was that a consociational system was only found in relatively small countries. Also, the EU does not satisfy all of Lijphart’s ten criteria although it is very close in a number of respects, particularly as regards the veto and secret deliberation. However, looking at the rhetoric of the Convention it is clear that the consociational element is strong. For instance, throughout the constitution, as well as the other interim documents produced by the Convention, the inhabitants are referred to as ‘the peoples of Europe’. This is a strongly consociational (and confederal) rather than federal discourse and even though the constitution is now unlikely to be ratified (certainly in its present form), there is nothing to suggest that this will change.

4.3.2. Normative Strengths of the Models In comparative political studies, the literature on democratic systems often distinguishes between presidential and parliamentary systems as if it were a matter of either one or the other. This is due to the fact that by nature, this literature is based on empirical studies and comparisons of democratic systems. However, from a normative perspective, the choice for the European Union is not a matter of either a presidential or a parliamentary system. From a no-demos perspective neither model is appropriate. Whether one of them will be increasingly suitable in future is not just a question of time, although there seems to be an unspoken consensus amongst many academics and practitioners which assumes that people will gradually get used to the EU and come to understand its benefits and somehow become increasingly comfortable with European institutions. However, these are not causal arguments and developments so far seem to point in the opposite direction. And as mentioned in Chapter 2, in time anything can happen, hence unless one has very good arguments why time should contribute to change, one should probably not rely so much on time to set the direction of change. The truth is that both the parliamentary model and the presidential model require a demos 133

Models of Democracy for the EU

in order to be legitimate solutions because both models rely considerably on majoritarian principles. The main difference between the models is that a presidential system sees a direct link between the electorate and both the legislature and the executive, whereas the parliamentary system links the executive to the legislature and only links the executive and the electorate indirectly. (This is also the case if one sees the Council as part of the executive because national electorates are only indirectly linked to the Council.) However, both cases treat the European electorates as one electorate, but from the no demos-perspective this is not viable precisely because if there is no demos then it follows that there is no electorate but multiple electorates. Unlike both of the above, the consociational model does address the nodemos issue by emphasizing the importance of the veto. However, from a normative perspective its weaknesses are the habits of elite bargaining as well as the worry that it may prevent the development of a demos. On the other hand, the latter could turn out to be a blessing in disguise, particularly if the demos necessitated the creation of an identity that relied on the EU defining itself against ‘others’ (a concern expressed by many authors, notably Smith 1992). It is recognized that networks are an extensive part of the EU governmental process. The network model is useful because it captures actors and processes not encompassed by the more ‘traditional’ models which tend to focus on representative institutions. However, as a normative model it is perhaps more useful as a critique of the federalist models than as an alternative solution. Having said that, even if there are networks that contribute to the democratic process in the sense that they serve to provide a link between the governed and the governing (or even to a blurring of that demarcation line), the EU still needs an institutional framework. Hence, it is still necessary to consider how the institutions best cope with the polity’s absent demos. Finally, because the EU is also battling with a perceived lack of transparency it would probably be unwise to credit the network model with too much democratic impact. In fact, the contrary may be the case. As regards the institutionalized—or perhaps even constitutionalized— two-tier model, this would do nothing for the European demos because it would mean that EU citizens were no longer equal. However, the flexibility of such a system could serve as a stabilizing force because national publics would have the security of knowing that they could opt-out of cooperation in certain policy areas. But the idea of a federal Europe with a hard core of members is troublesome seen from the no-demos perspective, 134

Models of Democracy for the EU

although in a confederal system it would be easier to accept case-by-case cooperation. This is due to the fact that a confederal system does not require the formal equality of the parts (because a confederal system is not binding), whereas this would be a prerequisite for a democratic federal system. From a normative perspective, the strength of the neo-medieval models lies in its insistence that democracy beyond the nation-state level should not rely on majoritarian institutions. Its weakness on the other hand lies in its lack of concrete suggestions as to how a neo-medieval democracy could be built. It proposes a range of solutions but at a higher level of abstraction than, say, a new constitutional convention could consider. On the other hand, it provides direction which those potentially responsible for proposing new legal structures could take into consideration. The question, however, is whether an increasingly puzzled European citizenry is prepared to accept such innovative measures. There is, of course, another solution to the socio-psychological problem which has not been included in the above list, namely, to dissolve the union and return transferred powers to the nation states: Discussions on the future of European integration will never come to an end, and proposals demanding an abandoning of the whole endeavour may always be heard. (Wessels 1996: 69)

Although this is obviously not a real solution, it should not be ignored that there are numerous voices across the European political scene that continue to advocate their country’s withdrawal from the union. (In fact, for integrationists, it is hugely important to listen to those voices so that the concerns they represent can be addressed.) Mostly, however, anti-integrationists do not call for the dissolution of the union, only for the withdrawal from the union of their particular country or even the withdrawal from participation in certain policies. However, if one state were to leave, particularly a large state—perhaps due to domestic public pressure during a crisis—the result could be an effective collapse. First of all, such an exit could stir anti-‘integrationism’ in other member states resulting in a domino effect which would only be exaggerated by the eventual loss of economies of scale that could be caused by exits. In addition to causing exorbitant administrative and governmental chaos, which in itself could lead to unrest, a collapse would obviously undermine the discourse, or mode, of transnational cooperation which has been the ‘European way’ since the end of the Second World War: 135

Models of Democracy for the EU . . . there is no alternative to continuing the process of European integration. Whether ‘Europe’ will become a supranational entity in accordance with the optimistic view or whether it will merely comprise co-operation between states and governments is of secondary importance. In both cases, the unification process provides the historical answer that a number of problems can no longer be resolved in the narrow decision-making framework of the nation-state. European integration is nothing more than an attempt to harmonize the decision-making structures of politics with the migration of problems. (Decker 2002: 257)

It is unknown why Decker describes the ‘supranational entity’ as a result of the ‘optimistic view’, but in fact, this is one of many statements in the literature on European integration which give observers the impression that only supranational integration is considered real progress and that those who speak a more confederal language are ‘second best’ or even setbacks to the integration process. However, a real setback would be withdrawals and ever-diminishing public interest and support. Such things are likelier to occur in an environment in which elites continue to ‘federalize’ with almost no regard for expressions of popular opinion. I shall return to this discussion in Chapter 7. In conclusion, although no abstract model of democracy can reasonably be proclaimed the perfect solution for the EU, these models are useful for those concerned with sources of legitimacy. If a system is desired which would necessitate the existence of some degree of an EU demos— as opposed to a multiple demoi legitimacy—the EU must adopt a model which does not obstruct the development of such a demos in which case the two-tier or consociational models may not be appropriate. On the other hand, if the EU were to accept that the demos would probably be relatively weak in future but that there exists strong backing for the continued institutionalized cooperation, a confederal or consociational system with flexibility of cooperation would be preferable. I shall return to this discussion after the empirical analysis.

Notes 1. Some of the concepts listed here as ‘models’ can also serve as prerequisites for each other. For instance, the consociational model is said to work best in a federal system. In such an argument, federalism is a prerequisite for a consociational democracy and is thus seen as structure rather than a system of government.

136

Models of Democracy for the EU 2. For instance, one scholar talks about a ‘consociational confederation’ (Chryssochoou 1994). However, conceptual overlaps or blends are not always this explicitly defined. 3. This is expressed convincingly by Sartori who argues that ‘Alertness to the distinction between is and ought immediately shows that it is nonsensical to argue that the ideal of something is superior to the reality of another’ (Sartori 1987: 13). 4. Qualified majority voting is now the most usual decision-making mechanism in the Council, although in practice there is still heavy reliance on consensus even when QMV is the rule (Hayes-Renshaw and Wallace 2006: 56). As regards, the application of the co-decision procedure this was expanded both in the Treaty of Amsterdam and to a lesser extent in the Treaty of Nice (Nugent 2003: 347–8). Nice added areas previously confined to pillar 3, including some antidiscrimination measures, judicial cooperation in civil matters (except family law), and some specific areas under industrial policy (Nugent 2003: 88). Also, the assent procedure now (since Nice) applies when enhanced cooperation concerns an area that falls under the co-decision procedure and when/if establishing whether a member state is at serious risk of breaching the principles on which the EU was founded (Nugent 2003: 89). 5. Provided, of course, that all do not agree on everything, all the time. 6. However, a confederation need not lead to a federation. 7. Whether a federal EU would mean the most integrationist position would depend whether a potential federal EU was to be considered a federal system or a federal state. 8. One could possibly argue that reluctance to accept a federal future for the EU does not only exist due to collective ignorance of the term ‘federal’. The fact that federalism is a way of securing segment autonomy might in itself provoke hostility because it may be taken as an indication that the central level of government needs to be kept in check, i.e. that ‘Brussels’ has already become too powerful. 9. In this rhetoric, the national level equates the federal level, which is not the case for the EU, where the national level equates the unit level. 10. The danger is that should it become increasingly unclear to the electorate who is ‘left’, ‘right’, government, and opposition, a crisis—such as high unemployment or recession—could lead to a loss of legitimacy of the structures rather than ‘merely’ the current power holders. This again could result in an activation of the search for authoritarian, populist or secessionist alternatives, or major institutional modification of the liberal democratic regime (Offe 1995: 28). 11. The idea that majoritarian democracy is closer to a democratic ideal is quite possibly flawed (Lijphart 1999: 6; Dahl 1989: 185) because majoritarian democracy is not suitable for all types of societies.

137

Models of Democracy for the EU 12. If participation is an indication of the mandate given to an elected assembly, the mandate of the EP is significantly smaller than that of national European assemblies. In the June 2004 EP elections, the turnout was 45.7%, whereas average turnouts in European national elections are in the region of 83%. (International Institute for Democracy and Electoral Assistance, 2004.) In fact, overall participation in EP elections has dropped from 63% since the first direct election in 1979. (For individual countries, this is not a linear decrease, however. For instance, between 1989 and 1994, five of the then 12 member states saw an increase in participation.) 13. In fact, David Easton, without using the actual word ‘consociationalism’ described the workings of such a system as far back as 1955: ‘. . . in highly stratified political systems, various social and political classes may communicate with each other only indirectly for the most part, intermediate elites acting as communication links and focal points of support. The membership of the whole system may have only vague feelings of belonging to the system but they may have strong attachments to the subunits of which they are part. Yet this low input of support from the ordinary members of the system may not pose a threat if each intermediary leadership itself evinces strong attachments to the next higher level of leadership’ (Easton 1955: 227). Lijphart himself, however, credits another author with having invented consociationalism stating that it was originally developed as a result of an inductive approach by Sir Arthur Lewis who—without knowing empirical cases—suggested a consociational model (without using the specific term) as a way of government in the ethnically divided societies of west Africa (see Lijphart 2002: 166). 14. To the question ‘who will do the governing and to whose interest should the government be responsive when the people are in disagreement and have divergent preferences?’ Lijphart offers two possible answers. The first is the majority of the people: ‘The majoritarian answer is simple and straightforward and has great appeal because government by the majority and in accordance with the majority’s wishes obviously comes closer to the democratic ideal of ‘government by and for the people’ than government by and responsive to a minority’ (Lijphart 1996: 2). However, the consensus model offers another way: ‘The alternative answer to the dilemma is: as many people as possible. This is the crux of the consensus model. It does not differ from the majoritarian model in accepting that majority rule is better than minority rule, but it accepts majority rule only as a minimum requirement: instead of being satisfied with narrow decision-making majorities, it seeks to maximise the size of these majorities. Its rules and institutions aim at broad participation in government and broad agreement on the policies that the government should pursue’ (Lijphart 1999: 2). 15. These characteristics were established empirically after studies of Switzerland and Belgium.

138

Models of Democracy for the EU 16. Note that here Lijphart uses the term ‘nation’ interchangeably with that of ‘member state’. 17. Note that Lijphart’s book from 1999 was published prior to the treaties of Amsterdam and Nice which both gave increasing powers to the EP. 18. Perhaps the assumption that the EU is either one or the other is not fully justified. 19. This is a perfect example of a clash between the descriptive and the normative strengths of a model. 20. The tensions that developed between some older member states and a number of the 2004 member states in connection with the Iraq war could tend to increase such worries. 21. In this context, it should be reiterated that the term ‘model’ refers to an ‘ideal type’ rather than a model that illustrates a causal relationship. 22. The literature on policy networks was initially developed as a strain of new institutionalism (see George and Bache 2001: 21). 23. Interestingly, Leonard uses the phrase ‘one of the other visions’ having just said that the network was a ‘fluke’ and therefore presumably not a vision. 24. The idea that we can start to reform the EU ‘rationally’ is possibly a slightly optimistic interpretation of the situation (see Olsen 2002). 25. Although this chapter does not make specific reference to the EU, the argument and typologies are arguably of a universal nature and hence applicable to a broad range of objects of study, including the European Union. 26. Clearly (as shown by the Leonard and Sørensen and Tofting articles) the literature on networks has developed to include the other questions since Marsh’s analysis was published; however, Marsh’s framework is still useful in highlighting different interpretations and reactions to the network model. 27. This is a similar distinction to that offered by Dunleavy and O’Leary, although these authors also include The New Right as a paradigm (1987). 28. One important difference between the emerging picture and the Middle Ages is that in the Middle Ages, the nations did not exist (Wæver 1993: 69). 29. Whether the emphasis in network approaches is on agency or structure depends on the definition one applies. Networks are occasionally described as referring to structures or structural configurations but at other times seen as having ‘relatively autonomous action units’. (See Thatcher 1998 for a discussion of this problem.) In the context of the EU, however, the emphasis is on agency due to the relatively loose configuration of the structures which exists as a result.

139

This page intentionally left blank

5 The framework for empirical testing

5.1. Introduction This chapter opens with a discussion of the methodological problems that face the next part of the project and of ways in which it is intended to reduce the impact of such problems, particularly the link between the theoretical and analytical definitions of the demos and its eventual operationalization. The latter raises a number of questions, the two most central ones being: How can a concept as abstract as the demos be operationalized to the extent that it is measurable; And how would one acquire data suitable to answer such a question?

5.2. Defining a Demos The theoretical point of departure is the conclusion from Chapter 3, which established that it is possible, and indeed necessary, to distinguish between the legal and the social demos. It was concluded that a legal demos can be created via the formation of civil rights and citizenship within a polity, but that a social demos is more complex because it entails such elements as common history, language, identity, and solidarity (towards the polity as well as amongst individuals). These are all elements that are harder to measure, let alone create. Having said that, there appear to be two categories of elements of the social demos. The ethno-cultural elements are matters such as language, history, political habits, ethnic origin, and religion. As these are relatively well documented, they will not be the focus of this project. Instead, Chapter 6 focuses on the other elements of the demos, that is the subjective emotional elements such as solidarity and identity.1 These are different to the ethno-cultural elements. Where the values of the latter can often be established without the actual 141

The Framework for Empirical Testing

involvement of the individuals being studied, establishing the values of the subjective emotional variables necessitates the direct cooperation of these individuals or a sample of the population. As was established in Chapter 3, there is consensus in the literature that no EU demos exists at the present time. However, there is some disagreement as regards the likelihood of a demos emerging and also as regards the desirability of such a development. Establishing the nature and existence of the demos thus necessitates a processing of this concept to translate it from the abstract into the concrete. Thus, one must begin by establishing a theoretical definition of the concept. This will be done on the basis of the established literature on the topic and although it is probably difficult to achieve a universally accepted definition of most concepts, the real problems start once the theoretical definition is operationalized. This process has a range of potential pitfalls. In the following paragraphs I explain how I aim to reduce their impact on the research.

5.2.1. Validity Once a theoretical definition of the concept to be measured is established (in this case the demos) it must be made suitable for the empirical world, that is, be operationalized. It is at this stage, when the researcher is moving from the theoretical to the empirical level, that problems with validity are likely to arise. And for operationalizations, similar reservations apply to those raised during the definitional stages in that it will rarely be possible to achieve consensus on one particular definition or operationalization. But again, all one can aim for is that others are able to follow the operationalization and hence are able to follow the logic of the subsequent argument which will be made by inferring from the measurable, operationalized variables to the concept. Inference in its general sense means that we use ‘the facts we know to learn about the facts we do not know’ (King, Keohane, and Verba 1994: 46). As regards inferential statistical methods these can be said to ‘provide predictions about characteristics of a population, based on information in a sample from that population’ (Agresti and Finlay 1997: 5). The question in this regard is how to ensure that the facts about the people one has spoken to really provide actual information about the people to whom one has not spoken. The problem in Chapter 6 is related although slightly different. Here the question is how one can ensure that the ‘facts we know’ really do provide information about the phenomenon we cannot measure. The 142

The Framework for Empirical Testing

answer is by trying to avoid indicators that are far from the original concept and by making sure that the abstract general terms are connected to the specific measurable concept in order to allow empirical testing (King, Keohane, and Verba 1994: 111). Thus, validity can be defined as the extent to which ‘the measurements reflect what the investigator is trying to measure’ (King, Keohane, and Verba 1994: 151), and can be measured as ‘the degree to which the operationalized definition corresponds to the theoretical definition’ (Hellevik 1997: 43). An explanation is provided below which repeats the definition arrived at in Chapter 3 and outlines the operationalizations used in Chapter 6, although at this stage it should be noted that the type of inference that takes place is predominantly descriptive rather than causal. A clear distinction between these types is offered by King, Keohane, and Verba: Scientific research is designed to make descriptive or explanatory inferences on the basis of empirical information about the world. Causal descriptions of specific phenomena are often indispensable to scientific research, but the accumulation of facts alone is not sufficient. Facts can be collected (by qualitative or quantitative researchers) more or less systematically, and the former is obviously better than the latter, but our particular definition of science requires the additional step of attempting to infer beyond the immediate data to something broader that is not directly observed. That something may involve descriptive inference—using observations from the world to learn about other unobserved facts. Or that something may involve causal inference—learning about causal effects from the data observed. (King, Keohane, and Verba 1994: 8)

5.2.2. Reliability Once the suitable indicators have been identified and data have been obtained, the next step is to make sure that one uses the data correctly so that the results are reliable, meaning that ‘different measurements of the same phenomenon yield the same results’ (King, Keohane, and Verba 1994: 151). By ensuring high degrees of validity and reliability, another important requirement can be fulfilled, namely that the logic, and steps that lead to one’s conclusions, can be followed and replicated by other researchers. Apart from ensuring a reasonable quality of research, this also makes it harder for researchers to manipulate science with the aim of promoting a specific agenda. Thus, to the extent that this is possible, all data and analyses should be replicable so that observations and experiments are intersubjective. (Gilje and Grimen 1993: 19; King, Keohane, and Verba 143

The Framework for Empirical Testing

1994: 26). In the paragraphs below, I outline the process that is used in Chapter 6. First the theoretical and analytical definitions of the demos are repeated from Chapter 3.

5.3. Operationalizing and Measuring a Demos It has so far been established that the demos consists of two main elements, that is, the legal and the social parts. Further, it has been established that the social demos is constituted of two main categories of elements, that is, the ethno-cultural elements (that will not be investigated here) and the subjective emotional elements (the focus of this book). These are equivalent to Weiler’s ‘subjective manifestations of peoplehood’ (Weiler, Haltern, and Mayer 1995: 12). However, in the present context, the term ‘subjective emotional elements’ is preferable because it would not be desirable to include in one’s definition the word that one is actually defining. Nevertheless, these subjective manifestations consist of: (1) (2) (3) (4)

shared sense of social cohesion; shared destiny; collective self-identity; and, loyalty (Weiler, Haltern, and Mayer 1995: 11).

In a similar vein, but even more concretely Chryssochoou offers the following definition: A European demos can be said to exist insofar as the constituent publics see themselves as parts of a democratic whole, and are given the institutional means to mark their impact on the governance of the composite polity, thus creating the basis for a self-directed and self-determining collectivity. (Chryssochoou 2001: 252)

This definition is useful in that it includes the essential characteristics of what constitutes as demos that were discussed in Chapter 3. Other than defining the demos in general theoretical terms, Chryssochoou has broken the concept down and split it into subdefinitions that are easier to operationalize than the more general definition. Thus, based on the two theoretical elements: (1) The constituent publics see themselves as parts of a democratic whole, and (2) the constituent publics are given the institutional means to mark their impact on the governance of the composite polity, 144

The Framework for Empirical Testing

Chryssochoou has derived four properties that can serve as indicators for the demos: (i) democratic self-consciousness of the collective citizen body; (ii) adherence to shared democratic values; (iii) public awareness of the transnational process; and, (iv) a desire to shape democratically the future of a plurality of interrelated peoples without endangering the existence of that plurality (Chryssochoou 2001: 252).

These are the indicators that are measured in Chapter 6. However, they have been amended slightly from Chryssochoou’s original proposals: ‘Adherence to shared democratic values’ has been expanded to ‘adherence to shared values’. There are two reasons for this. First of all, it is not always easy to distinguish between a ‘democratic’ value as opposed to any other value. There are values that some see as moral values and other would see as democratic values. For instance, to some commentators the death penalty could not be a feature of a democratic political system whereas others may have no problem with this (although they may oppose or support it on moral or religious grounds). The second advantage of excluding the word democratic is that there may be other values which are as important for successful transnational democratization, even though they are not directly related to democracy. This is because such values could be part of the glue that binds a group of people together. The second change has been made to indicator four, ‘a desire to shape democratically the future of a plurality of interrelated peoples without endangering the existence of that plurality’. The indicator has been reduced to ‘a desire to shape a common future’. First of all, it is impossible for anybody to predict whether the existence of the plurality will be endangered. Secondly, it would be extremely difficult to measure both the desire to shape a common future and the desire to do so democratically at the same time. Instead, people’s views on democracy in general will be investigated under indicator two. In sum, the four indicators tell us something about the existence of a demos in various, complementary ways. It must be stressed that it is recognized that others might have chosen different variables to highlight these relationships. However, it is my hope that the openness about the process whereby variables have been selected in this study will at least serve as ‘damage limitation’ in the sense that critics will be able 145

The Framework for Empirical Testing

to follow the process and hence identify potential weaknesses in the evidence.

5.4. Data, Indicators, and Variables All of the proxy variables are questions selected from the Eurobarometer surveys and the European Value Survey. Thus, the analysis relies entirely on secondary data. This raises potential criticism. First of all, a researcher using secondary data can find it hard to break free from the mindset and conceptual frameworks, the priorities and the perspectives imposed by those who originally defined and operationalized the topic (Miller 1997: 155). The danger is that the researcher may design his or her questions to ‘fit’ the data, rather than collect data after having defined the research question. On the other hand, this need not be a disadvantage if the new research question is a relevant one and the findings contribute something new to the research area (Miller 1997: 155). Furthermore, the secondhand research needs to ensure that the data are reliable in the sense that the data gathering has taken place appropriately since obviously one has no way of influencing the sampling procedure. However, if the researcher is able to follow the specific process that has been carried out by the primary researchers the implications of this problem are reduced.

5.4.1. Data Sources THE EUROBAROMETER SURVEYS The Eurobarometer surveys have been produced since 1973 and provide the best data available for comparing all EU member states as well as accession countries. The reports are biennial although at times special reports are produced covering particular policy areas such as health or defence, making it relatively easy to trace developments over time. The sampling and data-gathering methods used by Eurobarometer are generally accepted by the academic community as ensuring maximum reliability and validity. As regards sampling methods, weighting and other methodology, these are all publicized and technical information is available on the internet. Detailed files on methods are provided together with the data. This, however, does not mean that Eurobarometer can be regarded as infallible. First of all, the study is commissioned by an EU institution, the Commission, which could perhaps raise questions about independence. 146

The Framework for Empirical Testing

Secondly, the questions are—at times—slightly manipulative (examples will be given in Chapter 6).2 The mere fact that Eurobarometer is generally accepted as a valid and reliable data source by the academic community does not relieve the researcher from his or her responsibilities as regards checking the validity and reliability of the data provided by this source. A common problem is the actual wording of the questions. There are occasions when a question is asked which includes a bias that is likely to make the respondents favour a particular answer. (In the few cases where such a question has been included in Chapter 6, these problems are addressed in the analysis.) On the positive side, the wide availability of the Eurobarometer surveys, descriptive statistics, and actual data ensures that the results are scrutinized by those commentators who access the material. Furthermore, anybody can—via the Europa website where Eurobarometer publishes reports and the aggregate data—gain access to a statement that outlines Eurobarometer’s background and methodology. Thus, Eurobarometer fully adheres to the principle stated by King, Keohane, and Verba that in a quantitative opinion poll, recording the data-generation process requires that we know the exact method by which the sample was drawn and the specific questions that were asked (King, Keohane, and Verba 1994: 23). Hence, there may be times when researchers do not agree with the conclusions that have been printed in the Eurobarometer reports. However, due to the availability of the data it is possible for those interested to track the process that has been used by the Eurobarometer institutes. Furthermore, it is possible via a range of data archives to gain access to the actual data sets so that one can subject those data to secondary analysis and carry out one’s own analyses depending on the questions, should one’s research question be of a different nature to that which the data was originally intended to answer. The great benefit is that data can be easily compared between countries and over time. Three types of references will be made to the Eurobarometer material. Frequently, I refer to the Eurobarometer annexes, that is, the tables wherein Eurobarometer publish the results (aggregate by country). Occasionally, it will be necessary to refer to the actual Eurobarometer report, meaning the report that is produced on the basis of the result tables, which are again collated on the basis of the raw data. Finally, in a number of cases references will be made to Eurobarometer data, meaning the actual data files. As explained, there are occasions when it is relevant to process the data when analysis that may not have been interesting for 147

The Framework for Empirical Testing

the purposes of the original report for which the data was generated are particularly relevant for the current purposes. THE EUROPEAN VALUE STUDY The European Value Study is a large-scale cross-national survey programme which focuses on values. The data obtained are from the 1999– 2000 survey. In principle, similar reservations apply to the European Value Survey (EVS) as to Eurobarometer as both are secondary data. However, the EVS has perhaps a higher credibility as it is independent (whereas Eurobarometer is coordinated by the European Commission) and its focus is values in general as opposed to the EU. Furthermore, the survey is very relevant (especially for indicator two) because it contains a high amount of value-related as opposed to attitudinal questions. Finally, the EVS contains data from a larger range of countries, including the recent members, allowing for the inclusion of these countries as well. EXCLUDED MATERIAL In addition to accounting for the material that is included, it is perhaps prudent to dedicate a few comments to other sources of information that could have been included but were not. First of all, it could be argued that it might have been enlightening to look at some of those ‘real outcomes’ that have resulted from EUrelated referendums rather than merely polls. The actual count of the whole parameter (N rather than n) would, all things being equal, bring about more precise results. Thus, one could look at the Swedish Euro referendum, the Irish Nice-vote, etc. The problem with this approach is that all things are not equal. Although outcomes of referendums overcome the potential bias and error problems inherent in sample-based surveys, there are other problems relating to this type of data. First of all, there are indications that in the past, electorates have used a referendum for other purposes than the immediate one, which is to answer a, at least in semantic terms, relatively straightforward question. Occasionally, referendums are hijacked by the electorates to signal general dissatisfaction with their political leaders by voting against the side recommended by the government (Franklin, van der Eijk, and Marsh 1995; Taylor 1996; Marsh 1998). There is evidence to suggest that this was a strong factor in the Danish Euro-vote in September 2000. The government was coming to the end of its term and was not popular. A few days before the referendum, a German 148

The Framework for Empirical Testing

newspaper asked the then-Danish prime minister, Poul Nyrup Rasmussen, whether he intended to resign, should the result be negative. (The prime minister had campaigned forcefully for a ‘yes’ to the point where it could perhaps be argued that a ‘no’ would seriously affect his credibility, mainly because he had not hesitated to issue threats of what were to happen if Denmark did not join the Euro (i.e. doom and gloom).) His answer to the German journalist was that he had ‘threatened to stay’ because otherwise he might have been encouraging the voters to say no ‘just to get rid of me’ (Der Spiegel, 18 September 2000). A similar argument has been made about the French referendum on the Maastricht Treaty when the unpopularity of Mitterrand’s socialist government together with increasing distrust in the president led a surprising (at least to the government) number of citizens to vote against the treaty (Taylor 1996: 145). Thus, a study from 1995 which examined circumstances and quantitative data surrounding the Danish and the French Maastricht referendums concludes that: ‘In case of the Maastricht referenda in France and Denmark, the apparent unpopularity of the European Project in fact appears to be nothing of the kind, but instead to have reflected the unpopularity of ruling parties in both countries’ (Franklin, van der Eijk, and Marsh 1995: 101). Having said this, the above qualification is only valid in cases where the population is strongly divided on an issue and/or not particularly bothered. If the Danes had been enthusiastic about joining the Euro, they would presumably have voted for it regardless of how they felt about their government. It should be noted that this is not the case with respect to all referendums and that evidence points to the fact that the recent French and Dutch referendums on the constitution were very much an expression of concerns regarding European integration, albeit different concerns in the two countries. Another problem with using referendums for measuring EU public opinion lies in the fact that referendums do not take place in all countries and thus do not provide an inclusive data set. And finally, the questions asked in referendums—despite being obvious indicators of the level of support for European integration—do not focus on the demos issue and thus are not particularly useful in this case. The issue of the demos is effectively one that affects the legitimacy of EU government and not the integration process; hence, survey data where relevant questions are asked is likely to be more useful in illuminating this issue, whereas referendum results are instead an excellent indicator of support for integration at that given point in time in the specific context in one country. 149

The Framework for Empirical Testing

This leads me to the next type of excluded material, namely the large volumes of survey data that have not been included in this study. As mentioned previously, the Eurobarometer surveys have never been produced with the particular aim of measuring the existence of a demos but rather support for integration and particular policy initiatives. And although some of the questions are suitable proxy variables for the indicators, the majority of the Eurobarometer questions are not relevant for such analysis. Similarly, EVS variables have been selected which are directly relevant and appropriate as proxies. Explanations as to the inclusion of particular questions will be provided below with additional comments in the analysis itself.

5.4.2. Indicators and Variables A number of variables have been selected in an attempt to measure the indicators. It must be stressed that although it is not possible to quantify accurately the relative importance of each of the variables, some are clearly more important than others. This is mainly the case because the variables are not all included for the same reasons. For instance, under indicator three ‘Desire to shape a common future’, the variable that measures support for integration (good or bad thing), clearly has a different purpose than ‘age group’. Where the former is a proxy for the indicator, the latter is included in order to analyse whether there is a relationship between age and views on European integration. Another example of variables included for different purposes can be found under indicator four ‘awareness of the transnational process’. Some variables serve as direct proxies such as knowledge about the EU (the quiz questions), another ‘membership good/bad thing’ (which was a proxy under indicator three) is included in order to test whether there is a relationship between knowledge and support. Furthermore, some of the variables are ‘real proxies’ (i.e. questions that are very semantically close to the indicator), whereas others act as ‘related variables’ and are included with the aim of potentially uncovering reasons for the scores of the proxies. For instance, under ‘democratic self-consciousness of the collective citizen body’ one of the problems with the identification and attachment variables is that these are all relative, that is, respondents were asked to rank their relative attachment to a number of polities. Hence, other variables have been included that might add to the information, such as what it means to be a citizen of Europe. Thus, it should be stressed that this approach means that some variables have a more ‘natural’ connection to the indicator that 150

The Framework for Empirical Testing

is being measured than others. In the following section each indicator will be summarized. INDICATOR ONE: THE DEMOCRATIC SELF-CONSCIOUSNESS OF THE COLLECTIVE CITIZEN BODY This indicator can measure the extent to which people see themselves as part of a whole and also as part of a democratic whole. Hence, it draws on both Weiler’s demos definition and that offered by Chryssochoou. The indicator seeks to uncover the extent to which people feel they have a European (or rather an EU) identity, whether they believe a European cultural identity exists, and also whether they feel solidarity towards each other. INDICATOR TWO: ADHERENCE TO SHARED VALUES Under this indicator I look at the extent to which a range of different values of the different electorates are similar both as regards democratic practices but also as regards basic, lasting preferences. The proxies for indicator two are from the EVS, which is more appropriate than Eurobarometer in this case because it focuses on lasting preferences as opposed to attitudes. There are a number of reasons why shared values are important for a political system, but there are also limitations as to how ‘shared’ such values have to be for the polity to survive. As regards the nature of such values, the definition applied by Aberle in a study of the United States is appropriate: By a value is meant an effectively charged idea or attitude in terms of which objects, events, actions, individuals, etc. are judged on a scale of approval– disapproval, whether the approval and disapproval are moral, aesthetic, hedonic, or in terms of some other dimension. By a shared value is meant—for present purposes—one held in common by a plurality of individuals. By a system of values is meant a set of such ideas or attitudes which have a logical, meaningful, or affective constituency. (Aberle 1950: 495)

This definition, however, does not make a strict demarcation line between values and attitudes (of course, this may not be possible). However, in another study, Miller, White and Heywood defined values as ‘conceptions of the desirable, used in moral discourse, with a political relevance for behaviour’ (van Deth and Scarbrough quoted in Miller, White, and Heywood 1998: 1). In addition, values can be seen as ‘restricted to questions of broad political principle, avoiding questions of day to day politics’ (Miller, White, and Heywood 1998: 2) which form part of the rationale on which 151

The Framework for Empirical Testing

the formation of attitudes are based (see discussion in Katz 1991: 295). A further distinction is provided by Hechter who differentiates between instrumental values that provide means to other ends and immanent values which are desired purely for their own sake (Hechter 1993: 4). The reason that it is interesting to uncover the degree to which individuals across Europe share values is based on the hypothesis that the more different the values, the harder it will be to obtain the solidarity and identity the integration project necessitates. Hechter argues that: To the degree that values are shared by members of a collectivity, they take on the character of an objective reality that may be reflected in prices or in existing institutions and therefore constrain individual action in predictable ways. These kinds of shared values result from the internalization of norms that have to be countenanced in decision-making: they serve to limit expectations, and also may carry sanctions in their wake. At least two things distinguish instrumental values from other kinds of socially produced shared values. First, unlike other shared values (which vary across societies), instrumental ones are universal. Second, they are self-serving, which suggests that they may well have evolutionary origins (Hechter 1993: 4, n. 11).

Hechter proceeds to argue that whereas instrumental values are held commonly by the members of a society, immanent values are unlikely to be because ‘they afford us an opportunity to individuate ourselves in social life’ (Hechter 1993: 4). Whether there really is such a clearcut distinction between values is another issue, and this book does not make claim to providing parameters for such a distinction. Nevertheless, it is worth keeping it in mind. If the values that Europeans share appear to be of a particularly instrumental character (which they should be, according to Hechter, because he claims that instrumental values are universal), this could indicate that the basis for integration is rather more utilitarian (obviously this would only be the case if the immanent values diverged between countries). However, should we find a great heterogeneity between the immanent values, this would give a higher score for indicator two in the subjective-emotional sense.3 The values and their meaning are discussed in further detail under this indicator in Chapter 6. INDICATOR THREE: A DESIRE TO SHAPE A COMMON FUTURE Under this indicator the primary variables refer to attitudes rather than values. This distinction is based on the rationale in Miller, White, and Heywood who differentiate between enduring political values versus transient political attitudes and that hence a change in values is bound to be 152

The Framework for Empirical Testing

more significant than a change in preferences (Miller, White, and Heywood 1998: 1). For indicator three, the proxy variables include support for EU membership as well as views on the veto. INDICATOR FOUR: PUBLIC AWARENESS OF THE TRANSNATIONAL PROCESS The inclusion of this indicator is necessary because part of the definition of the demos is that there has to be a reasonable level of public awareness of the polity-wide decision-making process, in this case, the ‘transnational’ process. However, it is difficult to conclude much on the basis of the measurement alone because we have no established ‘level’, that is, the indicator is not quantified because we do not know exactly how much awareness is necessary. On the other hand, recent findings in research on changing attitudes to European integration may be useful in this context. Experiments have shown that for those with a relatively low level of knowledge, small amounts of negative or positive information can serve to swiftly change attitudes (Saris 1997). Hence, perhaps it would be possible to locate the level at which people’s views are not so easily changed and consider this level the ‘appropriate level’. But such matters are beyond the scope of this research. In the current case, the knowledge indicator is included as a means of testing whether people are generally aware of the process. Summary It is important to stress that none of the four indicators should be analysed in isolation. People’s ‘desire to shape a common future of the EU’ does not document the existence of a demos. Such a desire could be the result of a pragmatic cost–benefit analysis aiming at maximizing perceived national or personal utility. Likewise, ‘public awareness of the transnational process’ could be present without a desire for integration and is not, as an isolated indicator, capable of answering the demos question. Neither is ‘adherence to shared values’, as such a ‘sameness’ does not necessarily correlate with a desire for a common future, nor transnational solidarity. Finally, democratic self-consciousness of the collective citizen body is a term that despite its relevance is too diffuse to constitute the basis for a complete analysis. The Figure 5.1 below shows the steps that have been taken in the operationalization process. In Chapter 2 the demos was identified as an object of study (level 1). In Chapter 3 the demos was divided into a legal and a social demos 153

The Framework for Empirical Testing Demos

Level 1

Legal

Level 2

Social Subjective emotional

Ethnocultural Indicator 1 Proxies

Indicator 2 Proxies

Indicator 3 Proxies

Level 3

Indicator 4 Proxies

Level 4 Level 5

Figure 5.1. Operationalizing the demos

(level 2) and subsequently the social dimension was divided into ethnocultural and a subjective emotional dimension (level 3). On the basis of the preceding analysis and the categories developed by Chryssochoou, the four indicators were derived (level 4) and variables selected (level 5). It is between levels 3 and 4 that one moves from the theoretical level to the analytical level and between levels 4 and 5 that one moves into the empirical world. In this phase, the second step of the operationalization, the variables are selected that—together—can be used to measure the values of the indicators. These are also known as the proxy variables.

5.5. Descriptive Statistics and the Presentation of Results All graphs and tables presented in Chapter 6 are produced independently of the descriptive statistics in the Eurobarometer reports. The findings concentrate on post 1992, that is, after the Maastricht ratification crisis, although in a number of cases it is relevant to compare results with older material. For the remainder of the analysis the most recent data available has been used.

5.6. Conclusion This chapter has established the framework for the empirical investigations that are to follow in Chapter 6 by outlining the method and explaining how it is intended to ensure high degrees of validity and reliability. Potential pitfalls have been discussed and explanations have been provided as to how I intend to minimize the impact of such pitfalls. Clearly, some would argue that measuring feelings of identity and attachment and other such variables is not possible. However, despite 154

The Framework for Empirical Testing

recognizing that it is difficult to measure such things and that it probably is impossible to measure them 100 per cent accurately, this is not to say that it would not be possible to get an indication of the values of these variables and subsequently the state of the demos. One major reservation should be made at this stage: there is a large amount of data included in the analysis. When presenting the data I concentrate my analysis on issues that are directly or indirectly linked to the demos question. This means that other potentially interesting findings may be ignored if they are not directly related to the indicators, the aim being to avoid straying from the focus. It was concluded that the most important element in the method used in a research project is that one actually makes it clear what that method was. This means giving a detailed account of each step of the process in order that readers can subject the results to scrutiny by way of ‘process tracing’. With this in mind such explanations have now been offered and in the following chapter the actual findings will be described and analysed.

Notes 1. A similar concept is Easton’s political community which can be measured as the extent to which the ‘members of the group that we are examining are sufficiently oriented toward each other to want to contribute their collective energies toward pacific settlement of their varying demands’ (Easton 1957: 391). 2. Some are very sceptical towards the surveys. See Moravcsik and Sangiovanni, 2003: 141. 3. A similar distinction is provided by Hechter who argues that values and norms are durable and general, whereas preferences are labile and particular (Hechter 1993: 3).

155

This page intentionally left blank

6 The demos—empirical analysis

6.1. Introduction In this chapter evidence is provided which indicates that no social EU demos exists at the present time. On the basis of the analysis, the tentative claim will be made that an EU demos is not about to materialize in the near future although there is evidence to suggest a relatively high level of common values across states, particularly when it comes to democratic values.

6.2. Findings 6.2.1. Indicator One—Democratic Self-Consciousness of the Collective Citizen Body It is, of course, doubtful whether the EU has a ‘collective citizen body’. Indeed, this is the central question of the current chapter. Hence, the way to understand this indicator is to ask, do the EU citizens perceive themselves as part of a collective, transnational whole at the EU level, at least in a technical sense? And furthermore, to what extent do they identify with and feel solidarity towards each other? The extent to which EU inhabitants consider themselves part of a collective citizen body can be measured directly by simply asking respondents how they primarily identify themselves (Duchesne and Frognier 1995) or indirectly by measuring levels of pride in the polity (Duchesne and Frognier 1995) and transnational trust (Lindberg and Scheingold 1970; Niedermayer 1995). To a lesser extent, views on taxation can be used as proxies for levels of solidarity, assuming that a willingness to pay tax reveals an underlying sense of belonging to the polity, a shared 157

The Demos—Empirical Analysis

transnational solidarity or perhaps a belief that there is a transnational ‘common good’. To begin with the more direct approach, we can look at one of the consistently posed questions of the Eurobarometer surveys, that is, how the respondents identify themselves. IDENTIFICATION AND ATTACHMENT TO THE EU When asked whether they would tend to define themselves as Europeans, strictly in the terms of their nationality, or a combination, respondents replied as follows (Figure 6.1). Throughout the period most respondents classify themselves as either ‘nationality only’ or ‘nationality and European’. However, there does not appear to be any sign that people are feeling more European in 2005 than was the case in 1994. In fact, the number of people who define 100

80

Nationality only Nationality and European European and nationality European only Don’t know

60

40

20

0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

Figure 6.1. Nationality vs. European in future (1994–2005) Note: Respondents were asked: ‘In the near future do you see yourself as, “(nationality) only”, “(nationality) and European”, “European and (nationality)”, “European only”, “don’t know”.’ Source: Data compiled from EB42 (EU12), EB43, EB46, EB47, EB50, EB52, EB54, EB56, EB58, EB60 (EU15), EB62, and EB64 (EU25).

158

The Demos—Empirical Analysis

themselves only in terms of their nationality has risen. However, because the numbers change throughout the period, this increase is unlikely to be significant. The aggregate figures show that in 1994 (EB42) an average of 33 per cent defined themselves in terms of nationality only, whereas in 2005 this number had increased to 41 per cent. Because the trends change regularly, it makes sense to add a trend line for nationality only in order to uncover the developments between 1994 and 2005 (Figure 6.2).1 Despite the actual increase in respondents who see themselves in terms of their nationality only, the trend line for this category is nearly horizontal. This means that no discerning upward trend can be confirmed throughout the period. Even though the number of people who chose nationality only is higher in 2005 than it was in 1994, the fact that the trend line is almost horizontal indicates that this number is as likely to fall in the future as it is to increase. 100

80

Nationality only Nationality and European European and nationality European only Don’t know Linear (Nationality only)

60

40

20

0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

2005

Figure 6.2. Nationality vs. European in future with trend line for ‘nationality only’ (1994–2005) Source: Data compiled from EB42 (EU12), EB43, EB46, EB47, EB50, EB52, EB54, EB56, EB58, EB60 (EU15), EB62, and EB64 (EU25).

159

The Demos—Empirical Analysis 100 Nationality only Nationality and European European and nationality European only Don’t know Linear (Nationality and European)

80

60

40

20

0 1994

1995 1996

1997

1998 1999

2000

2001

2002

2003 2004

2005

Figure 6.3. Nationality vs. European in future with trend line for ‘nationality and European’ (1994–2005)

For the sake of good order, the line chart is shown with individual trend lines for each response category (Figures 6.3–6.6). Whether one interprets these figures as a low score on the indicator would depend on one’s definition of a low score. It is felt that establishing such a score in terms of percentages would be arbitrary and that attention should be paid to developments and change instead. What we can see is that identification with the national polity is still significantly stronger than the identification with the EU and that there is nothing to indicate that individuals are inclined to feeling more European in 2005 than they were in 1994. Having said that, the fact that the EU has enlarged three times during the period shown, means that the averages compared over the years do not contain figures from the same countries. Hence, it is possible that the aggregate scores have remained almost unchanged because older member states become more ‘European’ and that the average is decreased by the new member states. If, for instance, it were the case that people felt 160

The Demos—Empirical Analysis 100

80

Nationality only Nationality and European European and nationality European only Don’t know Linear (European and nationality)

60

40

20

0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

2004 2005

Figure 6.4. Nationality vs. European in future with trend line for ‘European and nationality’ (1994–2005) Source: Data compiled from EB42 (EU12), EB43, EB46, EB47, EB50, EB52, EB54, EB56, EB58, EB60 (EU15), EB62, and EB64 (EU25).

increasingly European the longer their country had been a member of the Union, the fact that the Union continues to enlarge could mean that the average was ‘hiding’ a large discrepancy between newer and older members. Recent figures from Eurobarometer show that there is no significant difference between the EU25 and the EU15 (EB63). The category ‘nationality only’ scored 41 per cent in both groups, ‘nationality and European’ was 48 per cent in the EU25 and 47 per cent in the EU15. Having said that, between 2004 and 2005, some developments can be spotted within individual countries. Most notably, in the Czech Republic ‘nationality only’ fell from 57 per cent in 2004 (EB62) to 37 per cent a year later. The score for ‘nationality and European’ increased to 48 per cent in 2005. Other states where ‘nationality only’ scored higher by 2 or more percentage points in 2005 than in 2004 were Belgium, Estonia, and Ireland (all by 161

The Demos—Empirical Analysis 100 Nationality only Nationality and European European and nationality European only Don’t know Linear (European only)

80

60

40

20

0 1994 1995

1996 1997 1998 1999 2000

2001 2002 2003 2004

2005

Figure 6.5. Nationality vs. European in future with trend line for ‘European only’ (1994–2005) Source: Data compiled from EB42 (EU12), EB43, EB46, EB47, EB50, EB52, EB54, EB56, EB58, EB60 (EU15), EB62, and EB64 (EU25).

5 percentage points), France and Cyprus (by 3 percentage points), Italy and Austria (both by 2 percentage points), and the United Kingdom, where the ‘nationality only’ category was selected by 55 per cent of respondents in 2004 and 63 per cent of respondents in 2005. Countries where ‘nationality only’ scored lower by 2 percentage points or more in 2005 than in 2004 were Greece (−10), Spain (−3), Latvia (−2), Luxembourg (−7), Hungary (−13), Malta (−6), Poland (−6), Slovenia (−8), Finland (−7) and Sweden (−4). In actual terms there does not appear to be a consistent relationship between length of membership and feelings of identity. In some newer member states there is a clear increase in the European feeling but the trend is the reverse in other states. At the time of the data collection, the ten most recent members had only been in the EU for one year so perhaps some ‘turbulence’ in the figures can be 162

The Demos—Empirical Analysis 100

80

Nationality only Nationality and European European and nationality European only Don’t know Linear (Don’t know)

60

40

20

0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

Figure 6.6. Nationality vs. European in future with trend line for ‘don’t know’ (1994–2005) Source: Data compiled from EB42 (EU12), EB43, EB46, EB47, EB50, EB52, EB54, EB56, EB58, EB60 (EU15), EB62, and EB64 (EU25).

expected. (These numbers should be read with caution because the scores change so regularly.) Nevertheless, for the older member states we can test the following hypothesis: People feel increasingly European the longer their country has been a member of the European Union. Table 6.1 below shows the results in percentages. The figures are derived by running individual cross tabulations for each country and calculating the average value for all countries that joined in that particular year. The numbers for the 2004 members have been included for comparative purposes, that is, to see what respondents said about their feelings of identity before they joined the Union (CCEB2003.4): 163

The Demos—Empirical Analysis Table 6.1. Year joined vs. European citizenship future feeling Year joined

1958 1973 1981 1986 1995 2004

Nationality only

Nationality & European

European & nationality

European only

Don’t know

Total∗

30.9 46.5 53.4 38.6 48.7 36.0

51.7 45.0 41.1 50.9 43.4 47.0

9.8 4.5 3.5 4.2 5.1 8.0

5.3 1.9 1.6 2.6 1.7 3.0

2.3 2.2 0.4 3.8 1.2 6.0

100 100 100 100 100 100

∗

Rounded Note: The original member states (1958) were Belgium, France, Germany, Italy, Luxembourg, and the Netherlands. The UK, Ireland, and Denmark joined in 1973, Greece in 1981, Spain and Portugal in 1986, and Austria, Finland, and Sweden joined in 1995. The ten countries who joined in 2004 are Malta, Cyprus, Lithuania, Latvia, Estonia, Poland, the Czech Republic, Slovenia, Slovakia, and Hungary. Source: Data compiled from EB58. For countries that joined in 2004 accumulated figures are from CCEB2003.4.

The results are similar to findings by previous studies (see for instance Duchesne and Frognier 1995: 201) that there is no significant relationship between length of membership and identification with the Union. Although there is no positive linear relationship between joining year and feeling European at first glance, there appears to be a pattern in the sense that the founding members feel more European than all other members. But when testing we find no significant relationship, in other words, there is no evidence to suggest an association between the length of a country’s membership of the EU and the extent to which its citizens feel European. The fact that 6 months before joining the Union, 8 per cent of the accession countries’ citizens felt almost as European as the founding members also contributes to falsifying the hypothesis. According to this analysis, length of membership does not determine how European citizens feel. The next question that arises is whether feeling European also means feeling attached to the European Union. This is because ‘Europe’ and ‘the European Union’ are not equal entities so it is more than conceivable that people could feel very European without feeling any particular attachment to the EU. The latter is also measured regularly by Eurobarometer. In 2002, Eurobarometer 58 (released in March 2003) asked respondents how attached they felt to a range of polities. The results are shown below (Figure 6.7). For the EU15 as a whole, 45 per cent claim to be either very (10%) or fairly (35%) attached to the EU, whereas a total of 52 per cent are either

164

The Demos—Empirical Analysis EU15 UK Sweden Finland Portugal Austria Very attached

Country

The Netherlands

Fairly attached

Luxembourg

Not very attached

Italy

Not at all attached Don’t know

Ireland France Spain Greece Germany Denmark Belgium 0

20

40

60

80

100

Per cent

Figure 6.7. Attachment to the EU Note: Figures from 2002. Source: Data compiled from EB58.

not very attached (36%) or not at all attached (16%) to the Union. As regards the question of the potential pattern in the length of membership of the EU and attachment to the polity, there does not appear to be an association. (For instance, Figure 6.7 indicates that the Netherlands and Belgium show a lower level of attachment to the EU than more recent members such as Ireland and Austria.) In comparison, attachment to own country—measured in the same survey—is shown below (Figure 6.8). For the EU15, 89 per cent are either very or fairly attached to their own country, whereas 10 per cent are not very or not at all attached to their country. Thus, people are approximately twice as likely to be very attached to their country as to the EU. Even in states that are normally considered to have a low level of overall national integration due to, for instance, strong regional loyalties, the attachment to country is higher than attachment to the Union. (Belgium and Spain have a high or fair attachment of 86 and 82 per cent, respectively, although it should be

165

The Demos—Empirical Analysis EU15 UK Sweden Finland Portugal Austria

Country

The Netherlands

Very attached Fairly attached

Luxembourg

Not very attached Italy

Not at all attached Don’t know

Ireland France Spain Greece Germany Denmark Belgium 0

20

40

60

80

100

Per cent

Figure 6.8. Attachment to own country Source: Data compiled from EB58.

kept in mind that the distribution is skewed towards the ‘fairly attached’ category rather than the ‘very attached’ response.) How much the regional differences matter in this case can be tested by looking at the next variable (Figure 6.9). The highest levels are seen in Greece (73% very attached) and Ireland (63% very attached). These strong levels of attachment both to regions and nation states are another sign that an EU-wide demos is not realistic in the near future. Although it is perfectly possible for people to feel attachment to a range of polities, it is harder for the European Union to build up a ‘Scharpfian’ thick identity as long as people’s emotions remain embedded in their more familiar environments. What the graphs do not show, however, is whether a person’s level of attachment to their state affects the level of attachment they feel towards the European Union. The pie graphs in Figure 6.10 highlight this relationship. Each pie shows the values of level of attachment felt

166

The Demos—Empirical Analysis EU15 UK Sweden Finland Portugal Austria

Country

The Netherlands

Very attached Fairly attached

Luxembourg

Not very attached Italy

Not at all attached Don’t know

Ireland France Spain Greece Germany Denmark Belgium 0

20

40

60

80

100

Per cent

Figure 6.9. Attachment to region Source: Data compiled from EB58.

towards the EU grouped by whether people felt very attached, fairly, not very or not at all attached, or did not know, respectively to their own country. There appears to be a clear pattern, but perhaps a slightly surprising one, in that the more a respondent is inclined to declare him or herself very attached to his or her country, the more that same person is inclined to declare themselves very attached or fairly attached to the EU. This relationship is true for each different category. Testing for significance confirmed the clear positive association between respondents’ level of declared attachment to their country and their level of attachment to the EU.2 What this shows is that the variable may be telling us more about a person’s willingness to declare themselves attached to a polity than their relative levels of attachment to different polities.

167

The Demos—Empirical Analysis Very attached to country

Very attached to the EU Fairly attached to the EU Not very attached to the EU Not attached at all to the EU Don’t know (EU)

Not very attached to country

Very attached to the EU Fairly attached to the EU Not very attached to the EU Not attached at all to the EU Don’t know (EU)

Fairly attached to country

Very attached to the EU Fairly attached to the EU Not very attached to the EU Not attached at all to the EU Don’t know (EU)

Not at all attached to country

Very attached to the EU Fairly attached to the EU Not very attached to the EU Not attached at all to the EU Don’t know (EU)

Don’t know (country)

Very attached to the EU Fairly attached to the EU Not very attached to the EU Not attached at all to the EU Don’t know (EU)

Figure 6.10. Relationship between level of attachment to own country and level of attachment to the European Union (EU15) Source: Data compiled from EB58.

168

The Demos—Empirical Analysis

ATTACHMENT TO EUROPE OR ATTACHMENT TO THE EUROPEAN UNION? It is not possible to include the 2004 accession countries (EU2004) in this particular comparison because the survey on the then-candidate countries does not ask this question. However, the survey includes a very similar question, that is, it asks people to quantify their level of attachment to Europe. An overall average of 35 per cent of the EU2004 respondents say they are very attached to Europe. In a Standard Eurobarometer survey of the EU15 carried out in the same months, the ‘attachment to the EU’ (as measured in previous surveys) was replaced by ‘attachment to Europe’ making it possible to produce a direct comparison of attachment levels between new and old members. The results are shown below (Table 6.2). Two interesting tendencies appear from this table. First of all there is the much stronger attachment to Europe seen in the 2004 joining countries (i.e. very attached responses). These results were found to be significant, meaning that new countries (who at the time the survey was carried out had yet to join the Union) were more inclined to declare themselves attached to Europe than the citizens of the EU15. (For the test, the categories were grouped so that the EU15 and the EU2004 were both compared to three categories of levels of attachment: very/fairly attached, not very/not attached, and don’t know).3 All things being equal, it is clear that respondents from countries who, when surveyed, had yet to join the Union express a stronger attachment to Europe. The numbers will obviously not tell us anything about the causality. However, it is tempting to conclude that the 2004 states felt a stronger attachment to Europe due to the fact that they did not take membership of the ‘club’ for granted.

Table 6.2. Attachment to Europe vs. attachment to the EU

Very attached Fairly attached Not very attached Not at all attached Don’t know Total

EU15—(to the EU) Fieldwork Autumn 2002

EU15 to Europe Fieldwork Autumn 2003

2004 countries to Europe Fieldwork Autumn 2003

10 35 36 16 3 100

14 44 29 10 3 100

35 39 23 (‘not very’ or ‘not at all’ accumulated) 3 100

Source: Data compiled from EB58, EB60, and CCEB2003.4.

169

The Demos—Empirical Analysis

For the EU15, there is an increase in the number of very attached and fairly attached respondents between the attachment to the ‘European Union’ question and the attachment to Europe question. However, this is only a small increase and the fact that there is a year between the two results makes it impossible to test whether the difference is due to the word ‘EU’ being replaced with Europe or simply coincidental. But it does mean that we cannot be sure that attachment to Europe necessarily equates to attachment to the EU, hence another good argument for not using these terms interchangeably. As regards subsequent developments (most recent figures available at the time of writing are from 2005) the following picture emerges of attachment to Europe for the EU25 (where earlier surveys talked about the EU in this context, recent surveys measure attachment to Europe) (Figure 6.11):

Country

Figure 6.11. Attachment to Europe EU25 (2005) Source: Data compiled from EB63.

170

UK EU2 5 EU1 5

Italy prus of Cy

Latvia Lithu ania Luxe mbou rg Hung ary Malta The Neth erlan ds Austr ia Pola nd Portu gal Slov enia Slov akia Finla nd Swe den

Don’t know Not at all attached Not very attached Fairly attached Very attached

Rep.

0

Spain Fran ce Irela nd

50

Belg ium Czec h Re p. Denm ark Germ any Esto nia G ree ce

Per cent

100

The Demos—Empirical Analysis

Twenty-two per cent of respondents are willing to declare themselves very attached to Europe. This is a significant increase compared to 2004 (EB58) where only 10 per cent of respondents selected this category. Even when allowing for the potential underlying effect of new member states (who as shown above are occasionally more likely to express attachment to Europe or the EU) this is a significant increase. However, bearing in mind the findings made above that people appear more willing to express attachment to Europe than to the European Union, comparing this table to a table with results for the EU would be likely to produce misleading interpretations. Having said that, the fact that 66 per cent of respondents declared attachment to Europe is very positive from an integrationist perspective. In order to compare like with like we can look at attachment to country (Figure 6.12). As in 2002 (Figure 6.8) most people express attachment to their country. Fifty-three per cent of the EU25 claim to be very attached and 38 per cent fairly attached to their country. In other words, 91 per cent of respondents

0

Don’t know Not at all attached Not very attached Fairly attached Very attached

UK EU2 5 EU1 5

50

Belg ium Czec h Re p. Denm ark Germ any Esto nia Gree ce Spain Fran ce Irela nd Italy Rep. of Cy prus Latvia Lithu ania Luxe mbo urg Hung ary Malta The Neth erlan ds Austr ia Pola nd Portu gal Slov enia Slov akia Finla nd Swe den

Per cent

100

Country

Figure 6.12. Attachment to country EU25 (2005) Source: Data compiled from EB63.

171

The Demos—Empirical Analysis

express attachment to their country. For the EU15 there are no significant changes between 2002 (reported in EB58) and 2006. Hence, we can again conclude that people are still much more attached to their country than to Europe. There are, however, significant differences between member states, with Danes, Maltese, and Greeks much more likely to declare themselves very attached than other nationalities. The latter are also significantly more proud of their country than the average EU25, with 80 per cent of Greek respondents declaring themselves ‘very proud’ to be Greek (see Figure 6.13). A similar level of national pride (very proud) can be seen in the Republic of Cyprus (79%) and Malta (75%). The respondents least willing to declare themselves very proud of their nationality are the Germans (19%). This result is significantly lower than the average and 12 percentage points lower than the second least proud nation, the Dutch (31%). The Eurobarometer Report explains this in terms of Germany’s

0

Don’t know Not at all proud Not very proud Fairly proud Very proud

Country

Figure 6.13. Pride in nationality EU25 (2005) Source: Data compiled from EB64.

172

UK EU2 5 EU1 5

50

Belg ium Czec h Re p. Denm ark Germ any Esto nia G ree ce Spain Fran ce Irela nd Italy Rep. of Cy prus Latvia Lithu ania Luxe mbou rg Hung ary Malta The Neth erlan ds Austr ia Pola nd Portu gal Slov enia Slov akia Finla nd Swe den

Per cent

100

The Demos—Empirical Analysis

past, stating that in Germany the ‘notion of national pride remains sensitive’ (Eurobarometer Report 64: 26). The report cites measurements of national pride from 2000 to 2005 and states that although the increases in pride from year to year are insignificant the overall increase during the 5 year period is almost 5 percentage points (Eurobarometer Report 64: 26). One could perhaps speculate whether this increase was due to the inclusion of ten new member states; however, the numbers for the EU25 and the EU15 do not differ significantly in any of the five response categories. Explaining a steady increase in national pride without any other data or follow-up questions would easily become a speculative enterprise, and there is some evidence that national pride does not hinder European pride. In fact, the contrary could be the case (Figure 6.14):

Don’t know I do not feel European Not at all proud Not very proud Fairly proud Very proud

UK EU2 5 EU1 5

of Cy

Rep.

0

Italy prus Latvia Lithu ania Luxe mbo urg Hung ary Malta The Neth erlan ds Aust ria Pola nd Portu gal Slov enia Slov akia Finla nd Swe den

50

Belg ium Czec h Re p Denm . ark Germ any Esto nia Gree ce Spain Fran ce Irela nd

Per cent

100

Country

Figure 6.14. Pride in being European EU25 (2005) Source: Data compiled from EB64.

173

The Demos—Empirical Analysis

As stated by the Eurobarometer researchers, ‘In the context of European public opinion, the relevant question to research is to what extent national pride has an effect on how people think about the European Union. Though this requires more extensive analysis, a brief examination reveals that people who are proud of their country also tend to be proud of the European Union and vice versa’ (Eurobarometer Report 64: 27). This is a similar pattern to the one seen with respect to declared levels of attachment. The problem with this citation from Eurobarometer 64 is that the question was not about pride in the European Union. The question was ‘Would you say you are very proud, fairly proud, not very proud, or not at all proud to be European’. Again, the mistake has been made of equating Europe to the European Union, which is incorrect at a number of levels, both with reference to geography but also to psychology. It is perfectly possible that some respondents may consider themselves very European and even proud to be European without feeling similarly towards the European Union (see Table 6.2 for a similar example).

A EUROPEAN CULTURAL IDENTITY? Aside from people’s personal feelings of identification and belonging, one can also approach this question from a more general perspective. In 1999, the Eurobarometer survey asked respondents to evaluate the following statement: ‘There is a European cultural identity shared by all Europeans’ (EB52). The respondents were made to select one of five responses, ‘completely agree’, ‘slightly agree’, ‘slightly disagree’, ‘disagree completely’, and ‘don’t know’. This is a different category of variable to those above, because rather than measuring identification and attachment to the European Union (variables often associated with the future), the ‘cultural identity’ is a historical variable that focuses on peoples’ perception of the past, that is, that there may be foundations to ‘build on’. In addition, it talks about the collective rather than the individual’s own feelings. In other words, respondents are asked their views on a general question rather than a personal one. The results are shown in Figure 6.15. Very few (7 per cent of the respondents) agree completely that there is a ‘distinct European cultural identity shared by all Europeans’. Nearly one-third slightly agrees, 27 per cent slightly disagree and 22 per cent disagree completely. Hence, only one-third agree to some extent whereas nearly half of the respondents disagree to some extent. There is some 174

The Demos—Empirical Analysis 100 90 80 70 Don’t know Disagree completely Slightly disagree Slightly agree Completely agree

Per cent

60 50 40 30 20 10

15 EU

rk rma ny Gre ece Spa in Fra nce Irel and Ital y Lux em The bou rg Net her lan ds Aus tria Por tug al Fin lan d Sw ede n UK Ge

ma

Den

Bel

giu

m

0

The Country

Figure 6.15. A shared European cultural identity? Source: Data compiled from EB52.

variation between countries, the highest score for completely agreeing being in Greece (16%) with the lowest found in Spain, the Netherlands, Portugal, and Finland (all 4%). TRANSNATIONAL SOLIDARITY In a number of Eurobarometer surveys, respondents have been asked directly whether they feel solidarity towards each other. Obviously, this is the most straightforward way of testing such a variable because the actual variable and the proxy are so semantically close. However, it is also the most problematic. First of all, it is problematic if the feeling of solidarity 175

The Demos—Empirical Analysis Table 6.3. Measurements of transnational EC solidarity 1973–8 (EU9), % Year

Should help

Should not help

Don’t know

Total

1973 1976 1977 1978

76 80 79 76

11 11 10 11

13 9 11 13

100 100 100 100

Source: EB10.

has a human base. If this is the case, solidarity could be just as strong with a person from Zimbabwe or Ontario and thus, not a proxy variable for quantifying a European demos. Secondly, it is an easy question to answer in the affirmative because a positive answer is ‘free’ in this case both theoretically and practically. The first question reads: ‘In the event of one of the member countries being in major economic difficulties, the others should/should not/don’t know help.’ (EB10, fieldwork 1978). The weighted average at community level is shown in Table 6.3. All things being equal, this gives a picture of a fair amount of transnational solidarity. However, in their report, the Eurobarometer auditors themselves conclude that there is ‘no doubt that the above question tends to induce a “conformist” response’ (Eurobarometer 10: 99). This led the authors to ask another question, which interestingly produced a different response. Respondents were asked how far they would be prepared to make some ‘personal sacrifice’ to help another member country. This produced the following average result for the EC (Table 6.4). Hence, if ‘helping’ is operationalized into ‘make a personal sacrifice’ there is clearly a big drop in the number prepared to assist another member state. In recent surveys, another type of question has been asked which could perhaps serve as an indirect indicator of transnational solidarity, that is, whether people feel that a transfer payment during a particular crisis Table 6.4. Prepared to make a personal sacrifice Yes—prepared to No—not prepared Don’t know Total Source: EB10.

176

41 43 16 100

The Demos—Empirical Analysis

would be acceptable. This eliminated the second error option because answering in the affirmative is, theoretically, not free. However, the first potential error remains because it is likely that respondents would feel just as strongly that any payments should also be made in case of a flood in China or a drought in Africa. Again, this depends whether solidarity has a human base. So, how is it possible to clarify whether people feel more solidarity towards their fellow nationals than to Europeans and other people in general? In order to look into this we can form and test another two hypotheses: — People are inclined to feel stronger solidarity with their own nationals than towards other Europeans and — People are inclined to feel stronger solidarity towards other Europeans than towards human beings in general. In order to test the hypotheses, the results per country for three statements were tested, that is, ‘To what extent do you feel concerned about the living conditions of ‘your fellow countrymen’, ‘Europeans’ and ‘human kind’ (data from EVS). The results are shown in Figure 6.16. If comparing concern for ‘human kind’ and ‘Europeans’, the former scores higher in almost all countries surveyed—a result which is highly significant.4 In twenty-five of the twenty-six countries shown in the graph (all the EU25 where Great Britain and Northern Ireland are shown individually), respondents are more concerned with the living conditions of ‘human kind’ than ‘Europeans’. Of course, ‘Europeans’ are human, so this category includes the Europeans. However, the fact that fewer respondents show specific concern for Europeans indicates the difficulty that Europe has in creating a transnational solidarity that does not end up including the whole world. Now, it is tempting to suggest that human beings are reluctant to show greater concern for a selected group than for humans in general. But when comparing the bar graphs for ‘fellow countrymen’ and ‘Europeans’, again the European category has the lowest score, this time in twenty-five of twenty-six countries—and again the result is significant. Hence, we cannot conclude that there is a reluctance to be more concerned about a selected group. As ever, one has to be careful about ‘over-concluding’ on the basis of one variable. But the figures do confirm that a specific pan-European solidarity is not manifest.5 177

The Demos—Empirical Analysis 5 4.5 4 3.5 3 2.5 2 1.5

d Ire l

lan

ern rth No

Th Fellow countrymen

an

d

en

Fin

ark

ed Sw

m

nm De

lgiu

an erl eth

eN

Be

ds

l ga rtu

ain

Po

Sp

Ita ly

ia Au

str

y Ge

rm

an

GB

Fra

nc

e

1

Europeans

Human kind

5 4.5 4 3.5 3 2.5 2 1.5 nia ve

Th

eC

ze

Lu

Slo

rg mb ou

lta

xe

Ma

ce ee Gr

tia oa Cr

ary Hu

ng

kia va

pu Re ch

Slo

blic

d lan Po

ia an hu Lit

tvia La

ia ton Es

Ire

lan

d

1

Fellow countrymen

Europeans

Human kind

Figure 6.16. Concern with fellow countrymen, Europeans, and human kind—a comparison of means (1 = very concerned, 5 = not at all concerned) Source: Data from EVS 1999.

TAXATION As an indirect measure of solidarity, the tax question as raised by Eurobarometer in recent surveys is deemed an appropriate indirect measurement of solidarity because it shows the level of commitment to transnational redistribution. In 2002 and 2003, Eurobarometer asked respondents whether they would prefer to pay a tax directly to the European Union (Figure 6.17): 178

The Demos—Empirical Analysis 100

Per cent

80

60 Yes 2002 Yes 2003 40

20

15

UK

EU

rtug al Fin lan d Sw ede n

Po

ds stri a

rlan the

The

Ne

Au

rg

y

bou em

Lux

and

Ital

Irel

ain

nce

Sp

Fra

ny

ece

Gre

ark

rma

Ge

nm De

Be

lgiu

m

0

Country

Figure 6.17. A direct EU tax? Note: The Question was phrased as follows: ‘At present, each member state passes on to the European Union a part of its tax revenue. Would you prefer to pay this contribution directly to the European Union, or not?’. Respondents could reply either ‘yes’, ‘no’, or ‘don’t know’. Source: Data compiled from EB57 and EB59.

A large majority of respondents do not want to pay a direct EU tax.6 It could, of course, be added that given the choice, a lot of people might not want to pay tax to any polity at all; however, the wording of the question reduces this potential error. It appears that there is a small change from autumn 2002—but since there are only figures for two years, one should not read much into this. SUMMARY OF INDICATOR ONE Europeans seem no closer to the polity, nor to each other, in 2005 than they were in 1994. But perhaps most importantly, only a small number of respondents completely agree that there exists a distinct European identity (1999 data). The findings as regards transnational social integration show that social integration is not at the same level as legal–institutional 179

The Demos—Empirical Analysis

integration. In fact, there are two particular findings, which seem to falsify the assumption that legal–institutional integration will eventually lead to increased social integration. First, on average citizens of the 10 countries who joined the EU in 2004 are more likely to feel proud to be European than the EU15. Secondly, there is no linear relationship between the length of a country’s membership and the extent to which its citizens feel European or attached to the European Union, although citizens of founding members are likelier to be both more European and more attached to the polity. In other words, to the extent that ‘feeling European’ is a proxy for social integration, legal and institutional means may not be having the desired effect. Finally, there is no particular pan-European solidarity. Naturally, the findings are limited by the fact that respondents have only a set number of categories to choose from. If given the choice respondents might have mentioned other groups. For example, they may feel similar or higher levels of concern for Americans, Manchester United fans or model railway enthusiasts, depending on personal preferences. However, this does not change the fact that, of the options given, concern for Europeans scores lowest. This does not mean that Europeans are not concerned for each other but it does mean that they are not more concerned for each other than for other groups. Finally, however, it is recognized that it would have been desirable to include variables that could highlight the level of trust between the different national groups. These variables, however, are not available at this stage although earlier Eurobarometers included such questions. One earlier and very comprehensive study which looked at cross-national trust between 1976 and 1990 concluded that although an increase in trust could be spotted over time the general level of trust was rather modest (Niedermayer 1995: 237).

6.2.2. Indicator Two—Adherence to Shared Values A number of studies have analysed European values (particularly Western European values) both in the context of European integration but also more generally to evaluate value change and connections to sociological or political developments (see for instance Inglehart and Reif 1991; Abramson and Inglehart 1995; van Deth and Scarbrough 1998). Many such studies are substantial in that they include long time series as well as a large number of countries, enabling country-by-country comparisons as well as the study of developments over time. 180

The Demos—Empirical Analysis

The study of values in this chapter is comparatively modest and looks only at a small amount of values at one given moment in time. The subsequent paragraphs are concerned mainly with EU member states although for some questions data from potential accession states are included. The results discussed under Indicator Two have been compiled from data selected from the third wave of the EVS collected in 1999.7 Shared values are important as part of the glue that bind a group of people to each other. However, one should not jump from this proposition to the conclusion that all actors within a political system need to share all the same values, all the time, and in all contexts. In fact, perhaps the most important ‘value’ is the belief system that guides the behaviour of individuals in a situation where they do not share the same values. For instance, a political system which rules a society composed of different religious groups is sustainable if the different groups share the overruling value that society should be able to accommodate such religious diversity (see Aberle 1950: 495). Hence, although a range of variables are measured under this indicator, the central variable is the view on democracy as the best political system (assuming that democracy is a system which accommodates diversity of values). To begin with, a number of basic values are examined. This is followed by attitudes to specific suggestions which are nevertheless matters of principle rather than views on policies. Finally, democratic values are discussed.

GOOD AND EVIL, FREEDOM, AND EQUALITY The EVS researchers asked respondents whether they believed that there are clear guidelines between good and evil (Figure 6.18). Such a question is relevant mainly when thinking about dealing with crime, terrorism both internally but also with respect to foreign policy, criminal justice systems, etc. If integration continued with respect to Justice and Home Affairs, values that affect views on justice can be expected to have a large impact on whether integration in these areas will be considered legitimate by European publics. From the demos perspective, such values are relevant because they are closely related to identity. Most respondents (62.3%) did not believe that clear guidelines exist between good and evil. There are, however, great and significant differences between countries. The lowest score for a positive answer was seen in Denmark where only 10.4 per cent replied in the affirmative as well as Hungary (14.8%), Sweden (15.6%), and Greece (16%). The highest scores 181

The Demos—Empirical Analysis Good and evil 100

Per cent

80

Clear guidelines

60

No clear guidelines

40

Disagree with both

20

Gr Fran ea c tB e rita in Ge rm an Au y str ia Ita ly Sp a i n Th P e N ortu ga eth l erl an ds Be lg De ium nm a Sw rk ed en No rth Finla ern n Ire d lan d Ire lan Es d ton ia

0

Country Good and evil 100

Per cent

80 60 40 20

Lit

La tvi a hu Th an ia eC P ze ch olan Re d pu bli c Slo va kia Hu ng ary Ro ma n Bu ia lga r Cr ia oa t Gr ia ee ce Ma Lu l xe mb ta ou Slo rg ve nia Tu rke y Av era ge

0

Country

Figure 6.18. Good and Evil Note: The full text of the question was as follows: Here are two statements which people sometimes make when discussing good and evil. Which one comes closest to your own point of view? A There are absolutely clear guidelines about what is good and evil. These always apply to everyone, whatever the circumstances B There can never be absolutely clear guidelines about what is good and evil. What is good and evil depends entirely upon the circumstances at the time C Disagree with both Source: Figures compiled from EVS 1999.

182

The Demos—Empirical Analysis

were seen in Poland (50.1%), Malta (48.7%), and Northern Ireland (47.1%). In all countries the lowest scoring category is ‘disagree with both’. Although the don’t knows are not reported, it can be concluded that those who decided to answer the question were inclined to agree with one of the value-laden statements. The only country in which more than 10 per cent disagreed with both statements was Italy (13.2%) and the average score for the category was 6.3 per cent as opposed to 31.4 per cent for ‘clear guidelines’ and 62.3 per cent for ‘no clear guidelines’. The countries who were internally most divided on the issue (measured as between 45% and 55% for one category) were Northern Ireland, Poland, and Malta. Another fundamental issue is the balance between individual freedom and equality. At some point all democratic societies have to deal with tensions between them. Assuming that increasing equality automatically leads to a restriction in freedoms, this is a political question, but a fundamental one rather than an attitudinal one. It must be assumed that a respondent’s views on this particular issue, to a very high degree, affect or even guide, his or her attitudes towards both procedural and substantive issues, for example, matters with respect to patterns of democratic participation and general structures of governance as well as policy issues such as, for instance, redistribution, education, and personal security. Hence, a great discrepancy between countries would be negative not only for further European integration but also for further EU democratization as these are principles which—amongst other things—guide the nature of a democratic political system. Such underlying values can have a significant impact on the degree to which legislation that affects either is considered legitimate by the public it affects. Respondents were asked whether, if they were to choose, they would rate freedom over equality and vice versa. The results are shown in Figure 6.19. Most respondents replied that they place freedom over equality. However, in some countries the differences were too small to be significant. This applies to France, Portugal, Northern Ireland, Ireland, and Slovenia. In those countries where most people favoured equality (Italy and Belgium) the differences are also too small to be significant. For ‘freedom above equality’ only 3 countries have a score that is more than ±10 percentage points from the average of 53.2 per cent (Italy with 39.7, Denmark with 69.4, and Slovakia 63.5). Hence,

183

The Demos—Empirical Analysis Freedom vs. equality 100 Per cent

80

Freedom Equality Neither

60 40 20

Gr

ea

Fra

nc e tB rita in Ge rm an y Au str ia Ita ly Sp ain Po rtu Th ga eN l eth erl an ds Be lgi um De nm ar Sw k ed en Fin No lan rth d ern Ire lan d Ire lan d

0

Country Freedom vs. equality

Per cent

100 80 60 40 20 ge

y rke

nia

era Av

Tu

ve

rg

Slo

lta

ou mb

Th

eC

ze

Lu

xe

tia

Ma

oa

ria

Cr

lga

ary

Bu

c

kia

ng Hu

va

bli

Slo

d lan ch

Re

pu

ia

Po

an hu

Lit

Es

ton

ia

0

Country

Figure 6.19. Freedom vs. Equality Note: No figures for Latvia and Romania. The full text of the question was as follows: Which of these two statements comes closest to your opinion? A I find that both freedom and equality are important. But if I were to choose one or the other, I would consider personal freedom more important, that is, everyone can live in freedom and develop without hinderance B Certainly both freedom and equality are important. But if I were to choose one or the other, I would consider equality more important, that is, that nobody is underprivileged and that social class differences are not so strong C Neither Source: EVS 1999.

184

The Demos—Empirical Analysis

there is a relatively high agreement across countries and divisions of similar size within each country. From an integrationist perspective, similar divisions within countries are positive because they mean that views are not divided along national lines. It is likely that it makes a difference to a minority whether any legislation that they oppose for philosophical reasons originates from their nation state or the European Union. Indeed, as has been argued in earlier chapters, because the EU is a ‘new’ regime it is likely that those who at any given time are in the minority will consider EU legislation that goes against them more illegitimate than if this legislation was to be passed at national level. Nevertheless, a high degree of similarity in internal cleavages leaves better scope for integration, although the best scenario from this perspective would be a high level of agreement both within and between the EU member states. From these rather abstract values we can now look at how close Europeans are when it comes to more specific moral or cultural questions. The variables chosen have been selected because they all deal with contemporary issues which are debated and relevant in most countries and on which opinions tend to be strong. This includes views on integration of immigrants, attitudes towards minorities, views on abortion, and views on the individual’s capacity to decide his or her economic/social fortune. On the integration of immigrants, respondents were asked whether they believed that immigrants should adopt the traditions and customs of their new country or whether they should maintain their own. Figure 6.20 shows the results. In 12 of the 28 countries reported, the majority of respondents believed that immigrants should take over the customs of the new country. In some countries this figure exceeded 70 per cent, that is, France, Germany, Austria, the Netherlands, Belgium, and Denmark. The only country where this number was less than 30 per cent was Greece (22.7%) and only in two other countries was this number lower than 40 per cent, that is, Romania (38.2%) and Bulgaria (37.7%). The ‘should adopt’ numbers were generally higher in the older EU member states. However, it should not be overlooked that many states were internally divided on this issue, particularly Great Britain, Spain, Portugal, Northern Ireland, Ireland, Estonia, Latvia, Poland, Slovakia, Croatia, Malta, and Luxembourg. Hence, disagreements on this issue do not only exist between the countries surveyed but also within them. The European Value Survey chose to evaluate people’s views on certain minority groups by asking them to mention those whom they would

185

The Demos—Empirical Analysis

Per cent

100

Adopt country’s customs Maintain own customs

50

ga l erl an ds Be lgi um De nm ark Sw ed en Fin No l a rth nd ern Ire lan d Ma lta

rtu Th

eN

eth

Po

Ita ly Sp ain

y

ia

an

str

rm

Ge

Au

in rita

Gr

ea

tB

Fra

nc

e

0

Country

Per cent

100

50

Th eC Po lan ze ch d Re pu bli c Slo va kia Hu ng ary Ro ma nia Bu lga ria Cr oa tia Gr e ec Lu e xe mb ou rg Slo ve nia

ia an

Lit

hu

ia tvi a La

ton Es

Ire

lan

d

0

Country

Figure 6.20. Should immigrants adopt the traditions and customs of their new country or maintain their own? Note: The table includes data from all EU15 as well as the majority of the recent member states (with the exception of the Republic of Cyprus) as well as Croatia, Bulgaria, and Romania. There are no figures for Turkey. Northern Ireland and Great Britain are listed individually. Source: Data compiled from EVS 1999.

prefer not to have as neighbours. In this study views on homosexuals have been included as well as views on people of a different religion. Figure 6.21a shows the figures per country with respect to homosexuals. 186

The Demos—Empirical Analysis 100

Per cent

80 60 Homosexuals 40 20

k Swe den Finl and Nor ther n Ire land Irela nd

Por

Den mar

Spa in t u g al The Net herl and s Belg ium

Italy

ain Ger man y Aus tria

Brit

Gre at

Fra nc

e

0

Country 100

Per cent

80 60 40 20

key Tur

Lux e

mbo urg Slov enia

ta Mal

atia

ece Gre

Cro

akia

ania Bulg aria

Rom

blic

Slov

nd

epu ch R

ia

The

Cze

Pola

ia

uan Lith

Latv

Esto

nia

0

Country

Figure 6.21a. Undesirable neighbours—homosexuals Source: Data compiled from EVS 1999.

Generally, Northern European countries have lower scores for ‘mentioned’, the lowest being Sweden, where 6 per cent would prefer not to have homosexuals as neighbours, the Netherlands (6.2%), and Denmark (8%). The notable exception is Finland where 20.1 per cent mentioned gays when asked who they would prefer not to have as neighbours. In Turkey, this group was mentioned by 90 per cent of respondents. Other high-scoring countries are Lithuania (67%), Poland (55%), Romania 187

The Demos—Empirical Analysis 100

Per cent

80 Jews Muslims Christians

60 40 20

nd Irela

land

and

n Ire

Finl

Nor

ther

k

den

mar

Swe

Den

s

ium Belg

The

Gre

Net

herl

and

l

in

tuga Por

Spa

Fra nce at B ritai n Ger man y Aus tria

Italy

0

Country 100

Per cent

80 Jews Muslims Christians

60 40 20

key Tur

enia Slov

ta

ourg

emb

Lux

Mal

ece Gre

atia Cro

aria

ania

blic

akia

Bulg

Rom

Slov

epu

The

Cze

ch R

ia

nd Pola

ia

uan

Latv

Lith

Esto

nia

0

Country

Figure 6.21b. Undesirable neighbours—religion Source: Data compiled from EVS 1999.

(65%), and Bulgaria (53.4%). In general—and there are exceptions to this rule—the more recent member states have higher scores for ‘mentioned’. Such differing views may serve as barriers to integration with respect to family law, for example, same sex marriages, inheritance law, pension rules, adoption rights, and other legal issues. As regards the religious groups mentioned by the respondents, the results can be seen in Figure 6.21b. 188

The Demos—Empirical Analysis

Three religious groups have been included, these being Jews, Muslims, and Christians (in Turkey where Muslims were understandably not mentioned). Here, variations between countries are generally small although there are exceptions. For Jews, scores range from 1.7 per cent of Dutch respondents having mentioned this group to 61 per cent in Turkey. Furthermore, in Lithuania, Poland, Romania and Malta the score was over 20 per cent. However, in 13 states the figure was under 10 per cent and in 10 states it was between 10 and 20%. Hence, the majority of respondents did not mention Jews as undesirable neighbours and similar patterns of division can be seen in almost all countries with the exceptions mentioned above. As regards Muslims, the lowest ‘score’ is Portugal where 7.6 per cent mentioned Muslims and the highest is Romania, where 31.4 per cent mentioned Muslims. On average the scepticism is higher in the most recent EU member states than in the EU15. In Turkey, the same question was asked but where ‘Muslims’ was replaced with ‘Christians’, 52.2 per cent of respondents declared that they would prefer not to have Christian neighbours. This is the second largest expression of scepticism towards a religious group measured for the religious categories represented here. ABORTION The issue of abortion has been included for two reasons. Abortion has been seen as a hugely divisive issue in other democratic countries, perhaps most so the United States (see Buchanan 2002) but also more recently in Ireland where membership of the European Union has led to a stirring in attitudes, although a large majority of the Irish population is still opposed to abortion (see Mercurio 2003). Secondly, the question has been included because people’s views on abortion tend to be deeply rooted in intrinsic personal values, be they of a religious or moral nature, but at the same time abortion is a concrete policy issue subject to legislation. Figure 6.22 shows the results from EVS 1999 where respondents have answered about their views on abortion under two different sets of circumstances: (a) if a woman is not married and (b) if a couple does not want any more children. This is clearly a divisive issue. In many countries there is an almost 50/50 split in the population. For instance, for question (a) the following splits exists (first number indicating those who approve): Great Britain (51.8%/48.2%), Austria (44.1%/55.9%), Spain (43.8%/56.2%), Portugal (43.5%/56.5%), Belgium (51.5%/48.5%), Estonia (48.8%/51.3%), Latvia (53.2%/46.8%), Slovakia (44.8%/55.2%), and Turkey (44.3%/55.7%). 189

The Demos—Empirical Analysis Per cent who disapprove of abortion 100

Per cent

80 Woman not married Couple wants no more children

60 40 20

Gr

Fra nc ea tB e rita Ge in rm an Au y str ia Ita ly Sp ain P Th e N ortu ga eth l erl an Be ds lg De ium nm a Sw rk ed en No rth Finla ern n Ire d lan Ire d lan Es d ton ia

0

Country Per cent who disapprove of abortion 100

Per cent

80 60 40 20

Th

eC

Lit

La

tvia

hu an ia P ze o l an ch d Re pu b Slo lic va Hu kia ng ary Ro ma n Bu ia lga ri Cr a oa ti Gr a ee ce Ma Lu l xe mb ta ou Slo rg ve n Tu ia rke y Av era ge

0

Country Figure 6.22. Abortion Source: Data compiled from EVS 1999.

In both categories the highest levels of disapproval can be seen in Malta, Ireland, Northern Ireland, Italy, and Poland—all countries with a large Catholic population. The Swedes are the most likely to approve of abortion under both circumstances, although approval rates are higher there if the woman is unmarried. Despite the large number of countries in which the population is clearly divided over the issue, there are some clear gaps between the extremes, for instance, in Sweden 90 per cent of respondents approve of abortion if the woman is not married whereas in 190

The Demos—Empirical Analysis

Ireland 84.4 per cent of respondents disapprove under such circumstances. Because abortion (in addition to being a useful variable when measuring values) is also an issue that is legislated it is hard to see how such differing views can be reconciled in an attempt at harmonizing legislation. (At least this is so if public opinion is taken into consideration by legislators.) The final question looks at life circumstances and an individual’s scope for influencing his or her economic destiny. The rationale behind including this question is that such views are likely to affect people’s stance when it comes to social legislation and distributive policies in general. Hence, from a socio-psychological perspective, the greater agreement, the better scope for developing a demos. The EVS asked people the following question: ‘Why do people live in need?’. Respondents could give one of the following reasons: unlucky, laziness or lack of willpower, injustice in society, part of modern progress, and none of these. The results are shown in Figure 6.23. 100 90 80

Per cent

70 60 50 40 30 20

Grea

0

Fran ce t Brit ain Germ any Aust ria Italy Spain Portu gal The Neth erlan d Belg s ium Denm ark Swe den Finla nd North ern I relan d Irela nd Esto nia Latvia Lithu ania Pola nd Czec h Re publi Slov c akia Hung ary Rom ania Bulg aria Croa tia Gree ce Malta Luxe mbo urg

10

None of these Modern progress Injustice in society Lazy or no willpower Unlucky

Country

Figure 6.23. Why do some people live in need? Source: Data compiled from EVS 1999.

191

The Demos—Empirical Analysis

Generally, the lowest scoring category is bad luck whereas the highest scoring category is injustice in society. Croatia was the country in which the smallest number of respondents chose ‘unlucky’ (4.3%) whereas the highest score for this answer can be seen in the Netherlands (33.6%). ‘Laziness and lack of willpower’ scored lowest in Sweden (7%) and highest in Malta (50.6%). The countries in which fewest respondents put the fact that some people live in need down to injustice in society are Denmark and the Czech Republic (17% and 19.2%, respectively) whereas the highest scores for this category can be found in Croatia and Turkey (68.9% and 66.7%, respectively). It should be noted that these responses do not necessarily reflect a general similarity in opinions as welfare models in the surveyed countries vary a great deal. It is very likely that responses will be empirically based rather than philosophically. For instance, when a Dane is unlikely to put personal lack of fortune down to ‘injustice in society’ this may be related to the fact that the Danes have a highly developed welfare system, free education, and health care. On the other hand, the Swedish welfare state is perhaps even more generous than that of Denmark and the Swedes are much more likely to mention injustice in society as a reason why some people ‘live in need’ (46.3%). DEMOCRATIC VALUES Perhaps even more important than the values discussed above which may all affect both views on policy and procedure, the final variables discussed are directly concerned with procedure. Three questions have been evaluated in this section, including the role of religious leaders, democracy as a suitable system for the respondent’s particular country, and democracy as a political system in general. A state’s (and its leaders’) commitment to democracy is a prerequisite for membership of the Union, but perhaps more importantly the sustainability of such a commitment is largely dependent on the views of the governed. Views on religion and leadership are important because they are likely to have an impact on political behaviour (see Jagodzinski and Dobbelaere 1998: 105). When asked whether religious leaders should influence government decisions most of the respondents say no. This result is shown in Figure 6.24. The categories have been combined so that those who replied ‘agree strongly’ or ‘agree’ are shown as agreeing and those who disagreed or disagreed strongly are shown as ‘disagree’. In all countries the number of people who would prefer religious leaders not to influence government 192

The Demos—Empirical Analysis Religious leaders should not influence government decisions 100 Per cent

80 Agree Disagree Neither

60 40 20 Gr Fran c ea tB e rita in Ge rm an y Au str ia Ita ly Sp ain P Th e N ortu ga eth l erl an d s Be lg De ium nm a Sw rk ed en No rth Finla ern nd Ire lan Ire d lan d Es ton ia

0

Country Religious leaders should not influence government decisions 100 Per cent

80 60 40 20 La t Lit via hu Th an eC Po ia ze lan ch d Re pu bli c Slo va k i a Hu ng ary Ro ma n Bu ia lga ria Cr oa t Gr ia ee ce Ma Lu xe mb lta ou Slo rg ve nia Tu rke y Av era ge

0

Country

Figure 6.24. Religion and government Note: Respondents were asked: How much do you agree with the following: Religious leaders should not influence government decisions. The following options were given: agree strongly, agree, neither agree nor disagree, disagree, disagree strongly. The chart shows accumulated strongly agree/agree and disagree/strongly disagree. (In the same line of questions respondents were also asked for their level of agreement with other questions about religion and politics, i.e. ‘politicians who do not believe in God are unfit for public office’, ‘religious leaders should not influence how people vote in elections’ and ‘it would be better for (country) if more people with strong religious beliefs held public office’ (EVS 1999). Source: EVS 1999.

193

The Demos—Empirical Analysis

decisions exceeded that of those who did. This includes more religious countries, for example, Ireland (72%) and Poland (81.4%). However, the merged categories hide some large differences in the degree to which people agree. Strongest levels of agreement are seen in France (63.9% strongly agreed), Belgium (57.1% strongly agreed), and Denmark (59.6% strongly agreed). Perhaps surprisingly, the Swedes are the least likely to agree with the statement (24.4% agreed strongly and 27.1% agree) and also in Great Britain the score for strongly agree is lower than in most other nations (20.6%). To the extent that this variable can be seen as a proxy for measuring support for secular government we can conclude that the surveyed populations support secular government to a very high degree. The high level of agreement in all countries shows that this issue is not amongst the divisive ones at EU level. When asked whether they prefer their country to be governed democratically most respondents replied in the affirmative. Figure 6.25 shows the results per country where ‘very good’ and ‘fairly good’ have been combined and ‘fairly bad’ and ‘very bad’ have also been combined. In no country did the accumulated agreement score under 80 per cent although again there are differences with respect to the levels of agreement. The score for very good is lowest in Estonia (16.6%), Latvia (17.5%), and Poland (23.6%). In 15 countries, more than 50 per cent of the respondents replied very good. These were France (52.1%), Germany (50.6%), Austria (62.8%), Italy (57.3%), Spain (50.2%), Belgium (53.6%), Denmark (75.9%), Sweden (71.8%), Northern Ireland (51.7%), Ireland (51.1%), Croatia (54.9%), Malta (59.5), Luxembourg (51.9%), and Turkey (51.9%). By far the strongest commitment to democracy was found in Greece where 82.9 per cent of respondents replied very good. Of these fifteen countries, twelve are EU15 (Northern Ireland and Great Britain counted separately) and two are not members of the EU (Turkey and Croatia). Only in one of the EU2004 did more than 50 per cent of the respondents reply very good (Malta). Hence, despite the strong commitment to democracy there are some indications that the recent EU member states show a lower degree of conviction that democracy is suitable for their country than the older member states, as well as Turkey and Croatia. Asked whether democracy is the best political system, respondents (per country including EU 2004) replied as follows (Figure 6.26). There is overwhelming support for democracy amongst the respondents. 85.4 per cent of the sample agreed with that democracy is the best political system. 194

The Demos—Empirical Analysis

Democracy in our country 100

Per cent

80 60

Good way of governing

40

Bad way of governing

20

Gr

ea

Fra nc tB e rita in Ge rm an y Au str ia Ita ly Sp ain P Th e N ortu ga eth l erl an Be ds lgi De um nm Sw ark ed en No F i nla rth ern nd Ire lan Ire d lan Es d ton ia

0

Country Democracy in our country 100

Per cent

80 60 40 20

an ia l h R and ep ub Slo lic va Hu kia ng ary Ro ma n Bu ia lga ri Cr a oa ti Gr a ee ce Ma Lu xe mb lta ou Slo rg ve ni Tu a rke y Av era ge

hu

Th

eC

ze c

Po

Lit

La

tvi

a

0

Country

Figure 6.25. Democracy in our country Note: Respondents were asked: I’m going to describe various political systems and ask what you think about each as a way of governing this country. For each one, would you say it is a very good, fairly good, fairly bad, or very bad way of governing this country. A Having a strong leader who does not have to bother with parliament and elections B Having experts, not government, make decisions according to what they think is best for the country C Having the army rule the country D Having a democratic political system In the chart ‘very good’ and ‘fairly good’ are shown combined and ‘fairly bad’ and ‘bad’ are combined. Source: EVS 1999.

195

The Demos—Empirical Analysis Agree and strongly agree

Disagree and strongly disagree

Don’t know

100 80 60 40 20

Th eN

Po rtu ga l eth er lan ds Be lgi um De nm ar k Sw ed en Fin No lan rth d er nI re lan d

Ita ly Sp ain

ia Au str

Gr

Fr an ce ea tB rita in Ge rm an y

0

Agree and strongly agree

Disagree and strongly disagree

Don’t know

100 80 60 40 20

ia

tal To

ou

rg

Sl

ov en

lta

mb

Ma

xe Lu

ce ee

y ar ng

Hu

Gr

ia

c

ak

Sl

pu Re

ch Th

eC

ze

ov

bli

d lan Po

ia an

a hu

La

tvi

Lit

nia Es to

Ire

lan

d

0

Figure 6.26. Is democracy the best political system? (Percentage who replied in the affirmative when asked ‘Democracy may have problems but it’s better than any other form of government?’) Source: Data compiled from EVS 1999. (Please note that EVS does not include an aggregate UK result).

196

The Demos—Empirical Analysis

Although this is not entirely obvious from looking at the bar graph, it becomes apparent when testing for significance that there is an association between a respondent’s nationality and his/her level of agreement. The Greeks do justice to their democratic heritage with 65.1 per cent strongly agreeing. Only in Denmark is this percentage higher (although the numbers are too close to reflect a significant difference). But the clearest example is Great Britain, which stands out more than any other, with 15.6 per cent of the sample disagreeing that democracy is the best political system. Another issue is that the 2004 members have yet to become convinced about the virtues of democracy (it should be kept in mind that the data is from 1999). Most of those who agree say that they agree, rather than agree strongly. For instance in Poland, 22.1 per cent of respondents believe that democracy is incapable of maintaining order. (Having said that, this number in France is 18.3%.) But it is, of course, quite possible to believe that democracy is incapable of maintaining order or that the system is indecisive while at the same time believing that it is the best political system. And in France, only 1.3 per cent of respondents strongly disagreed with this statement compared with 15.6 per cent in Great Britain. SUMMARY OF INDICATOR TWO When it comes to values there are clearly some differences between the surveyed countries. However, very often these differences are also present within nation states and hence cannot be interpreted as dividing EU member states into different segments. The highest levels of agreement within as well as between countries can be seen in matters of democratic government. Ideally, both from an integrationist and from a humanist perspective, it would have been preferable if all the responses in the ‘undesirable neighbours’ category had been zero, meaning that nobody minded living next door to any of the groups mentioned. This, however, is far from the case. Furthermore, responses to such questions may not even show the real levels of scepticism and there is a real risk that the negative feelings could be more widespread and that the figures show effects of ‘political correctness’, that is, that some individuals might have been more concerned about certain groups but are reluctant to admit so due to an awareness that such statements are not considered acceptable. Hence, for these categories it is possible that the ‘real’ negative scores are higher than reported. For the same reason, some differences between countries could be due to the fact that political correctness is not equally present 197

The Demos—Empirical Analysis

in all countries. Having said that, the figures show differences between who is ‘unpopular’ in certain countries. Despite differences between scepticism towards religious groups by far the strongest reaction is the Turkish reaction to Christians. At a time when Turkish membership of the European Union is highly debated, this figure is perhaps slightly worrying. The highest levels of agreement can be seen in the commitment to democratic government. First of all, it is important to note the strong reluctance to involve religious leaders in government decisions. This is a clear result in favour of secular government. Another finding is the clear and strong commitment to democracy. Despite a general acceptance that such a system is not perfect, it has significantly more support than its alternatives. (Although it must be kept in mind that definitions as to what constitutes democracy may vary between countries as well as within them.) Even if this is not a surprising finding, it is still an important one. First of all, the EU proclaims to be founded on democratic values. This means that—in Lipset’s terminology—Europeans can identify with the values of the political system and hence are more likely to find the system legitimate. Thus, even if there are diverse attitudes and belief-systems within the Union, there is agreement for ‘the rules of the game’. The importance of such an agreement is summarized by Aberle in his article on values in the United States, in which he states that the political system ‘can tolerate a large number of apathetic voters, but not a large number of individuals who believe that installation of a candidate in office is to be accomplished by storming the White House’ (Aberle 1950: 499).

6.2.3. Indicator Three—A Desire to Shape a Common Future Public opinion on the EU is crucial to future integration, both for reasons related to legitimacy in the sense that public attitudes provide the political foundation for integration (Gabel 1998b: 333), but also because referendums on integration present a crucial constraint on future integration (Palmer and Gabel, forthcoming: 39). This was most recently felt when electorates in France and Holland rejected the European Constitution in 2005. In the current context, however, this variable is only a small part of the larger picture because support for integration in itself does not tell us whether a European demos is emerging. The variable is important for measuring the ‘desire to shape a common future’, hence one central 198

The Demos—Empirical Analysis

variable under this indicator is the extent to which people believe that their country’s membership of the EU is a good or a bad thing. In the Eurobarometer reports this variable equates to ‘support for European integration’, although there is a large conceptual gap between ‘support for European integration’ and ‘membership of the EU good or bad thing’. It is possible that some people are quite happy to support European integration in principle but that they would prefer another model to the one offered by the EU; thus they may say that they think the membership is a bad thing although in principle they ‘support European integration’. For instance, some committed Eurosceptics, those labelled the ‘market Europeans’ by Jürgen Habermas, are fond of integration as long as this equates to a free internal market to compete with the United States, Japan, and increasingly China (Habermas 2000, 2003). In that sense it would have been preferable had the authors asked ‘In your view, should “your country” be a member or not be a member of the European Union’ and then given three response options, ‘yes’, ‘no’ or ‘don’t know’. With this reservation in mind, we can now look at the results with trend lines for each response category (Figures 6.27–6.30). 100 90 80 70 Good thing Bad thing Neither good nor bad Don’t know Linear (Good thing)

Per cent

60 50 40 30 20 10 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Year Figure 6.27. Is membership of the EU a good thing? 1994–2006 with trend line for ‘A good thing’ Source: EB42, EB43, EB46, EB47, EB49, EB52, EB54, EB56, EB58, EB60, EB62, EB63, and EB65.

199

100 90 80 70 Good thing Bad thing

Per cent

60

Neither good nor bad Don’t know Linear (Bad thing)

50 40 30 20 10 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Year

Figure 6.28. Is membership of the EU a good thing? 1994–2006 with trend line for ‘bad thing’ Source: EB42, EB43, EB46, EB47, EB49, EB52, EB54, EB56, EB58, EB60, EB62, EB63, and EB65.

100 90 80

Per cent

70 Good thing Bad thing Neither good nor bad Don’t know Linear (Neither good nor bad)

60 50 40 30 20 10 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Year

Figure 6.29. Is membership of the EU a good thing? 1994–2006 with trend line for ‘neither good nor bad’ Source: EB42, EB43, EB46, EB47, EB49, EB52, EB54, EB56, EB58, EB60, EB62, EB63, and EB65.

200

The Demos—Empirical Analysis 100 90 80 70 Good thing Bad thing Neither good nor bad Don’t know Linear (Don't know)

Per cent

60 50 40 30 20 10 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Year

Figure 6.30. Is membership of the EU a good thing? 1994–2006 with trend line for ‘don’t know’ Source: EB42, EB43, EB46, EB47, EB49, EB52, EB54, EB56, EB58, EB60, EB62, EB63, and EB65.

There are variations throughout the period and no significant trend exists as far as ‘good thing’ is concerned. The ‘neither good nor bad’ group constitutes almost one-third of the respondents and this has been the case throughout the period although this number was lower in the mid-90s. The ‘don’t know’ category is, very low, that is, currently 4 per cent down from its highest of 10 per cent in 1999. Despite blips, the development in the numbers is quite insignificant. There are no major shocks. Furthermore, there is nothing to indicate that the most recent members are more or less positive about their country’s membership of the EU than the publics in ‘older’ member states. The EU25 and EU15 numbers are nearly identical with the only differences being between ‘neither good/nor bad (twenty-seven for the EU15 and twenty-eight for the EU25) and ‘bad thing’ (thirteen for the EU25 and fourteen for the EU15). Having said that, there are variations between individual countries. If we look at the support by country (most recently available data at time of writing), the picture looks as follows (Figure 6.31): 201

The Demos—Empirical Analysis

The

0

Pola nd Portu gal Slov enia Slov akia Finla nd Swe den UK

50

Belg ium Czec h Re p. Denm ark Germ any Esto nia Gree ce Spain Fran ce Irela nd Italy The Rep. of Cy prus Latvia Lithu ania Hung ary Luxe mbo urg Malta The Neth erlan ds Aust ria

Per cent

100

Don’t know Neither good nor bad Bad thing Good thing

Country

Figure 6.31. Country’s membership good or bad—2006 figures (by country) Source: Data compiled from EB65.

In most member states a high proportion of respondents declare that their country’s membership of the European Union is a good thing. In 15 of the 25 member states over 50 per cent of respondents have selected this category. This applies to Belgium (65%), The Czech Republic (52%), Denmark (65%), Germany (57%), Estonia (51%), Greece (53%), Spain (72%), Ireland (77%), Italy (56%), Lithuania (59%), Luxembourg (72%), The Netherlands (74%), Poland (56%), Slovenia (54%), and Slovakia (55%). In no member state has the ‘bad thing’ category been chosen by the highest number of respondents. Overall, only 13 per cent of the respondents believe that their country’s membership of the EU is a bad thing, although this figure ranges from 6 per cent (in Spain, Slovakia, and Slovenia) to 26 per cent (in Finland). Other countries with a high score for ‘bad thing’ are Austria (24%) and the United Kingdom (25%). There is nothing to suggest that more recent members are more or less supportive 202

The Demos—Empirical Analysis

of their country’s membership of the Union. Unlike the measurements of identification with the polity (under indicator one) where founding members appeared to be feeling more European, support for membership is not stronger in founding members than later joiners. Although Belgium, Luxembourg, and the Netherlands see scores of over 70 per cent for ‘good thing’, so does Spain with 72 per cent. Ireland enjoys the highest ‘good thing’ score of all countries surveyed with almost 4/5 of respondents declaring that they believe their country’s membership of the EU is a good thing (77%). Even in states whose populations have traditionally been restrained in their enthusiasm for European integration, the score for ‘good thing’ is higher than ‘bad thing’. In Denmark, 65 per cent of respondents believe their country’s membership of the union is a good thing as do 42 per cent in the United Kingdom. Hence, on this basis it appears that membership of the EU enjoys strong popular backing across the board. However, the category which scores the second highest average should not be overlooked. On average 28 per cent of respondents believe that their country’s membership of the Union is neither good nor bad. If we add to this category the 4 per cent who replied ‘don’t know’ this amounts to almost a third of Europeans who are somewhat indifferent. If we add to this the number of respondents who believe their country’s membership of the Union is a bad thing, we can conclude that 45 per cent of the respondents do not forcefully back their country’s membership of the Union. To get a view of how this looks in individual countries, Figure 6.32 shows the division of respondents in two groups. ‘Good thing’ naturally consists of those who selected the category ‘good thing’ whereas ‘other responses’ covers those who selected either of the other three categories, ‘bad thing’, ‘neither good nor bad’, and ‘don’t know’.8 In ten member states those who did not select ‘good thing’ are in the majority. And in some states they are so by quite a large margin. This applies to 5 states, Latvia (63% not good thing), Malta (56%), Austria (65%), Finland (61%), and the United Kingdom (58%). Again, however, it must be noted that most people are indifferent rather than downright hostile. On the other hand, the fact that 45 per cent of the aggregate do not reply ‘good thing’ cannot be interpreted as a particularly strong backing for European integration and it remains a fact that approximately onethird of respondents (average) are indifferent (combined ‘neither good nor bad’ and ‘don’t know’). Indifference is also apparent when we look at the next variable, that is, how would people feel if the EU were to be scrapped? The respondents 203

The Demos—Empirical Analysis Other responses Good thing

Per cent

100

50

Belg The ium Cze ch R Den ep. m Ger ark man y Esto nia Gre ece Spa i Fra n nce Irela nd The Rep Italy . of Cyp rus Latv ia Lith uan Hun ia ga Lux emb ry ourg The Net Malta herl and s Aus tria Pola nd Por tuga Slov l en Slov ia akia Finl and Swe den UK

0

Country

Figure 6.32. Country’s membership a good thing or other (2006 figures) Source: Data compiled from EB65.

were asked: ‘If you were told tomorrow that the European Union had been scrapped, would you be very sorry about it, indifferent or very relieved?’. The results are shown in Figure 6.33. In just under half the member states the highest scoring category is ‘indifferent’. This is the case in 8 of 10 states that joined in 2004 (Estonia, Hungary, Slovenia, Slovakia, Lithuania, the Czech Republic, Latvia, and Poland) and in 5 of the EU15 (the United Kingdom, Sweden, Finland, Spain, and Austria). Also, stronger feelings seem to exist in older member 204

The Demos—Empirical Analysis Don’t know Very relieved Indifferent Very sorry

Per cent

100

50

Italy prus Latvia Lithu ania Hung ary Luxe mbo urg Malta The Neth erlan ds Aust ria Pola nd Portu gal Slov enia Slov akia Finla nd Swe den UK of Cy

Rep.

The

The

Belg ium Czec h Re p. Denm ark Germ any Esto nia Gree ce Spain Fran ce Irela nd

0

Country

Figure 6.33. Feelings if the EU were to be scrapped Source: Data compiled from EB62.

states in that amongst these, respondents are more likely to declare themselves ‘very relieved’. If we look at countries in which this category scored over 20 per cent, this applies to the United Kingdom (27%), Sweden (26%), and Finland (23%). The only 2004 accession state which in this reply was chosen by over 20 per cent or more of the respondents was Malta (20%). In three states over half the respondents declared that they would be very sorry to see the end of the EU. This applies to Luxembourg (63%), Ireland (54%), Portugal (53%), and in another 11 member states more than 40 per cent of respondents selected this answer, these being Germany and the Netherlands (both 49%), Belgium and the Republic of Cyprus (both 48%), Italy (45%), Greece (44%), France and Malta (both 43%), Spain (42%), and Denmark and Hungary (both 40%). Hence, a rather large 205

The Demos—Empirical Analysis Table 6.5. Feelings if the EU were to be scrapped

Very sorry Indifferent Very relieved Don’t know

1995

1998

2001

2004

39 38 13 9

36 39 13 13

28 45 14 13

39 43 13 5

Source: Data from EB43, EB50, EB55, and EB62. Some inconsistencies due to the way EB calculate the average.

number of people would be very sorry to see the end of the EU. However, if we add the indifferent to the very relieved and the ‘don’t knows’, only three member states have a higher proportion of respondents who would be very sorry than display an attitude of indifference or outright relief. Table 6.5 shows the developments over time. As can be seen, the trends change, although overall the results are relatively stable and variances may be too small to matter. Although there was a significant drop in ‘very sorry’ in 2001, the figures have reached 1995 levels in 2004 where also the indifferent group has grown. (The ‘don’t know’ group has fallen by 4 percentage points, which may be able to explain the higher score for indifferent.) As always, one should resist the temptation to ‘over-conclude’ on the basis of numbers, but what we can say is that in most states most people do not care whether the EU continues to exist and that this group, together with the ‘relieved’ group, outnumbers the ‘very sorry’ group. This cannot be said to constitute strong backing for the European Union. It should, however, be noted that the numbers refer to the specific institution of the EU and not European integration as an idea. It is quite possible that there are individuals sympathetic to the idea but unconvinced by the performance, institutional set-up, or other credentials of the Union. As always, there are divisions within each country. There are some groups who are historically more sceptical towards the EU than others. Eurobarometer Report 64 summarizes this neatly, stating those who are most supportive of their country’s membership of the Union are those who have studied the longest, men, the younger groups, people living in urban areas, and people who have a higher level of knowledge about the Union (Eurobarometer Report 64: 55). Furthermore, there are also figures that show that white collar workers are more likely than blue collars to be supportive of their country’s membership of the union. 206

The Demos—Empirical Analysis

The question of age is particularly interesting. If it could be demonstrated that there is a positive relationship between youth and the extent to which a person supports membership and/or feels attached to the EU (or Europe), it is likely that there will be a positive development towards a demos in the future. This, however, will only happen if people do not change their minds as they grow older. Previous studies on the relationship between age and support for European integration are divided. Some find that young people are more positive towards European integration than older generations (for instance Inglehart 1970: 788; Lindberg and Scheingold 1970: 256). Others find no correlation. In their study on the relationship between age and support for European integration, Bosch and Newton concluded that they had found no evidence to show ‘ . . . that support grows because an older and less favourable group is replaced by a younger and better disposed one’ (Bosch and Newton 1995: 102). Similarly, a study of the relationship between a range of socio-demographic variables and sense of European identity in the period 1983–91 concluded that, with the exception of Greece, Spain, and Portugal, age ‘is not perceptibly related to feelings about European identity’ (Duchesne and Frognier 1995: 212). The reasons for the differences in results are difficult to uncover, although there appears to be a pattern that older research (from 1970s and early 1980s) is more likely to show a correlation between age group and support/identification. For this purpose the most recent numbers are from 2004 (EB61) and only includes the EU15. The figures below show the results for six age bands (15–24, 25–34, 35–44, 45–54, 55–64, and 65+) and for two questions, that is, age versus support for integration, and age versus degree of ‘feeling European’ (Figure 6.34). First of all, the youngest group is the most likely to have an opinion at all and also more likely to feel that membership of the Union is a good thing. The 65+ group is the least likely to have an opinion and also the least likely to select ‘good thing’. As regards the other four groups, the differences are very small. ‘Bad thing’ is chosen by 18 per cent of the 65+ group, but there are more people choosing ‘bad thing’ in the two younger groups, the 55–64-year olds and the 45–54-year olds. Although these differences are too small to be significant we cannot conclude that there is a linear relationship between age increase and ‘bad thing’. Having said that, it remains a fact that 59 per cent of the 15–24-year olds chose ‘good thing’ as their reply compared to only 45.7 per cent of the 65+, indicating that membership of the Union is more popular amongst the youngest people surveyed than amongst the oldest people surveyed. 207

The Demos—Empirical Analysis 65+ 55−64 Good thing Bad thing Neither good nor bad Don’t know

45−54

Age 35−44 25−34 15−24 0

20

60

40

80

100

Per cent

Figure 6.34. Age vs. good thing or bad thing Source: Data compiled from EB61 covering the EU15.

Hence, there is a clear trend that the older the people, the smaller the proportion who respond ‘good thing’. The findings are somewhat similar in the next category, age versus geographical identification. The result of the cross tabulations can be seen in Figure 6.35. For the two marginal groups this shows a similar pattern to the one seen in Figure 6.34. Of the youngest group, 39.6 per cent identify themselves purely in terms of their nationality whereas for the oldest group this applies to 56.8 per cent. Furthermore, there is a consistent development in terms of the four other groups in the sense that the older people are, the more likely they are to define themselves in terms of their nationality 65+ 55−64 Nationality only Nationality and European European and nationality European only

45−54

Age 35−44 25−34 15−24 0

20

40

60

80

Per cent

Figure 6.35. Age vs. feeling European Source: Data compiled from EB61 covering the EU15.

208

100

The Demos—Empirical Analysis Table 6.6. Feeling European vs. membership good thing or bad thing

Nationality only Nationality and European European and nationality European only Don’t know

A good thing

A bad thing

Neither + don’t know

37.2 64.5 70.5 60.5 34.5

25.8 8.1 7.8 10.0 21.6

37.0 27.4 21.7 29.6 43.0

only. For the category ‘European only’ we see a different pattern. The oldest (65+) are least likely to declare themselves European only (2.8% of respondents). The youngest (15–24) are more likely to put themselves in this category (3.6%), but they are not the most European. The proportion answering ‘European only’ rises with age, peaking at 5.2 per cent for the 45–54-year olds, before dropping to 2.9 per cent for the 55–64-year olds (and 2.8% for the 65+ group). One can only speculate as to why this is the case; however, it may be possible that the youngest have yet to take on a more international lifestyle which occasionally follows a career in business or other professions. Nevertheless, the trend is the same and there are signs that the older groups feel less European than the younger groups. With respect to the relationship between feeling European and support for membership we can test the following hypothesis (Table 6.6). The more European a person feels the more likely he or she is to support his or her country’s membership of the Union. There is a clear relationship in the sense that those who declare themselves ‘nationality and European’ are much more likely to choose ‘good thing’ than ‘bad thing’. This tendency is even stronger for those who claim to be ‘European and nationality’. However, for those who declare themselves ‘European only’ this is not the case with respect to ‘good thing’. In fact, more people in this group believe their country’s membership is a bad thing than is the case for the ‘European and nationality’ group as well as the ‘nationality and European’ group. Also, those who define themselves only in terms of their nationality are more likely to reply ‘neither good nor bad’ or ‘don’t know’ than those who replied ‘nationality and European’, ‘European and nationality’ or ‘European only’. The tendency is clearer when we look at whether people believe their country has benefited from EU membership. Table 6.7 shows this relationship. 209

The Demos—Empirical Analysis Table 6.7. Benefited vs. good thing or bad thing

Benefited Not benefited Don’t know

A good thing

A bad thing

Neither + don’t know

80.9 15.2 51.2

2.9 42.7 16.4

16.2 42.1 67.9

Of those who believed so, 80 per cent replied that membership of the EU was a good thing as opposed to only 15.2 per cent of those who did not believe their country had benefited. Interestingly, 16.2 per cent of those who believe their country has benefited from EU membership replied either ‘neither good nor bad’ or ‘don’t know’. It is likely that ideology and identification can be involved in the sense that some individuals may be inclined to support integration for ideological or political reasons whereas others may recognize the benefits but not have their heart in it (Inglehart and Reif 1991) or be opposed due to concerns about politics or democracy (Hix 2005: 150). For instance, one study has shown that if people feel under-represented in the EU context, their support for the EU is reduced independently of economic perceptions (see Rohrschneider 2002). Having said that, this is not to say that perceived benefit is the only factor influencing support for integration, neither that the perceived benefit is the same across member states. There are many different factors that affect support for European integration (Gabel 1998b). Gabel tested five explanations of public support for European integration, that is, cognitive mobilization, political values, utilitarian appraisals, class partisanship, and support for current domestic government (Gabel 1998b). This study concluded that the utilitarian theory and class partisanship were ‘robust’ explanations for variation in support (Gabel 1998b: 351), and that support for domestic government provides a ‘systematic explanation for support for integration except in the original member-states in the post-SEA period’. As regards political values and cognitive mobilization, the study found that such theories only provided explanations in the original EU member states and that compared to the utilitarian element had only ‘relatively small impact on support’ (Gabel 1998b: 352). Subsequent studies confirm that support for integration depends to some extent on economic factors, but that varying degrees of support are more likely to depend on nation-specific differences (Palmer and Gabel, forthcoming: 8). This is a similar result to that found in an earlier 210

The Demos—Empirical Analysis

study, which concluded that ‘So far as economic variables do appear to be associated with support or opposition to the EC and European unification, they seem to have different effects in different countries and in different years. This makes it difficult to draw out any general conclusions’ (Bosch and Newton 1995: 101). As mentioned elsewhere, scholars have shown a clear relationship between strength of national identities and lack of support for European integration (see for instance Carey 2002). Others find that those inclined to feel identification with one polity (e.g. their nation state) are also inclined to identify with the EU (Hooghe and Marks 2004). In a similar vein, it has been shown in this chapter that those who feel attachment to one polity are also inclined towards declaring a higher level of attachment to another. The latter supports the theoretical arguments made by Easton and Smith (see Chapter 3) that individuals often have multiple identities (as in consider themselves members of more than one group). If this is the case, it is perhaps hard to see exactly how these identities affect support for a given polity. In the case of the EU, Hooghe and Marks suggest that the level of support is determined by the type of identification the individual feels. The authors also argue that it is possible that strong national identities can lead in both directions because national histories also act as independent variables. Quoting a study by Diez Medrano the authors argue that: Constructing patterns of discourse in the UK, Spain and Germany, Medrano finds that English Euroskepticism is rooted in Britain’s specific history of empire, that West German pro-Europeanism reflects World War II guilt, and that the Spanish tend to support European integration as proxy for modernization and democratization (Hooghe and Marks 2004: 416).

This is a simple yet relatively convincing argument. If true, it means that those who wish to address Euroscepticism, for instance, in order to win a referendum on the constitution, may need to emphasize different advantages of integration in different countries. It must be stressed again that this book does not seek to explain support or lack of support for European integration. Hence, favouring one explanation over another on either theoretical or empirical grounds will not lead to a conclusion that can highlight whether a demos is forming. The truth is probably that all these factors, and perhaps also other factors, are determinants of an individual’s support for the EU and that the force of each explanation may vary between individuals. Thus, 211

The Demos—Empirical Analysis Table 6.8. The meaning of being a citizen of Europe 1. Right to move 2. Right to work 3. Local elections 4. National elections 5. EP elections 6. Health care 7. Studies 8. None of the above SPONTANEOUS 9. Don’t know

51 57 21 17 22 32 42 3 10

Source: Data compiled from EB55.

as regards this variable (support for European integration) it is the actual score that is important, albeit seen in relation to the variables discussed below.

THE MEANING OF BEING A CITIZEN OF EUROPE? What does it mean to be a ‘citizen of Europe?’ To answer this question, respondents were given a list of options. Table 6.8 shows the EU15 score for each choice. It must be stressed that the options given to respondents are predominantly instrumental and appeal to a sense of rational choice and individual gain maximization. There are no real symbolic or moral statements that people could have chosen from. As can be seen, most people mention the right to work, relocate and study in other member states and the right to health care when visiting other countries. However, there is a pattern in the answers that the Eurobarometer Report does not pick up on. It is clear that the list contains two types of choices, namely the pragmatic ones that apply to the opportunities of the individual and the civic ones that apply to the taking part in the collective democratic process. To that extent 1, 2, 6, and 7 are the pragmatic ones and 3, 4, and 5 are civic, that is, all matters of elections. All of these choices have far lower scores than the utilitarian, pragmatic choices such as health care and the right to move, work, and study. It would be necessary to carry out more research into these findings but they clearly indicate that Europeans are interested in travel and spending time in other European countries but without engaging or even integrating at the civic level. Whether there is a spurious effect here in the sense that perhaps Europeans do not care about civic matters, full stop, does not really affect the result because that 212

The Demos—Empirical Analysis

would still mean that they were not integrating at the civic European level. Subsequent data of a similar nature, including results from all twentyfive current member states, confirm these tendencies. In a special Eurobarometer from 2006, respondents were asked to spontaneously say what came to their mind when they heard the words ‘European Union (SEB 251). People mentioned a variety of things such as cooperation, unity, equality, common decisions, and rules, legislation (EU25 = 22%) and Euro, one single currency (EU25 = 15%). Answers that could be categorized under ‘feeling European, European citizenship’ came to 0 per cent for the EU25 (1% in Denmark, Greece, Spain, Ireland, Republic of Cyprus, Hungary, the Netherlands, Austria, and Sweden, 0% in all other countries). Hence, Europeans do not spontaneously think of themselves in terms of identification with Europe, whether in civic terms or psychologically.

THE VETO The veto is a crucial indicator of a respondent’s feeling of identification because ‘giving up’ the veto constitutes an effective acceptance of the EU level as the legitimate space for decision-making. Thus, whether a person feels that the veto should be retained can be seen as a very important factor in how integrated he or she feels with the EU polity or, at least, how integrated he or she would like to be. This is, of course, assuming that respondents know what the veto is and to an extent how, where and under what circumstances it can be applied (Table 6.9).9 Of the EU15, 71.8 per cent believe that the veto should be retained, although 21.7 per cent only believe so in ‘few essential areas’. Only 10.9 per cent would be willing to sacrifice the national veto in favour of efficiency. Almost one-fifth of the respondents replied don’t know. Hence, there is a strong majority in favour of retaining the national veto. This issue is addressed further under indicator four. Table 6.9. The veto

EU15

Be retained in order to preserve national interest

Be limited to a very few essential areas

Be given up for all decisions to make the EU more efficient

Don’t know

50.1

21.7

10.9

17.3

213

The Demos—Empirical Analysis

SUMMARY OF INDICATOR THREE On the one hand, more people believe that their country’s membership of the European Union is a good thing than a bad thing. On the other hand, they would not be particularly sorry if the EU were to be dissolved. This is good news to those integrationists who assume that those who are not sure are really in support. But, of course, this cannot be taken for granted, particularly in the light of the fact that a majority would not be sorry if the EU were to be scrapped tomorrow. This does not constitute a great desire to form a common future. There are, however, differences between age groups. And although this is a disputed issue, the data used in this particular study show—in general terms—that the youngest have more positive feelings about the Union than the oldest generations. However, what this means for the future is still uncertain. If levels of support are generational, there will arguably be a shift towards more supportive older people as time goes by. If, on the other hand, views are age-specific this may not be the case. The strong desire to retain the national veto indicates that respondents, even those keen on integration, prefer to maintain their formal national sovereignty. As far as this indicator goes, it seems fair to conclude that people are very fond of cooperation although perhaps not equally supportive of irreversible integration. This again is illustrated by the fact that people are fond of the fact that they can travel (for whatever purposes) across the region but that they do not feel any particular desire to integrate at the civic level despite their increased fondness for international lifestyles.

6.2.4. Indicator Four—Public Awareness of the Transnational Process The inclusion of this indicator is based on the logic that a certain amount of knowledge is necessary if people are to develop a stronger attachment to the Union (Haubrich 1997: 45). However, perhaps more importantly, much research into feelings about the EU rely on survey data. If levels of knowledge are low, it must be assumed that survey data on specific questions about the Union are likely to become less reliable. And although public support for the EC in general has been described as widespread, although ‘rather flimsy and without conviction’ (Bosch and Newton 1995: 103), there is evidence that fluctuating views are more common amongst individuals who possess little knowledge about the EU. This is particularly the case when low levels of knowledge are combined with low levels of interest, which is—unsurprisingly—often the case. Hence, apathy suggests that public opinion regarding European integration is superficial and 214

The Demos—Empirical Analysis

transitory (Palmer and Gabel, forthcoming: 6) and low levels of knowledge mean that public opinion is easily swayed in either direction (Saris 1997). In a study published in 1997, Dutch researchers carried out experiments which showed that the less people know about the EU, the easier it is to change their opinion. The researchers divided respondents into groups according to levels of knowledge, asked a range of questions about respondents’ views on European integration and subsequently subjected them to positive information about the EU before asking the same questions again. The result showed significant change of opinion for the groups with little knowledge, who now responded more positively to the questions. Finally, the researchers subjected the same respondents to negative information and again asked them the same set of questions. Those with little knowledge about the EU changed their opinions once more and responded more negatively to the same questions (see Saris 1997). For instance, if people are not aware of the institutions of the Union, it makes no sense to ask them whether they are satisfied with democracy in the Union. Even if one takes the view that the central parameter is perception, e.g. if one takes the view that a positive score can be interpreted to mean that the Union enjoys legitimacy, an opinion based on perception rather than knowledge may be more easily swayed than an opinion based on knowledge. Thus, in cases where knowledge is low, credibility of answers to normative questions may suffer and thereby compromise reliability of such survey results. Hence, it appears that a higher level of knowledge is likely to produce more sustainable answers and hence, make the survey responses more stable and less subject to random answers. It is perhaps improper to use the phrase ‘a well-known fact’ in an empirical analysis. However, when it comes to people’s knowledge about the EU, its institutions and processes, it really is a well-known fact that there is widespread ignorance amongst EU citizens. As mentioned earlier, this is likely to be connected to the fact that the majority of people are not particularly interested in matters related to the EU and that they feel the EU is remote and perhaps not particularly important to them. On the other hand, lack of interest does not equate hostility or even apathy. One must assume that lack of interest is as likely to be due to lack of hostility as to lack of support. Finally, lack of interest could be due to the fact that many salient policy areas are not primarily governed by the EU—that is, health care provision, education, law and order, pensions, social security, and tax (Moravcsik 2002: 615). Under the assumption that there is a relationship between an individual’s knowledge about the European Union and his or her feelings 215

The Demos—Empirical Analysis

towards integration, Eurobarometer regularly seeks to establish the level of knowledge each respondent possesses about the Union, its institutions, and its policies. There are three indicators which are used for this purpose. The first is known as ‘self-perceived knowledge’. In this category respondents are asked to place their own level of knowledge on a scale from 1 to 10, 10 being the maximum level. The second indicator used to test this is to ask respondents whether they have or have not heard about a specific institution. Finally, recent surveys regularly pose a list of specific questions with respect to the Union. This section is sometimes referred to as a ‘quiz’. The following section will discuss the self-perceived knowledge at the time of the Convention, as well as levels of knowledge surrounding the Maastricht ‘crisis’, although the emphasis will be on recent quiz results.

KNOWLEDGE ABOUT THE EU AT THE TIME OF THE EUROPEAN CONVENTION Asked this question: ‘Using this scale, how much do you feel you know about the European Union, its policies, its institutions and bodies’, respondents were shown a card with the scale 1–10 (EB58). By far the largest group of respondents across countries (EU15 = 52%) perceive themselves as knowing ‘a bit’ about the EU with the EU15 average being nineteen for ‘know (almost) nothing’ and a quarter of the inhabitants knowing ‘quite a lot’. The United Kingdom is at the bottom of the list, with 36 per cent admitting to knowing almost nothing, and Denmark and Austria having the lowest proportion of respondents who claim to know almost nothing (both 8%).

THE CONVENTION When asked with which EU institutions they were familiar, almost twothirds of the respondents claimed they had never heard of the European Convention (65%), a figure which is similar in EB59 (63%). The fact that at the time the fieldwork for EB59 was carried out (March–April 2003), the European Convention’s work was almost complete confirms Schmitter’s comment cited elsewhere that the Convention went largely unnoticed. This means that people will not have had an opportunity to offer their views or input, contact members of the Convention for a dialogue on the matter or even just have informal discussions about the Convention with other citizens. Considering the fact that one of the

216

The Demos—Empirical Analysis

aims of the Convention was to bring the EU closer to the people, one could say that even before producing its final draft, the Convention had already failed in one of its major tasks. (Unless of course, one takes the view that one-third is sufficient.) At the same time, almost one-fifth had never heard of the European Commission and a third had never heard of the Council of Ministers, a table topped by the United Kingdom with a score of 56 per cent. MAASTRICHT Going back in time to one of the biggest recent crises to hit the European integration process, it is interesting to look at the results from Eurobarometer 38 (fieldwork September–October 1992), more specifically the question: ‘Have you recently heard anything about the Treaty of Maastricht?’ The response for the EU12 is 85 per cent comparatively high with a view to the results shown above regarding the Convention. Perhaps this could indicate that when there are great controversies and disagreements and hence additional media attention, people cannot avoid hearing about the EU but that they would not go out of their way to get information. This assumption is partly backed by the follow-up question, which asked respondents how much they knew about the Maastricht Treaty. Most replied ‘know just a little’ (EU12 = 47 %), but 28 per cent said they had heard of the treaty, but knew ‘nothing else’. A similar question was raised in EB48 regarding the Treaty of Amsterdam. Here only 34 per cent of the EU15 had heard about the treaty. Outliers are the Netherlands (74%) and Denmark (91%). The Dutch result could probably be explained by the fact that the treaty was signed in Amsterdam. As regards the Danish result, the Danes had to ratify the treaty in a referendum, which is likely to account for a large proportion of the high positive response rate. Thus, the fact that people have heard about a treaty does not mean that they know anything about it or that they even care about it. In many cases, the reason people have heard about a treaty is that they could not have avoided hearing it mentioned, for instance, on the radio or on TV. RECENT QUIZZES ABOUT THE EUROPEAN UNION Table 6.10 shows the questions, correct answers, and proportion of correct answers per country for the most recent quiz. For the first question, ‘the European Union currently consists of fifteen member states’ the correct answer is false. At the time of data

217

EU25 Quiz question

The European Union currently consists of fifteen member states The members of the European Parliament are directly elected by the citizens of the European Union (our country) has a European Commissioner Source: Eurobarometer 65 annexes.

% of % of % of % of % of respondents The Correct answer respondents respondents respondents respondents who gave the incorrect who replied who replied who replied who gave the answer or answered ‘True’ ‘False’ ‘Don’t know’ correct answer ‘Don’t know’ False

30

48

21

48

51

True

49

30

19

49

49

True

72

9

19

72

28

The Demos—Empirical Analysis

218 Table 6.10. Quiz questions and results 2006

The Demos—Empirical Analysis

collection the EU consisted of twenty-five member states. However, less than half of the respondents (48%) answered this correctly. A similar result can be seen for the next question (whether members of the EP are directly elected). Forty-nine per cent of respondents replied in the affirmative. Thirty per cent believed this statement to be false. Finally, most people know that their country has a European Commissioner (72%). Only 9 per cent gave the incorrect answer whereas 19 per cent did not know. The more extensive quiz from 2004 provides better scope for thorough analysis. This quiz asked respondents ten questions about the European Union. The results are shown in Table 6.11. Again, people don’t know how many member states the EU has. At the time of data collection for this survey the Union had fifteen member states. Nevertheless, only 33 per cent answered this question correctly. Only 22 per cent know that the Commission’s headquarters are not in Strasbourg and only a quarter of the respondents know that the EU has its own anthem. Sixty-seven per cent believe that the EU has one star per member state. (This is perhaps not surprising as the flag has 12 stars and most people believe that the Union has 12 members and most respondents—81%—were aware that the EU flag is blue with yellow stars.) Again, knowledge about the European parliamentary elections is low. Only 29 per cent knew that there were not going to be elections to the EP in 2006. In the next few paragraphs we look at whether the levels of knowledge (as measured by right answers in the quiz) affect people’s views about the Union. A range of relationships have been tested, including level of knowledge versus support for membership, level of knowledge versus ‘feeling European’, level of knowledge versus feelings about the veto, level of knowledge versus satisfaction with democracy, and level of knowledge versus trust in the European Union. Figure 6.36 illustrates the relationship between knowledge and support for membership. (Level of knowledge is measured as the number of correct quiz answers given by a respondent. Support is measured as earlier, that is, does the respondent think his or her country’s membership of the EU is a good thing, a bad thing, neither good nor bad, or doesn’t know.) There is a clear relationship between these two variables. Those with a higher number of correct answers are more likely to believe that their country’s membership of the EU is a good thing. This, however, does not necessarily illustrate a causal relationship. Although it is likely that increased information leads to more positive feelings, it is equally likely 219

EU15 Quiz question

The European Union currently consists of twelve member states The European Community was created just after the First World War, in the early 1920s The European flag is blue with yellow stars On the European flag, there is one star for each member country The headquarters of the European Commission are in Strasbourg The members of the European Parliament are directly elected by the citizens of the European Union The president of the European Commission is directly elected by the citizens of the European Union The European Union has its own anthem Each year, there is a Europe Day in all the countries of the European Union The next elections to the European Parliament will take place in June 2006 Source: Data compiled from Eurobarometer 61 annexes.

% of The Correct % of % of % of % of respondents answer respondents respondents respondents respondents who gave the incorrect who replied who replied who replied who gave the answer or answered ‘True’ ‘False’ ‘Don’t know’ correct answer ‘Don’t know’ False

47

33

23

33

56

False

13

55

32

55

45

True False

81 67

5 13

14 20

81 13

19 87

False

49

22

29

22

78

True

50

21

29

50

50

False

17

48

35

48

52

True True

25 34

36 24

39 42

25 34

75 66

False

25

29

46

29

71

The Demos—Empirical Analysis

220 Table 6.11. Quiz questions and results 2004

The Demos—Empirical Analysis Total right vs. support for membership 100

Per cent

80 Good Bad Neither Don’t Know

60 40 20 0 0

1

2

3

4

5

6

7

8

9

10

Number of correct answers given

Figure 6.36. Quiz results vs. support for membership Source: Data compiled from EB61 (ZA4056).

Total right vs. feeling European 100 Nationality only Nationality and European European and Nationality European only Don’t know

Per cent

80 60 40 20 0 0

1

2

3

4

5

6

7

8

9

10

Number of correct answers given

Figure 6.37. Quiz results vs. feeling European Source: Data compiled from EB61 (ZA4056).

that more positive feelings, perhaps grounded in idealistic thinking, can lead to more active seeking of information and hence contribute to a higher level of knowledge. Nevertheless, there is a clear relationship, particularly for those who gave the correct answer to eight or more questions. Those with a higher number of correct answers are also more likely to have a view at all. 221

The Demos—Empirical Analysis

As regards the level of knowledge versus feeling European, Figure 6.37 shows the results of cross tabulating these variables. Again, there is a clear relationship although it is not a linear one and the trend changes around those who do moderately well in the quiz. Of those whose gave incorrect answers to all 10 questions less than 20 per cent define themselves in terms of nationality and European whilst over 70 per cent define themselves only in terms of their nationality. This number declines until nine correct answers, whereas those who gave the correct answer to ten questions were more likely to define themselves strictly in terms of their nationality. There is a steep fall between 9 and 10 also for nationality and European. However, this is balanced by an increase in those who replied only European or European and nationality. Hence, for those who answered 10 questions correctly, 14.5 per cent declared themselves European and nationality and 12.7 per cent only European. These numbers are significantly higher than for any other groups. Thus, again we must conclude that despite variations and the lack of linear relationships per category, the overall conclusion is that people who know more about the EU also tend to feel more European than others.

THE VETO How does this influence views on national sovereignty? In order to analyse this we can replace feeling European with views on the veto. Figure 6.38 shows this relationship. Total right vs. veto 70 60 Per cent

50 Veto retained Veto essential areas Veto given up Don’t know

40 30 20 10 0 0

1

2

3

4

5

6

7

8

9

10

Number of correct answers given

Figure 6.38. Quiz results vs. feelings about the veto Source: Data compiled from EB61 dataset.

222

The Demos—Empirical Analysis 100

Per cent

80 60

Veto retained combined

40 20 0 0

1

2

3

4

5

6

7

8

9

10

Number of correct answers given

Figure 6.39. Quiz results vs. veto retained (combined categories) Source: Data compiled from EB61 dataset.

At the lower end of the scale, it appears that the more correct answers a respondent gives, the more he or she is inclined to think that the veto should be ‘retained to preserve the national interest’. However, for those who gave six correct answers or more there is a fall in the number likely to have this view. This fall is accompanied by an increase in the number inclined to believe that the veto should be retained in essential matters. As regards whether to give up the veto for the sake of efficiency, this number is relatively stable from two correct answers and upwards. There appears to be a relationship between questions answered correctly and the belief that the veto should be retained in essential areas. Hence, it cannot be concluded that the better a person did in the quiz, the more he or she was inclined to believe the veto should be given up. This trend only exists between 0 and 2. Figure 6.39 shows the accumulated categories ‘retain veto’ and ‘retain veto in essential areas’. The relationship is that the more a person knows about the union, the more he or she is inclined to believe that the veto should be retained. However, from six questions onwards this increase is almost zero. The quiz does not contain any questions about the council, although we can test the following hypothesis: People who have heard about the Council are more likely to favour giving up the veto than people who have not heard about the Council. The result is shown in Table 6.12. Of those who have heard about the Council 11.2 per cent believe the veto should be given up whereas of those who have not heard about 223

The Demos—Empirical Analysis Table 6.12. Heard of the Council of Ministers vs. maintain veto

Heard about Not heard about Don’t know

Be retained in order to preserve national interest (%)

Be limited to a very few essential areas (%)

Be given up for all decisions to make the EU more efficient (%)

Don’t know (%)

53.0 43.7 37.5

23.4 17.9 14.2

11.2 10.9 6.0

12.5 27.4 42.3

the Council this figure is 10.9 per cent. So at face value the hypothesis is confirmed, however, due to the small difference in the numbers it would probably be fairest to say that this does not make much of a difference. The other thing that could confirm this is that there are more people in the group who claim to have heard about the Council (53%) who believe that the veto should be retained to preserve national interest than in the group who say they have not heard about the Council (43.7%) (Table 6.12). It does not appear to make much difference whether respondents trust the council or not (Table 6.13). Of those who tend to trust the Council 52.4 per cent believe the veto should be retained, whereas the figure for those who tend not to trust the Council is 52.7 per cent. The scores for ‘giving up the veto’ are identical for the two groups. In the other two categories, there are differences. Those who do not trust the Council are less likely to express an opinion and are also less likely to believe that the veto should be limited to essential areas. Furthermore, knowledge about the Union also affects people’s levels of satisfaction with EU democracy. Figure 6.40 shows the result of cross tabulating the number of correct answers given per person with their declared level of satisfaction with EU democracy. The general pattern is that the level of satisfaction with EU democracy increases with the number of correct answers given in the quiz. Of those who gave a total right of 0 answers, 3.4 per cent were very satisfied with Table 6.13. Trust in the Council of Ministers vs. maintain veto

Tend to trust Tend not to trust Don’t know

224

Be retained in order to preserve national interest (%)

Be limited to a very few essential areas (%)

Be given up for all decisions to make the EU more efficient (%)

Don’t know (%)

52.4 52.7 43.1

25.7 20.8 15.8

11.7 11.7 8.8

10.3 14.8 32.2

The Demos—Empirical Analysis Total right vs. democracy satisfaction EU 100

Per cent

80 60

Very satisfied Fairly satisfied

40

Not very satisfied Not at all satisfied Don’t know

20 0 0

1

2

3

4

5

6

7

8

10

9

Number of correct answers given 100

Per cent

80 60 Satisfied Not satisfied

40 20 0 0

1

2

3

4

5

6

7

8

9

10

Number of correct answers given

Figure 6.40. Satisfaction with democracy in the EU

EU democracy, and for those who gave 10 correct answers this applied to 11.5 per cent. Those who gave 0 correct answers were also the most likely to say that they did not know whether they were satisfied with EU democracy. In the second figure both the positive and negative categories have been accumulated and this graph shows the same tendency. According to these figures we can conclude that there is a positive relationship between a person’s level of knowledge of the Union and that person’s satisfaction with EU democracy. For comparisons the figures for the same sample are illustrated in Figure 6.41 with respect to national democracies. The relationship is rather similar in the sense that the more questions a respondent answered correctly the more satisfied is he or she with national democracy. This level of satisfaction is higher than the levels 225

The Demos—Empirical Analysis Total right vs. democracy satisfaction country 100

Per cent

80

Very satisfied Fairly satisfied Not very satisfied Not at all satisfied Don’t know

60 40 20 0 0

1

2

3

4

5

6

7

8

9

10

Number of correct answers given

Figure 6.41. Satisfaction with democracy (own country)

of satisfaction with democracy in the EU for each group and with respect to both ‘very satisfied’ and ‘fairly satisfied’ (Figure 6.42). Total right vs. trust in the EU 100

Per cent

80 60

Tend to trust Tend not to trust Don’t know

40 20 0 0

1

2

3

4

5

6

7

8

9

10

Number of correct answers given

Figure 6.42. Quiz results vs. trust in the EU

Again, there is a clear pattern. The more successful in the quiz, the more respondents are likely to trust the Council. SUMMARY OF INDICATOR FOUR It will not be denied that it may be possible to set a standard for what would be considered an appropriate level of knowledge about a political system amongst the groups of people who live in it. However, it is likely 226

The Demos—Empirical Analysis

that the parameters of such a model of standards would be arbitrary or at least that it would be hard to agree on what constituted an appropriate level of knowledge. It is also likely that this would vary on the question at hand. However, as a minimum, one would assume that the citizens of a democracy knew the rules of the game of that democracy. Also, it may be reasonable to expect the citizens of a polity to be able to identify their ‘co-citizens’. When asked in 2006 whether the members of the European Parliament were directly elected by the citizens of Europe less than half the respondents gave the correct answer. Fewer knew that the EU had twenty-five member states (at the time). As regards the popularity of the Union (as in support for integration, satisfaction with democracy, trust in the Union, and the level of identification) there are clear indications of a relationship between lack of knowledge and lack of positive feelings.

6.3. Conclusion People feel much stronger attachment to their country than to the European Union. Similarly, most people identify themselves primarily with their nation state and secondarily in terms of their nationality and as European. These results are perhaps not surprising taking into consideration conclusions made on the basis of the literature review and analysis in the preceding chapters. However, the fact that continued membership of the European Union does not seem to make people feel more European shows that the assumption that institutional and legal integration will eventually lead to civic or social integration may be flawed. Furthermore, Europeans are more likely to express concern for their fellow human beings rather than specifically other Europeans. This finding confirms the assumption made earlier that it may be difficult to create a European demos because transnational solidarity appears to have a human base. There are, however, signs that that younger generations are likely to feel more positive about the Union, perhaps indicating that views are likely to become increasingly positive in future. (Unless, people change their minds as they get older.) The evidence shows that Europeans agree that the EU is improving their chances of studying and working abroad. However, they rate their individual transnational ‘rights’ higher than their transnational democratic opportunities, again giving rise to doubts that European civic integration is a priority. 227

The Demos—Empirical Analysis

Arguably, the most important finding is that Europeans do not really believe that they share a distinct cultural identity. This is particularly serious, because the EU itself claims to be building on such an identity even though claims to that effect have been toned down since the Maastricht ratification crisis. For instance Article I, 5.1 (http://europa. eu/constitution/en/ptoc2_en.htm) of the constitution states that ‘The Union shall respect the equality of the member states before the constitution as well as their national identities. . . .’ (C 310/40). This is a different language to that of the TEU articles 2 and 27a, which both refer to the Union’s identity and in which the preamble includes a paragraph that reads that the heads of state have resolved to ‘implement a common foreign and security policy including the progressive framing of a common defence policy, which might lead to a common defence in accordance with the provisions of article 17, thereby reinforcing the European identity and its independence in order to promote peace (C 325/9) (italics added). A similar reference is made in the White Paper (COM(2001) 428, 127). In addition to the fact that these references show a formal reference to the concept of European identity (Mayer and Palmowski 2004: 576) one should also note the language, that is, the fact that the Union is the subject in the sentences, indicating its existence as a personified entity. The potential compensation offered by ‘public awareness of the transnational process’ is at best sporadic. If people had taken a great interest in the technicalities of the European Union and its decision-making processes, it could perhaps have been valid to claim that this display of interest served as a sign that they were keen to build a common future and thus took particular interest in the polity. This is not the case. Although there is a high level of support for transnational cooperation, over half the EU citizens would not be particularly sorry if the EU were to be dissolved. On the positive side, there is clearly a strong commitment to democratic values and very substantial support for democracy as a political system. As regards other values there are some divisions but often in areas where similar levels of division can be seen within the national context. It could perhaps be argued that it would have been useful to compare the civic feeling in a number of nation states. Such an analysis could perhaps have provided a comparative basis for measuring whether the lack of a demos is a situation that a number of different societies suffer from, thus perhaps reducing the impact that the lack of the demos could have on the EU. It will not be contested that such comparisons could have been both interesting and enlightening. However, two conclusions drawn in 228

The Demos—Empirical Analysis

earlier chapters serve to reduce the impact of this criticism. First of all, the four indicators and derived hypotheses deal almost exclusively with the subjective–emotional elements of the demos (as established in Chapter 3). This means that a weak outcome is not compensated for by, for instance, the existence of a common media or a common language (which, despite two known cases to the contrary most European countries have). Indeed, as regards the—very few—ethno-cultural variables described here, those that dealt with the question of whether people believed that Europe has a common cultural identity came out with a low score. The findings appear to support the theoretical argument made earlier, that the EU does not have a demos (as defined in earlier chapters) and that despite strongly shared democratic values this does not seem to be a situation which is likely to change in the near future. In Chapter 7 the findings are discussed in the light of the argument developed in the theoretical chapters.

Notes 1. As a matter of caution it is worth noting the ‘in the near future’. The problem with this reference to time is that intuitively it could perhaps make respondents believe that there is about to be a change in their identity. To that extent the question could be interpreted as slightly manipulative. Particularly with a view to the fact that if the authors of the survey were trying to measure the status quo, the reference to the future should have been left out. 2. The ˜ 2 value of 3,276.2 is highly significant with a p value of less than .001. 3. The ˜ 2 test was significant at the 5% level (comparing the ˜ 2 statistic of 6.068 with a ˜ 2 distribution of 2 degrees of freedom gives a p value of .048). 4. Paired t test. 5. A similar finding has been reported by Haubich who analysed the identifications of 6,000 young people from 21 European countries although the data-sets are not directly comparable to the one European Value Survey as Haubrich’s study included only the 18–22-year olds (see Haubrich 1997). 6. The variable should be treated with some caution. It is more than likely that respondents would be concerned that the EU contribution would be an additional tax rather than a different distribution system. This could perhaps have made some respondents reluctant to reply in the affirmative. 7. The data is distributed by Zentralarchiv für Empirische Sozialforschung an der Universität zu Köln. Outside Germany the data can be obtained via national data archives, in the United Kingdom the UK data archive. Countries surveyed in the 1999–2000 wave are Austria, Belarus, Belgium, Bulgaria, Croatia, the Czech Republic, Denmark, Estonia, Finland, France, Germany, Great Britain, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta,

229

The Demos—Empirical Analysis the Netherlands, Northern Ireland, Poland, Portugal, Romania, Russia, Slovakia, Slovenia, Spain, Sweden, Turkey, and the Ukraine. 8. The justification for adding the ‘bad thing’, ‘neither good nor bad’ and ‘don’t know’ responses is that although ‘neither good nor bad’ and ‘don’t know’ are not anti-support (because then the respondents would presumably have said ‘a bad thing’) they are not statements of support. 9. Other studies have found that respondents give very varying replies to questions about the veto depending on the wording of the actual question (see for instance, Saris, 1997).

230

7 Conclusions

7.1. The EU and Democracy The EU presents analysts with a dilemma with regards to the future of European democracy. The preceding chapters have focused on the problems, thus it may be appropriate to begin this chapter by summarizing the positive aspects:

r

r

The EU has increased the collective acting-capability of the member states in a range of areas. To that extent the EU is clearly greater than the sum of its parts. This in itself does not entail an increase in democracy but when it comes to issues such as environmental matters, the ‘pooling’ of sovereignty benefits democracy because even if one were to argue that a national decision on, say, CO2 emissions was democratic, the decision might have very little impact because the effects of pollution are likely to be felt beyond the polluter’s borders. And even though not all countries are members of the EU, collectively the EU has more muscle to place pressure on third parties when it comes to environmental affairs although, of course, the EU cannot dictate policy to other states as the controversies over the Kyoto Protocol serves to show. Furthermore—using the model of Alesina and Spolaore—the economies of scale are high and at the same time the heterogeneity of the preferences of EU citizens in this policy area are also high. Secondly, a large number of scholars credit the Union with the democratic stability of those member states that have only a very short history of democracy and whose transition to democracy was strongly linked to their joining the Union. These states include Greece, Spain,

231

Conclusions

r

and Portugal. As regards the 2004 accession states, it is clear that one of the aims of including these states is the hope that their membership will help sustain democracy in what continues to be relatively fragile political systems. The third major way in which the EU can be said to contribute to democracy in Europe is perhaps more controversial. But if one is prepared to credit the Union with the peace amongst EU members which Europe has seen since 1945 (subsequent European wars have, after all, been amongst non-members), then the EU has contributed to the continued democratic existence of the member states.1 On the other hand, when the European Commission generously credits the EU with the avoidance of war in Europe since 1945, this is a moderated truth. First of all, there have been wars in Europe—although as mentioned not amongst EU member states—and secondly, it is more than doubtful whether the EU can be persuasively credited with the absence of wars amongst the member states. Such an argument overlooks the possible effects of nuclear deterrence and the presence—until the early 1990s—of an identifiable external enemy. Nor does it take account of the strong US presence in Europe and the overlay resulting from bipolarity during the Cold War. Interestingly, the end of ‘permissive consensus’ collided with the demise of the Eastern bloc and the fall of the Soviet Union, possibly indicating that the external security situation played a strong part in motivating public support for European integration. But whatever the reason, the fact that the EU region has been stable and prosperous is likely to have prevented the emergence of forceful anti-democratic tendencies and hence may have contributed to democracy by keeping people contented and thereby unlikely to revolt by becoming attracted to fringe politicians who offer easy solutions to complex problems.

However, despite these achievements, the EU also presents a potential cost to democracy. The discussion about the EU’s democratic deficit(s) is extremely complex. Hence, I have divided the lenses of the literature into four paradigms, thereby providing a system in which the different issues are categorized. In essence, I am arguing that the research into democracy in the EU is still in a state that Thomas Kuhn would have described as

232

Conclusions

pre-paradigmatic. This is not an unnatural condition but merely a result of the fact that the pan-European democracy issue has not been debated for a particularly long time. The intensive debate about democracy in the EU accelerated in connection with the crisis that occurred as a result of the events surrounding the ratification of the Maastricht Treaty. Since then, the literature—both academic and popular—has grown enormously and many recommendations have been made although no broad consensus has been reached. In fact, although most scholars and practitioners believe that the EU is in need of reform, not least in order to accommodate institutionally the most recent members, there is no consensus in academia as to what should be the substance of such reforms. Aside from the normative and ideological reasons for this lack of consensus, which must be assumed to exist in any debate about reform, I have argued that there are specific reasons for the almost scattered debate about EU reform, that is, that scholars use a range of different ‘lenses’ when analysing the EU polity which determine what problems they identify. A systematic analysis of the debate was carried out in Chapter 2, which could have resulted in a range of different categorizations of arguments. For instance, one could have categorized scholars in groups in accordance with their conclusions as to whether the EU had a democratic deficit or not. But this would not have brought useful insights into the debate because such categorizations would be bound to hide the fact that scholars very often reach similar conclusions for different reasons. The reason this matters is that if a number of scholars all conclude that the EU lacks democracy, it is (at least partly) their reasons for reaching this conclusion that determines what remedies they suggest to improve the situation. Hence, I chose to categorize the arguments according to their foci and the parameters and relationships with which each was preoccupied. I named the categories ‘paradigms’ because of their resemblance to the sociological understanding of ‘paradigms’ as defined by Thomas Kuhn, thereby creating a system where all arguments have been analysed in context, that is, held up against a directly opposing argument rather than—as is often the case in the EU debate—against a completely different relationship. Although in one sense the paradigm is used to describe the debates (as four subdebates), the paradigms really are sets of analytical tools and interests that scholars begin with when forming their research questions—in the words of Kuhn, the ‘scientific matrix’. Although this book has used the socio-psychological paradigm, each contributes with important insights into the problem.

233

Conclusions

7.1.1. The Four Paradigms The first paradigm is the efficiency paradigm. This concentrates on the relationship between efficiency and democracy. According to one school of thought there will always be a trade-off between the two, hence, the debate about democracy is really a debate about finding the balance between them (see Wessels 1996). In practice, one could use the Council of Ministers to understand this argument. In a situation where decisions are made unanimously it may be very difficult to reach a decision or, alternatively, decisions may be technically poor due to the need for compromise. This is the situation Scharpf describes as the ‘joint decision trap’. The argument also works at the procedural level, that is, nobody would probably be prepared to suggest that decisions in the EU should be made by consensus so that all citizens had to come to agreement. Hence, the idea of representation is in itself a result of a trade-off because, although in an ideal democracy all decisions might be reached by consensus, this is impossible in practice. So, citizens trade ideal democracy for less ideal democracy with some level of acting-capability. Seen in this light, the efficiency/democracy trade-off relates to the democratic process rather than policy substance. This idea of a trade-off stems from a strain of democratic theory, which argues that legitimacy consists of a certain amount of democracy and a certain amount of efficiency. But it is perfectly possible to agree with this statement without subscribing to the idea that the two are interrelated in the sense that adding to one implies subtracting from the other and vice versa. In concrete terms, it is quite possible to imagine a polity in which government is neither very efficient nor very democratic.2 Similarly, a polity can also be both very democratic and very efficient. But even in a polity where the intentions of leaders are genuinely positive and there is a real wish to provide good government for the people, one should not overestimate the value of efficiency. There is a very fine—although by no means firm—line between democracy and guardianship. Thus, the danger in the efficiency argument is that taken to the extreme it can become an advocacy for guardianship. But if we accept the assumption that democracy is superior to its alternatives, it follows that we must not overestimate the value of policy output.3 Indeed, from a liberal perspective, efficiency cannot make up for lost democracy because it is precisely the democratic nature of the polity that grants governments the authority to act on behalf of the citizens, i.e. make policy. The reason that the relationship between efficiency and democracy is so hard to settle is that the assumption that democracy is preferable to other systems

234

Conclusions

confuses the analysis. The fact that the lens I am using is a concern with democracy means that I am already intuitively inclined towards having more concern for that particular issue. Therefore, it is necessary to take a step back and instead of asking ‘why do we want democracy?’ ask ‘why do we want government?’. The reason is well known. That individuals prefer to exchange the impossible conditions inherent in the state of nature for a certain amount of security. They trade personal autonomy for collective self-government under the condition that everybody else does the same. The result is the social contract between individuals, which authorises the institution of government. As has been argued throughout this book, this situation requires a level of recognized similarity between the individuals involved in the contract. In this understanding government and democracy go hand in hand. If we add to this rationale the assumption that most people intuitively understand this agreement, the scepticism towards the EU may be interpreted more as a scepticism towards what is seen as an additional layer of government. (Naming it governance will not change such perceptions.) Another development which could point in this direction is the fact that the same reluctance can occasionally be found with respect to sub-national systems. For instance, the process of devolution in the United Kingdom suffered at setback when the proposed establishment of a North East Assembly was rejected by the voters of the region. In this case, voters rejected the notion of a ‘closer democracy’ although it is clear that some policy areas could perhaps (following the rationale behind the model of multilevel governance, Hooghe, et al. 2006) have been better managed at the regional level. This problem is one of the elements in the discussion of the vertical democracy paradigm which concentrates on the distribution of powers between vertical rather than horizontal levels of government. This paradigm sees the EU as part of a large political system with different levels and debates. The research question here is how, and if the EU has affected democracy in nation states. On the one hand, perhaps the EU strengthens democracy because the pooling of sovereignty enables the EU-level to act collectively with more effective results than could have been achieved at national level. Endless amounts of internal freedom and democracy will, after all, be of no consequence if the political circumstances are so heavily influenced by exogenous factors that the polity has effectively lost autonomy (Dahl and Tufte 1973: 128). As mentioned, environmental policy is a good example. Others include competition and the fight against organized crime and international terrorism. The 235

Conclusions

conflicting argument under this paradigm is that the EU has in fact caused great damage to national democracies because of lack of transparency in the Council. Seen in this light, national executives experience an increase of their acting-capabilities without an increase in their mandate, the result being an erosion of democracy. Reforms have been implemented to reduce this gap, such as the direct election of MEPs and the subsequent introduction and increased expansion of co-decision. Such measures then affect the horizontal division of powers because they increase the powers of the EP relatively to the Council. And the relative distribution of powers between the EU institutions has—until very recently—been the dominant issue in the EU debate on democracy. I have named this lens the horizontal democracy paradigm because it focuses on the relationship between EU level institutions, both supranational, intergovernmental, and mixed. The common characteristic of arguments under this debate is an analogy to national systems, often Germany or the United States, but also other domestic political systems. The idea is to create a balance of powers to ensure that no institution becomes too dominant and this paradigm relies much—probably too much—on the nation state analogy. Most recently the developments have been in the direction of increased powers to the EP whereas the Commission has lost powers relatively to both the Council and the EP. Despite addressing these important issues, the trouble with using the nation state analogy is that the EU is not an all-encompassing political system. Thus, the weakness of the horizontal democracy paradigm is not its focus on the relative distribution of powers between EU institutions (which is a perfectly sensible thing to worry about), but the fact that when focussing too much on these relationships it becomes very easy to overlook other more fundamental ones. Unlike a national system, which is likely to be more stable if checks and balances on institutions are in place, the presence of such checks in the EU system would not alone ensure its stability. In that sense the trouble with the horizontal democracy paradigm is not what it includes, but what it does not include. Although the above discussions are important, this book has argued that the most fundamental of the paradigms is the socio-psychological paradigm. Scholars who use this ‘lens’ point to the fact that the lack of an EU demos must be addressed so that appropriate channels of representation can be identified. Even scholars who do not believe that the EU has a democratic deficit agree that there is no strong European identity and no EU demos. The absence of such a demos means that there are limits to the freedom that reformers of the Union will have in 236

Conclusions

designing, or redesigning, the system, even in the presence of utilitarian support.

7.2. The Nature of the EU Demos It has also been shown that a demos need not be understood in a strictly ethno-cultural sense. Whilst such a demos is perhaps not impossible to construct, doing so is a very difficult enterprise. The means available to European polity builders do not allow coercion, and in any case, the risk that an attempt to manipulate identities will be unsuccessful is great. Many scholars argue that creating a European demos based on a European cultural identity will be risky because such an identity would necessitate an ‘other’, creating scope for a new European nationalism. Such fears are expressed by Weiler, Smith, and to an extent, Habermas, the latter being concerned about the creation of a ‘fortress Europe’.4 Hence these deal with the fears of the relationship between Europeans and ‘outsiders’. However, the other side of the coin is the reaction likely to be seen within, that is, the tendency of cultures to defend themselves, as pointed out by Wæver. Tendencies towards such developments have been evident in some European countries where fringe parties resisting EU membership attract voters by referring to old national symbols and ideals. Finally, the developments that were expected to result as a form of social spillover (as expressed by Taylor) never took place. Although many Europeans accept European integration (albeit to varying degrees), a European identity has not materialized as a result. There are no real signs that this is changing. When asked, only 7 per cent of the sample in the Eurobarometer survey said they completely agreed that there exists a distinct European cultural identity. The EU does not have a demos in the sense that most stable democratic nation states do. However, Europeans are in principle supportive of transnational cooperation and those who are not are more likely to be indifferent than downright hostile. The evidence also showed a high degree of instrumentalism in support for the Union, with people likely to see the EU as providing increased transnational individual opportunities such as the chance to study and work abroad. The empirical evidence suggests a high degree of shared democratic values but no particular European solidarity. It is clear that, despite a strong desire for cooperation, Europeans do not see themselves a part of one democratic whole, but rather constituent units cooperating within 237

Conclusions

an institutional framework called the EU, a framework which many do not consider indispensable. Seen in the light of the conceptualized demos this means that the subjective-emotional dimension of the demos is weak but that the idea of a European procedural political identity is not entirely utopian. Again, the consequence for integration is that the areas that demand a high degree of solidarity should not be subjected to supranational government. To that extent, the idea of multi-level policymaking (if this means that different decisions are made at different levels) is a sensible one, not least from a normative perspective. Furthermore, the shared commitment to democracy also potentially hinders the building of the European polity because the fact that public commitment to democracy is so strong, taken together with the fact that the EU itself cannot be a democracy (for the reasons stated throughout this book), means that democratic legitimacy has to be embedded in the member states. It also means that the complexity and lack of transparency inherent in the EU system must be reduced. This is best done by maintaining, in very strong terms, that the nation states are still the primary polities, with local and supranational polities as secondary, and only in areas where the pragmatic argument for delegating to the supranational level from the state is convincing.

7.3. Models of Democracy It was concluded that although no model can serve as a perfect description of the EU, all the eight models analysed in Chapter 4 have contributions to make, each at different levels of analysis. Although at present the EU looks neither like a presidential nor a parliamentary system, it is—even without the direct election of a president—more presidential than it is parliamentary (if we regard the president of the Commission as the EU’s chief executive). As long as the EP does not form the EU executive, this will continue to be the case. As regards the future, many commentators have advocated the direct election of the Commission president, notably Hix and Decker. If the president of the Commission can be considered the chief executive of the EU, such a change would be a major step towards an increasingly presidential model, because the chief executive would have a fixed electoral mandate, which would—technically—be his or her own source of legitimacy. Since the legislature, that is, the EP, also has a fixed electoral mandate of its own, the effective result would be a presidential model as described by Stepan and Skach (see 238

Conclusions

Chapter 4). However, this analysis leaves out the role of the Council. If, following Hix, we consider the Council part of a ‘dual executive’ then the Commission president may be the chief executive on paper, but only so. If instead, the Commission president were to be elected by the EP (creating a link between the legislature and the executive and hence an indirect link between the electorate and the executive), the EU would be closer to a parliamentary model because it would satisfy the requirement that the chief executive power was supported by a majority in the legislature. The set-up would only be parliamentary if it also satisfied the other conditions, that is, that the president could subsequently be subjected to a vote of no confidence, and he or she had the capacity to dissolve the legislature and call new elections. Again, this analysis leaves out the Council. Hence, what the parliamentary and the presidential models have in common, when applied to the EU, is that neither provides an adequate analytical framework that accommodates the Council and the European Council. The president of the Commission is not a chief executive in the same way as, for instance, a British prime minister or an American president. Hence, for understanding the role of the Council and the European Council, both the parliamentary and the presidential models are inappropriate. For understanding the roles of these institutions, the more appropriate model could appear to be the confederal model, particularly when it comes to agreeing the broader terms and scopes of the Union. On the other hand, the fact that part of these overall strategies is to ‘overcome’ confederal limitations obviously means that this model can only be applied at the micro level, that is, to analyse interstate bargaining. Instead, the federal model can be seen as more suitable as a descriptive model because of the power-sharing between the two levels of government. Despite the fact that the EU has neither federal taxation nor a supranational power to declare war on third parties, the EU looks like a federation and hence it is no surprise that so many scholars have decided to use other federal states as comparisons. Related to federalism is the notion of consociational democracy. Again, the model takes into account the institutions that are hard to accommodate in the frameworks offered by the parliamentary and presidential models in the sense that the Council and European Council can be seen as elite representatives. However, the weaknesses are that it may not be appropriate to treat states as ‘segments’ particularly because— unlike segments in known consociational systems—EU member states have all enjoyed formal sovereignty prior to joining the Union. This is a major difference which should not be taken lightly, particularly because 239

Conclusions

citizens of these segments still identify very strongly with their ‘primary’ polity. The neomedieval empire analogy is an innovative description of the situation created by the fact that acting capabilities have been transferred to EU level in a range of policy areas and its advocates make convincing arguments why the EU could not and should not become a state. The network model contributes to the study of the EU by pointing to the involvement of non-governmental actors and thus provides a range of insights into processes that would simply not be captured by more traditional lenses. Its strength is in that it captures these relationships and identifies areas which are problematic for those concerned with representative democracy. And finally, the two-tier model is one of the only models which highlights the fact that EU level policymaking does not include all members in all areas. Each model provides valuable insights into the nature and functioning of the EU. And regardless of one’s chosen paradigm or normative stance on European integration and institutional matters, one can benefit from applying the analytical tools provided by the models and increase conceptual clarity. However, each of the models has a ‘following’ of one or more authors who sees it (or a version of it) as an appropriate way to solve the EU’s legitimacy crisis. Models of democracy are ideal types, which do not exist empirically in their purest forms. But even at the theoretical level, no abstract model can serve as a blueprint for the EU, although this does not mean that these are not useful when it comes to bringing to light ideas for improvement and reform. The choice one makes will—to a very large extent—depend on (a) what paradigm one belongs to and (b) what argument one takes under that specific paradigm. In this book, the paradigm has been the socio-psychological paradigm and the viewpoint has been the particular nature of the EU demos. The nature of the demos (its non-existence as a social phenomenon, the lack of pan-European solidarity, as well as the low level of interest in transnational democratic participation) can only lead one to conclude that, certainly in the short term, the Union should continue to be legitimised via its parts, that is, the nation states. Not only do Europeans have strong ties to their country, they also have significantly stronger ties to their fellow countrymen than to other EU citizens. For these reasons, neither a parliamentary system, nor a presidential system is desirable for the EU. Both of these systems require a level of solidarity that is not present in the EU. Contrary to Hix, I do not believe that the direct election of a president of the Commission will do anything to 240

Conclusions

provide increased legitimacy of his or her role. It is easy to overestimate the value of direct elections. The same problem faces the federal model, if (as argued by Weiler) a federal system requires a constituent demos. This leaves the confederal model which, seen from the socio-psychological perspective, is the most fitting, the problem being that it effectively means a setback to the integration process and hence would not be considered an alternative by federalist reformers. As regards the network model, it was concluded that this solution, advocated so strongly by Leonard, would not solve the socio-psychological deficit. The network model is interesting in the sense that it offers an alternative to the majoritarian principle, the appropriateness of which is questionable at EU level. However, from the same perspective it is questionable whether the network model is an appropriate alternative. At most, networks can serve as additional links between governed and governing, but not as alternatives to representative institutions that must take into consideration the nature of the demos. In fact, one could argue that the presence of networks is a major contributing factor to the perceived democratic deficit of the Union due to the inherent lack of transparency and limited access. Finally, the two-tier model (which could be seen as a system of flexible multi-level policymaking) is perhaps more appropriate. It takes into consideration that support for a policy can be hard to achieve in specific countries. However, if a demos is to develop it may be harmful that EU citizens are not all part of the exact same political and institutional framework. From a normative perspective, the strength of the neomedieval model lies in its insistence that democracy beyond the nation state level should not rely on majoritarian institutions. On the other hand, its weakness lies in its lack of concrete suggestions as to how a neomedieval democracy could be built. It proposes a range of solutions but at a higher level of abstraction than what reformers are seeking. Conversely, it provides direction which those potentially responsible for proposing new legal structures could take into consideration. The question, however, is whether an increasingly puzzled European citizenry is prepared to accept such innovative measures. This is where there is a potential clash between addressing two problems, the no demos issue and the fact that a treaty change would almost certainly be subjected to referendums in a number of states where it would be difficult to ratify. Unlike hitherto seen consociational systems, the EU (to the extent that it is a consociation) is intuitively and legally associated with the notion of freedom of exit (as in peaceful freedom of exit). This underlines the 241

Conclusions

voluntary history of the cooperation. The second operational criticism of the consociational model is that it is likely to reproduce the segments that it was designed to overcome. However, was the consociational model really designed to overcome the segments? That may be the critics overestimating the scope and intention of the consociational model and in essence depends on the analyst’s understanding of the word ‘overcome’. If one takes this to mean that eventually the system eradicates the meaning of the segments, perhaps even the segments, per se, the criticism is valid. As mentioned earlier, the consociational model is very likely to keep reproducing fragments. However, if one understands ‘overcome’ in the ‘work around’ sense, this need not be a problem. And despite disagreements in the literature, one issue that almost enjoys an element of consensus is the issue that it would be at worst risky and at best probably useless to try to manipulate European identities, so the fact that the segments are reproduced is not a problem at all. It may be so if one is a federalist in the ‘pure’ (as in not consociational sense) but if not, the EU polity can live with its segments and probably be better off with them than without them. Seen in this light, the durability of the segments poses no problem. The question then becomes whether the reproduction of the segments also means that there will be no emerging European demos. This will probably be the case, but developing a European ethno-cultural demos is—as argued—not likely to be possible, but even if it were, not particularly desirable either, a point argued convincingly by Joseph Weiler (see Chapter 3). Hence, what such developments would necessitate is a strong utilitarian support (which again is conditioned upon the fact that people have to see the EU as an institution that actually contributes to their well-being). This is when our more meticulous definition of the demos becomes a useful tool. Because if a consociational model could be constructed which, despite maintaining the ethno-cultural identities (which could not be changed in the short term in any case), contributes to the development of partly a utilitarian support as well as an increase in the subjective– emotional side of the demos, this would probably be the ideal scenario. However, so far there are no signs of a particular transnational European solidarity developing although there are many commonly held values. Maybe in the long run the utilitarian element could lead to increased common subjective–emotional identity, that is, transnational solidarity without the ethno-cultural elements. However, in the meantime, the foundation for the cooperation must be a division of labour between the national and EU levels of government. 242

Conclusions

In addition, it is clear that support for policymaking at the EU level must rest on support for the process rather than substance because different groups cannot always rely on achieving their desired policy outcome. However, if citizens respect the overall structures of authority (as described in Chapter 2), the perceived illegitimacy that may arise when they are disappointed with respect to direction of policy is likely to be reduced. The fact that the EU does not have ‘a people’ means that the EP cannot be considered the primary source of the polity’s legitimacy. Such statements rely on inappropriate analogies to nation states. However, this does not have to lead one to the conclusion that the EP is an illegitimate institution. The extent to which this is the case depends in which policy areas the EP is involved. Hence, the solution will have to be that the EP maintains its involvement in those policy areas which do not require pan-European solidarity and identity. However, because EU citizens share certain values there are policy areas in which supranational government (or governance) can be regarded as legitimate even with the weak or nonexistent demos. Following the logic of Majone, this means that redistributive policies should remain within the realms of the member states. This is, in fact, already the case with very few exceptions. Policies, however, which do not require high levels of transnational solidarity (i.e. within the Union) can remain within the realm of the EU’s supranational structures and derive legitimacy from their ability to produce output.

7.4. The Future of the EU One is often presented with the argument that it is unreasonable (and undemocratic) that the voters of a few, or one, state can hold back the integration process, and hence ‘unfair’ that the French and the Dutch voters can stall a constitution which has been ratified by many other states. Aside from the fact that such are the rules of the game, this is a poor and thoughtless argument. Had it not been for the occasional referendum which produces an unwelcome result, I am convinced that the current important debates about democracy would not have materialized anywhere other than on the fringe of the literature and amongst diehard eurosceptics. Hence, such results should be seen as blessings. After the rejections of the Constitution in Holland and France, many EU statesmen expressed the view that the ratification process should continue. Hence the European Commission posted the following message on its website: 243

Conclusions The people of France and the Netherlands rejected the text of the Constitution on 29 May and 1 June respectively. In the light of these results, the European Council, meeting on 16 and 17 June 2005, considered that we do not feel that the date initially planned for a report on ratification of the Treaty, 1 November 2006, is still tenable, since those countries which have not yet ratified the Treaty will be unable to furnish a clear reply before mid 2007. A period of reflection, explanation and discussion is currently under way in all countries, whether they have ratified the Constitution. The state of discussions on ratification of the Constitutional Treaty will be examined by the European Council under the Austrian Presidency (in the first half of 2006). The process of ratification by the member states has therefore not been abandoned. If necessary, the timetable will be adjusted to reflect the circumstances in the countries which have not yet ratified the treaty. (European Commission Website.)

This message can be found to this day (accessed 30 August 2006) which must mean that ‘no’ does not mean ‘no’ to the European Commission. The eventual ratification of the constitution is simply not questioned. References to the ‘timetable’ are evidence of this. Instead, the Commission is making increased effort at communicating with EU citizens. Although the Commission still appears rather attached to the notion of the institutionalized ‘civil society’, new avenues are being pursued which seem to demonstrate a clearer understanding that dialogue with individuals is also needed.5 Such initiatives are, of course, welcome. However, language used in leaflets is at best humorous (unintentionally), and at worst, patronising, and one is often left with the impression that whilst there is a desire to educate, there is also a desire to manipulate. This is confirmed by the often ludicrous literature posted on the website, such as the story of the Raspberry Ice Cream Wars, which is for children but nevertheless deeply patronising, and the bizarre proposals to mark the Unions fiftieth anniversary with a pan-European singing contest, possibly accompanied by a cake-baking competition. It may be true that the European publics are largely uneducated about the technicalities of European integration, but to assume that their views will be positively influenced by such absurd initiatives is rather naïve and does give associations to propaganda ‘celebrations’ occasionally seen in authoritarian nations. It also constitutes a striking example of the gap which exists between ‘the EU elite’ and the European publics. A majority of EU citizens are supportive of a European constitution. But the evidence from opinion polls also shows that a large majority do not believe the current constitution should be adopted. This, together with 244

Conclusions

the fact that the French and Dutch referendum results gave the EU muchneeded publicity, presents the EU institutions with a welcome opportunity to address both the no-demos issue and the failed constitution. It would indeed, as argued by The Economist, be preposterous to ask the Dutch and French electorates to vote again on the same constitution (25 May 2006). Instead, the European Council must go ‘back to the drawing board’ and negotiate a realistic document which takes into consideration the lessons learnt as well as the socio-psychological deficit. Of course, input can be sought from experts and think-tanks but it is important not to rely solely on the usual groups of idealist supporters. In order for such exercises to be legitimate, those with different views should also be consulted, such as the more eurosceptic or euro-‘critical’ organizations and individuals. This would give a more balanced input and increase credibility in the eyes of the publics. The fact that the EU system is not perfect does not mean that it cannot get worse. Trying to create institutions which themselves require the existence of a demos will not solve the EU’s legitimacy problems. The EU can rely on output legitimacy in those areas which do not require high levels of pan-European solidarity and identity. Policies which require high levels of solidarity and a common identity should remain within the member states, or in some cases, within the EU’s intergovernmental framework (such as pillar 2). In conclusion, redistributive policies must remain the responsibility of the primary polities (the states). Whilst being a satisfactory solution from the no-demos perspective, such a division of labour would also be very likely to satisfy those publics who are unconvinced that the EU provides a better space for settling political matters than the nation state. Whilst it may be disappointing for some that there is no such thing as a European political people, this a reality that has to be faced. The fact that Europe’s future is unlikely to include pan-European celebrations of Europe Day with teary-eyed citizens engaged in sing-songs and cake-baking does not mean that the EU is obsolete. It need not even be thought of as in a crisis.

Notes 1. Unless one counts Northern Ireland as a warzone which—during some periods of time—may not be entirely unreasonable. 2. This entails defining efficiency as providing good policy as in delivering welfare, security and generally positive policies for the people. If one were to define

245

Conclusions efficiency as the ability to keep order, the states that one would judge to be efficient might include states that would not qualify under the first definition. 3. There is, of course, another more shrewd idea behind the policy-delivery argument, i.e. that as long as people get what they want (bread and circus) they will not be too concerned with democratic processes. 4. Fortress Europe can be understood strictly in terms of trade and as such refers to the notion of protectionism but also in a broader sense, i.e. to denote a more general exclusion of outsiders. 5. See for instance http://europa.eu/debateeurope/index_en.htm

246

Bibliography

Primary Sources European Union Documents Commission of the European Communities (2001) European Governance. A White Paper, Brussels, COM(2001) 428 The European Convention, The Secretariat (2003) Consolidated report from the working party of experts of the Legal Services of March 2003—Volume I, CONV 618/03, Brussels European Council (2001) SN 300/1/01 REV, Presidency Conclusions, European Council Meeting in Laeken 14/15 December European Council (2006) 10633/1/06 REV 1, Presidency Conclusions, European Council Meeting in Brussels, 15/16 June 2006 Official Journal of the European Communities (2001), Notice No. 80/C01 Official Journal of the European Union (2004), Vol. 47, Notice No. 2004/C 310/01 Databases Eurobarometer 58.1 (2002). Eurobarometer 61.0 (2004). European Value Surveys (1999). Each database can be obtained by request from the UK Data Archive at http://www.data-archive.ac.uk Surveys Commission of the European Communities, Eurobarometer (EB) 42, EB43, EB46, EB47, EB49, EB50, EB52, EB54, EB55, EB56, EB57, EB58, EB59, EB60, EB61, EB62, EB64 and EB65. Special Eurobarometer Report 251. The reports and annexes can be downloaded from the Commission’s public opinion website http://ec.europa.eu/public_opinion/index_en.htm. Newspapers Der Spiegel, 18 September 2000

247

Bibliography The Economist, 10 October 2002, 11 October 2003, 14 February 2004, 13 March 2004, 4th May 2006, 25th May 2006 The Guardian, 14 July 2003 World Wide Web Resources http://news.bbc.co.uk/1/hi/world/europe/3032833.stm, accessed 23rd August 2006 http://news.bbc.co.uk/1/hi/world/europe/4615781.stm, News 6 June 2005 http://www.europarl.europa.eu/news/public/default_en.htm The official website of the European Parliament, last accessed 22nd March 2005 http://europa.eu/index_en.htm The European Union’s official website, last accessed 22nd March 2005 (http://europa.eu/youth/index.cfm?l_id=en). The European Union’s Youth website, last accessed 19 August 2006 http://www.eu-oplysningen.dk/euo_en/ The EU Information Centre of the Danish Folketing, last accessed 19 August 2004 http://europa.eu/abc/keyfigures/growing family/index_en.htm European Commission 2003

Secondary Sources Aberle, David F. (1950). ‘Shared Values in Complex Societies’, American Sociological Review, 15(4): 495–502. Abramson, Paul R. and Inglehart, Ronald (1995). Value Change in Global Perspective. Ann Arbor, MI: The University of Michigan Press. Agresti, Alan and Finlay, Barbara (1997). Statistical Methods for the Social Sciences, 3rd edn. New Jersey: Prentice-Hall. Alesina, Alberto and Spolaore, Enrico (2003). The Size of Nations. Cambridge, MA: The MIT Press. Andersen, Svein S. and Eliassen, Kjell A. (1996). ‘Democracy: Traditional Concerns in New Institutional Settings’, in Svein S. Andersen and Kjell A. Eliassen (eds.), The European Union. How Democratic Is It?. London: Sage. Anderson, James and Goodman, James (1999). ‘Transnationalism, Territorialities and Democracy in the EU’, Explorations in Sociology, 51: 17–34. Aristotle (1992). The Politics. London: Penguin Books. Ball, Terence (1976). ‘From Paradigms to Research Programs: Toward a Post-Kuhnian Political Science’, American Journal of Political Science, 20(1): 151–77. Bauböck, Rainer (1994). Transnational Citizenship: Membership and Rights in International Migration. Aldershot: Edward Elgar. Berlin, Isaiah (1958). Two Concepts of Liberty. An Inaugural Lecture Delivered before the University of Oxford. Oxford: Clarendon Press.

248

Bibliography Bogaards, Matthijs (1998). ‘The Favourable Factors for Consociational Democracy. A Review’, European Journal of Political Research, 33(4): 475–96. (2002). ‘The Consociational Analogy of the European Union. A Rejoinder to Crepaz with a Comment on Kaiser’, Forum Section, European Union Politics, 3(4): 501–3. and Crepaz, Markus M. L. (2002). ‘Consociational Interpretations of the European Union’, Forum Section, European Union Politics, 3(3): 357–81. Bogdanor, Vernon (1995). ‘European Union and the People’, European Business Journal, 7(3): 57–63. (1997). ‘Devolving for a Stronger Union’, The World Today, 53(11): 288–90. Bosch, Agusti and Newton, Kenneth (1995). ‘Economic Calculus or Familiarity Breeds Content?’, in Oskar Niedermayer and Richard Sinnott (eds.), Public Opinion and Internationalized Governance. New York: Oxford University Press, pp. 73– 105. Brok, Elmar (2001). ‘The Institutional Reforms of Amsterdam: A View from the European Parliament’, in Jörg Monar and Wolfgang Wessels (eds.), The European Union after the Treaty of Maastricht. London/New York: Continuum. Buchanan, Sidney (2002). ‘The Abortion Issue: An Agonizing Clash of Values’, Houston Law Review, 38(5): 1383–1584. Buzan, Barry (1993). ‘Introduction. The Changing Security Agenda in Europe’, in Wæver et al. (eds.), Identity, Migration and the New Security Agenda in Europe. London: Pinter Publishers Ltd, pp. 1–16. Caporaso, James (1996). ‘The European Union and Forms of State: Westphalian, Regulatory or Post-Modern?’, Journal of Common Market Studies, 24(1): 29–52. Carey, Sean (2002). ‘Undivided Loyalties. Is National Identity an Obstacle to European Integration?’, European Union Politics, 3(4): 387–413. Chryssochoou, Dimitris N. (1994). ‘Democracy and Symbiosis in the European Union: Towards a Confederal Consociation?’, West European Politics, 17(4): 1–14. (2001). ‘The Nature of Democracy in the European Union and the Limits of Treaty Reform’, Current Politics and Economics of Europe, 10(3): 245–64. Stavridis, Stelios and Tsinisizelis, Michael J. (1998). ‘European Democracy, Parliamentary Decline and the “Democratic Deficit” of the European Union’, The Journal of Legislative Studies, 4(3): 109–29. Collier, David and Levitsky, Steven (1996). ‘Democracy “with adjectives”: Conceptual Innovation in Comparative Research’, The Helen Kellogg Institute for International Studies Working Paper Series, #230, University of Notre Dame. http://www.nd.edu/∼kellogg/ Curtin, D. M. (1999). ‘Transparency and Political Participation in EU Governance: A Role for Civil Society?’, Cultural Values, 3(4): 445–71. Dahl, Robert A. (1989). Democracy and its critics. New Haven, CT: Yale University Press.

249

Bibliography Dahl, Robert A. and Tufte, Edward R. (1973). Size and Democracy. Stanford, CA: Stanford University Press. Dahrendorf, Ralf (2003). ‘The Challenge for Democracy’, Journal of Democracy, 14(4): 101–14, reprinted version of chapter four in ‘Dopo la Democrazia’ (2001) by Antonio Polito. De Schutter, Olivier (2002). ‘Europe in Search of Its Civil Society’, European Law Journal, 8(2): 198–217. Decker, Frank (April 2002). ‘Governance Beyond the Nation-State. Reflections on the Democratic Deficit of the European Union’, Journal of European Public Policy, 9(2): 256–72. Dehousse, Renaud (1997). ‘European Integration and the Nation-State’, in Martin Rhodes, Paul Heywood, and Vincent Wright (eds.), Developments in West European Politics. London: Macmillan, pp. 37–54. Deth, Jan de van and Elinor Scarbrough (1998). The Impact of Values. Oxford: Oxford University Press. www.oxfordscholarshiponline.com Dinan, Desmond (1999). Ever Closer Union. London: Macmillan. Duchesne, Sophie and Frognier, Andrè-Paul (1995). ‘Is There a European Identity?’, in Oskar Niedermayer and Richard Sinnott (eds.), Public Opinion and Internationalized Governance. New York: Oxford University Press, pp. 193–225. Dunleavy, Patrick and O’Leary, Brendan (1987). Theories of the State. The Politics of Liberal Democracy. London: Macmillan. Easton, David (1955). A Systems Analysis of Political Life. New York: John Wiley & Sons. (1957). ‘An Approach to the Analysis of a Political System’, World Politics, 9(3): 383–400. (1964). The Political System: An Inquiry into the State of Political Science. New York: Knopf. Elazar, Daniel J. (1996). ‘From Statism to Federalism—A Paradigm Shift’, International Political Science Review, 17(4): 417–29. Etzioni-Halevy, Eva (2002). ‘Linkage Deficits in Transnational Politics’, International Political Science Review, 23(2): 203–22. Fischer, Joschka (2000). ‘From Confederation to Federation—Thoughts on the Finality of European Integration’, in Mark Leonard (ed.), The Future Shape of Europe. London: The Foreign Policy Centre, pp. 7–20. Franklin, Mark N., VanderEijk, Cees and Marsh, Michael (1995). ‘Referendum Outcomes and Trust in Government: Public Support for Europe in the Wake of Maastricht’, West European Politics, 18(3): 101–17. Fukuyama, Francis (2004). ‘The Imperative of State-Building’, Journal of Democracy, 15(2): 17–54. Gabel, Matthew (1998a). ‘The Endurance of Supranational Governance. A Consociational Interpretation of the European Union’, Comparative Politics, 30(4): 463– 76.

250

Bibliography (1998b). ‘Public Support for European Integration: An Empirical Test of Five Theories’, The Journal of Politics, 60(2): 333–54. George, Stephen and Bache, Ian (2001). Politics in the European Union. Oxford: Oxford University Press. Giddens, Anthony (2000). ‘A Third Way for the European Union?’, in Mark Leonard (ed.), The Future Shape of Europe. London: The Foreign Policy Centre, pp. 69–76. Gilje, Nils and Grimen, Harald (1993). Samfunnsvitenskapenes forutsetninger. Innføring i samfunnsvitenskapenes vitenskapsfilosofi. Oslo: Universitetsforlaget. Green, David (1999). Who Are “The Europeans”? European Identity in the Context of the Post-War Integration Project. Winner of the prize for Best 1999 ECSA Conference Paper. www.eustudies.org/DavidGreenPaper.html Habermas, Jürgen (1998a). ‘Remarks on Legitimation through Human Rights’, Philosophy and Social Criticism, 24(2/3): pp. 157–71. (1998b). ‘The European Nation-State: On the Past and Future of Sovereignty and Citizenship’, Public Culture, 10(2): 397–416. (2000). ‘Crossing Globalization’s Valley of Tears’, New Perspectives Quarterly, 17(4): 51–7. (2003). ‘Toward a Cosmopolitan Europe’, Journal of Democracy, 14(4): 86–100. Haubrich, H. (1997). ‘Identity, Solidarity and the New Geography of Europe’, Geography, 82(11): 45–57. Hayes-Renshaw, Fiona and Wallace, Helen (2006). The Council of Ministers, 2nd edn. London: Palgrave. Hazan, Reuven Y. (1996). ‘Presidential Parliamentarism: Direct Popular Election of the Prime Minister, Israel’s New Electoral and Political System’, Electoral Studies, 15(1): 21–37. Hechter, Michael (1993). ‘Values Research in the Social and Behavioral Sciences’, in Michael Hechter, Lynn Nadel, and Richard E. Michod (eds.), The Origin of Values. New York: Aldine De Gruyter. Held, David (1987). Models of Democracy. Cambridge: Polity Press. Hellevik, Ottar (1997). Forskningsmetode i Sosiologi og statsvitenskap, 5th edn. Oslo: Universitetsforlaget. Hix, Simon (1994). ‘The Study of the European Community: The Challenge to Comparative Politics’, West European Politics, 17(1): 1–30. (1998). ‘Elections, Parties and Institutional Design: A Comparative Perspective on European Union Democracy’, West European Politics, 21(3): 19–52. (1999). The Political System of the European Union. New York: St. Martin’s Press Inc. (2005). The Political System of the European Union, 2nd edn. London: Palgrave Macmillan. Hobsbawm, Eric J. (1992). Nations and Nationalism since 1780: Programme, Myth, Reality, 2nd edn. Cambridge: Cambridge University Press.

251

Bibliography Hooghe, Liesbet and Marks, Gary (2003). ‘Unravelling the Central State, but How? Types of Multi-Level Governance’, American Political Science Review, 97(2): 233– 43. (2004). ‘Does Identity or Economic Rationality Drive Public Opinion on European Integration?’, PS: Political Science & Politics, 37(3): 415– 20. Inglehart, Ronald (1970). ‘Public Opinion and Regional Integration’, International Organization, 24(4): 764–795. (1990). Culture Shift in Advanced Industrial Society. Princeton, NJ: Princeton University Press. and Reif, Karlheinz (1991). ‘Analysing Trends in West European Opinion: The Role of the Eurobarometer Surveys’, in Karlheinz Reif and Ronald Inglehart (eds.), Eurobarometer. The Dynamics of European Public Opinion. London: Macmillan, pp. 1–27. International Institute for Democracy and Electoral Assistance (2004). Voter Turnout in Western Europe since 1945. Stockholm: International IDEA Publications Office. Jachtenfuchs, Markus (2001). ‘The Governance Approach to European Integration’, Journal of Common Market Studies, 39(2): 245–64. Jagodzinski, Wofgang and Dobbelaere, Karel (1998). ‘Secularization and Church Religiosity’, in Jan W. Van Deth and Elinor Scarbrough (eds.), The Impact of Values. Oxford: Oxford University Press. Jordan, Andrew (2001). ‘The European Union: An Evolving system of Multi-Level Governance . . . or Government?’, Policy and Politics, 29(2): 193–208. Kaldor, Mary, (2003). ‘The Idea of Global Civil Society’, International Affairs, 79(3): 583–93. Kassim, Hussein (1994). ‘Policy Networks, Networks and European Union Policy Making: A Sceptical View’, West European Politics, 17(4):15–27. Katz, Richard E. (1991). ‘Materialist and Postmaterialist Values: Different Viewpoints or Different Worlds?’, in Karlheinz Reif and Ronal Inglehart (eds.), Eurobarometer. The Dynamics of European Public Opinion. London: Macmillan, pp. 295– 309. Katz, Richard S. (2001). ‘Models of Democracy. Elite Attitudes and the Democratic Deficit of the European Union’, European Union Politics, 2(1): 53–79. Kerkhofs, Jan (1991). ‘Values and the Challenge of Europe’s Identity’, in Karlheinz Reif and Ronald Inglehart (eds.), Eurobarometer. The Dynamics of European Public Opinion. London: Macmillan, pp. 377–85. Kiernan, Annabel K. (1997). ‘Citizenship—The Real Democratic Deficit of the European Union?’, Citizenship Studies, 1(3): 323–34. King, Gary, Keohane, Robert O., and Verba, Sidney (1994). Designing Social Inquiry. Scientific Inference in Qualitative Research. New Jersey: Princeton University Press. Krasner, Stephen (1999). Sovereignty. Princeton: Princeton University Press.

252

Bibliography Kuhn, Thomas S. (1962). The Structure of Scientific Revolutions. Chicago: Chicago University Press. Lehning, Percy B. (2001). ‘European Citizenship: Towards a European Identity?’, Law and Philosophy, 20: 239–82. Leonard, Mark (2000). ‘Democracy and Legitimacy in Network Europe’, in Mark Leonard (ed.), The Future Shape of Europe. London: The Foreign Policy Centre, pp. 89–100. Lijphart, Arend (1996). ‘The Puzzle of Indian Democracy: A Consociational Interpretation’, The American Political Science Review, 90(2): 258–68. (1999). Patterns of Democracy. Government Forms and Performance in Thirty-Six Countries. New Haven, CT: Yale University Press. (2002). ‘The Evolution of Consociational Theory and Consociational Practices’, Acta Politica, 37(1/2): 11–22. Lindberg, Leon and Scheingold, Stuart A. (1970). Europe’s Would-be Polity. Patterns of Change in the European Community. New Jersey: Prentice-Hall Inc. Lipset, Seymour Martin (1969). Political Man. London: Heinemann. Lord, Christopher (1998). Democracy in the European Union. Sheffield: Sheffield Academic Press. (2001). ‘Assessing Democracy in a Contested Polity’, Journal of Common Market Studies, 39(4): 641–661. and Beetham, David (2001). ‘Legitimizing the EU: Is There a “PostParliamentary Basis” for Its Legitimation?’, Journal of Common Market Studies, 39(3): 443–62. Magnette, Paul (2003). ‘European Governance and Civic Participation. Beyond Elitist Citizenship?’, Political Studies, 51: 144–60. Majone, Giandomenico (1998). ‘Europe’s ‘Democratic Deficit’: The Question of Standards’, European Law Journal, 4(1): 5–28. (2005). Dilemmas of European Integration. Oxford: Oxford University Press. www.oxfordscholarshiponline.com Mancini, G. Federico (1998). ‘Europe: The Case for Statehood’, European Law Journal, 4(1): 29–42. Marquand, David (1979). Parliament for Europe. London: Jonathan Cape. Marsh, David (1995). ‘State Theory and the Policy Network Model’, Strathclyde Papers on Government and Politics, Strathclyde. Marsh, Michael (1998). ‘Testing the Second-Order Election Model after Four European Elections’, British Journal of Political Science, 28(4): 591–607. Maurer, Andreas (2001). ‘Democratic Governance in the European Union: The Institutional Terrain after Amsterdam’, in Jörg Monar and Wolfgang Wessels (eds.), The European Union after the Treaty of Amsterdam. London/New York: Continuum. Mayer, Franz C. and Palmowski, Jan (2004). ‘European Identities and the EU—The Ties That Bind the Peoples of Europe’, Journal of Common Market Studies, 42(3): 573–98.

253

Bibliography McCormick, John (1999). Understanding the European Union. London: Macmillan Press Ltd. Mény, Yves (2003). ‘De la démocratie en Europe: Old Concepts and New Challenges’, Journal of Common Market Studies, 41(1): 1–13. Mercurio, Bryan (2003). ‘Abortion in Ireland: An Analysis of the Legal Transformation Resulting from Membership of the European Union’, Tulane Journal of International and Comparative Law, 11: 141–80. Miller, William L. (1997). ‘Quantitative Methods’, in David Marsh and Gerry Stoker (eds.), Theory and Methods in Political Science. New York: MacMillan Press Ltd. Miller, William L., White, Stephen, and Heywood, Paul (1998). Values and Political Change in Postcommunist Europe. London: Macmillan. Moravcsik, Andrew (1998). ‘Europe’s Integration at Century’s End’, in Andrew Moravcsik (ed.), Centralization or Fragmentation—Europe Facing the Challenge of Deepening, Diversity and Democracy. New York: Council of Foreign Relations. (2001). ‘Despotism in Brussels?’, Foreign Affairs, 80(3): 114–19. (2002). ‘In Defence of the Democratic Deficit: Reassessing Legitimacy in the European Union’, Journal of Common Market Studies, 40(4): 603–24. and Sangiovanni, Andrea (2003). ‘On Democracy and “Public Interest” in the European Union’, in Renate Mayntz and Wolfgang Streeck (eds.), Die Reformierbarkeit der Demokratie: Innovationen und Blockaden. Frankfurt am Maintz: MaxPlanck-Institut fur Gesellschaftsforschung, pp. 122–48. Myrdal, Gunnar (1956). An International Economy. Problems and Prospects. London: Routledge and Kegan Paul Ltd. Nicolaïdis, Kalypso (2003). ‘The New Constitution as European Demoicracy?’, The Federal Trust for Education and Research, online paper 38/03. http://www.fedtrust.co.uk/main.asp?pageid=166&groupid=6 Niedermayer, Oskar (1995). ‘Trust and Sense of Community’, in Oskar Niedermayer and Richard Sinnott (eds.), Public Opinion and Internationalized Governance. New York: Oxford University Press, pp. 227–47. Nugent, Neill (2000). June 1999 European Parliament Elections. Basingstoke: Palgrave Publishers Ltd. at www.palgrave.com/politics/eu/elections.htm (2003). The Government and Politics of the European Union, 5th edn., Houndmills: Palgrave Macmillan. Offe, Claus (1995). ‘Main Problems of Contemporary Theory of Democracy and the Uncertain Future of its Practice’, Theoria-Pietermaritzburg, 86: 21–33. (2000). ‘Democracy and Trust’, Theoria-Pietermaritzburg, 96: 1–13. Olsen, Johan P. (2002). ‘Reforming European Institutions of Governance’, Journal of Common Market Studies, 4(4): 581–602. Palmer, Harvey D. and Gabel, Matthew (forthcoming). ‘Accounting for NationalLevel Public Support for European Integration: Comparing a New Method to Existing Methods for Time-Series Cross-Section Data’, unpublished working paper cited with permission from Dr. Matthew Gabel.

254

Bibliography Parsons, Wayne (1995). Public Policy. An Introduction to the Theory and Practice of Policy Analysis. Cheltenham: Edward Elgar. Peters, B. Guy (1992). ‘Bureaucratic Politics and the Institutions of the European Union’, in Alberta M. Sbragia (ed.), Euro-politics. Institutions and Policymaking in the ‘New’ European Community. Washington: The Brookings Institution. and Jon, Pierre (1998). ‘Governance without Government? Rethinking Public Administration’, Journal of Public Administration Research and Theory, 8(2): 223– 43. Pharr, Susan, Putnam, Robert D, and Dalton, Russell J. (2000). ‘A Quarter-Century of Declining Confidence’, Journal of Democracy, 11(2): 5–25. Plattner, Mark (2003). ‘Competing Goals, Conflicting Perspectives’, Journal of Democracy, 14(4): 43–56. Pollack, Mark A. et al. (2000). ‘Democracy and Constitutionalism in the European Union’, in Mark A. Pollack (ed.), ECSA Review. Three leading EU scholars/ECSA members debate whether, and if so, by what standards, the European Union has democratic legitimacy. www.ecsa.org/DemocracyForum.htm Rasmussen, Hjalte (1986). On Law and Policy in the European Court of Justice. Dordrecht: Martinus Nijhoff. (1996). Folkestyre, Grundlov og Højesteret. Grundlovens Paragraf 20 på prøve. Copenhagen: Christian Ejlers’ Forlag. Rhodes, R. A. W. (1997). Understanding Governance. Buckingham: Open University Press. Riker, William H. (1964). Federalism. Origin, Operation, Significance, Boston, MA: Little, Brown & Co. Rittberger, Berthold (28–31 March 2003). ‘The Behavioural Implications of the EU’s ‘Democratic Deficit’ (and What Students of the ‘Democratic Deficit’ Forgot to be Bothered About)’, Paper Presented at the EUSA Eighth Biennial International Conference, Nashville, TN (quoted with permission from the author). Rohrschneider, Robert (2002). ‘The Democracy Deficit and Mass Support for an EU-Wide Government’, American Journal of Political Science, 46(2): 463–75. Rosenau, James N. and Czempiel, Ernst-Otto (eds.) (1992). Governance without Government: Order and Change in World Politics. Cambridge: Cambridge University Press. Rumford, Chris (2001). ‘Social Spaces beyond Civil Society: European Integration, Globalization and the Sociology of European Society’, Innovation, 14(3): 205–18. Saris, Willem E. (1997). ‘Public Opinion about the EU Can Easily be Swayed in Different Directions’, Acta Politica, 32(4): 406–35. Sartori, Giovanni (1987). The Theory of Democracy Revisited. Part One: The Contemporary Debate. New Jersey: Chatham House Publishers Inc. Saward, Michael (1994). ‘Democratic Theory and Indices of Democratization’, in David Beetham (ed.), Defining and Measuring Democracy. London: Sage.

255

Bibliography Sbragia, Alberta M. (1992). ‘Thinking about the European Future: The Uses of Comparison’, in Alberta M. Sbragia (ed.), Euro-politics. Institutions and Policymaking in the ‘New’ European Community. Washington: The Brookings Institution. (2002). ‘The Dilemma of Governance with Government’, Jean Monnet Working Paper 3/03. New York School of Law. http://www.jeanmonnetprogram.org/ papers/02/020301.html Scharpf, Fritz W. (1988). ‘The Joint-Decision Trap: Lessons from German Federalism and European Integration’, Public Administration, 66: 239–78. (1996). ‘Negative and Positive Integration in the Political Economy of European Welfare States’, in Gary Marks et al. (eds.) Governance in the European Union. London: Sage, pp. 15–39. (1999). Governing in Europe. Effective and Democratic. New York: Oxford University Press. Schild, Joachim (2001). ‘National v. European Identities. French and Germans in the European Multi Level System’, Journal of Common Market Studies, 39(2): 331– 51. Schmidt, Manfred G. (2002). ‘The Consociational State. Hypotheses Regarding the Political Structure and Potential for Democratization of the European Union’, Acta Political Special, 37: 213–27. Schmitter, Philippe (2000a). How to Democratize the European Union. And Why Bother?. New York: Rowman & Littlefield. (2000b). ‘Review of Beate Kohler-Koch and Rainer Eising (eds.), The Transformation of Governance in the European Union’, New York, Routledge, 1999, Acta Politica, 35: 343–46. (2000c) ‘Federalism and the Euro-Polity’, Journal of Democracy, 11(1): pp. 40– 47. (2003). ‘Democracy in Europe and Europe’s Democratization’, Journal of Democracy, 14(4): 72–85. Siedentop, Larry (2000). Democracy in Europe, London: Penguin Books. Smith, Anthony D. (1992). ‘National Identity and the Idea of European Unity’, International Affairs, 68(1): 55–76. Sørensen, Eva and Torfing, Jacob (2003). ‘Networks, Political Capital, and Democracy’, International Journal of Public Administration, 26(6): 609–43. Stepan, Alfred (1999). ‘Federalism and Democracy. Beyond the US Model’, Journal of Democracy, 10(4): 19–34. and Skach, Cindy (1993). ‘Constitutional Frameworks and Democratic Consolidation: Parliamentarism versus Presidentialism’, World Politics, 46(1): 1–22. Taylor, Paul (1996). The European Union in the 1990s. Oxford: Oxford University Press. Thaa, Winfried (2001). ‘Lean Citizenship—the Fading Away of the Political in Transnational Democracy’, European Journal of International Relations, 7(4): 503– 24.

256

Bibliography Thatcher, Mark (1998). ‘The Development of Policy Network Analyses. From Modest Origins to Overarching Frameworks’, Journal of Theoretical Politics, 10(4): 389– 416. Thomassen, Jacques (2002). ‘Parties and Voters. The Feasibility of a European System of Political Representation’, in Bernard Steunenberg and Jacques Thomassen (eds.), The European Parliament. Moving toward Democracy in the EU. Lanham, MD: Rowman & Littlefield. Verdun, Amy and Stavridis, Stelios (2001). ‘Introduction—Democracy in the European Union’, Current Politics and Economics of Europe, 10(3): 213–26. Wæver, Ole (1997). ‘After Neo-Medievalism. Imperial Metaphors for European Security’, Postmodern Studies, 24: 321–68. Buzan, Barry, Kelstrup, Morten, and Lemaitre, Pierre (1993). Identity, Migration and the New Security Agenda in Europe. London: Pinter Publishers. Wallace, Helen (1993). ‘European Governance in Turbulent Times’, Journal of Common Market Studies, 31(3): 293–301. Watson, Adam (1992). The Evolution of International Society. A Comparative Historical Analysis. London: Routledge. Weber, Max (1968). Economy and Society. Berkeley and Los Angeles: University of California Press. Webster’s New Encyclopedic Dictionary (1995). New York: Black Dog & Leventhal Publishers, Inc. Weiler, Joseph H. H. (1981). ‘The Community System: The Dual Character of Supranationalism’, Yearbook of European Law, 1: 268–306. (1998). ‘Bread and Circus: The State of the European Union’, Columbia Journal of European Law, 4(2): 223–48. (2001). ‘Federalism and Constitutionalism. Europe’s Sonderweg’, Jean Monnet Working Paper No. 3/01. New York School of Law. http:// www.jeanmonnetprogram.org/papers/00/001001.html (2002). ‘A Constitution for Europe? Some Hard Choices’, Journal of Common Market Studies, 40(4): 563–83. , Haltern, Ulrich R. and Mayer, Franz C. (1995). ‘European Democracy and Its Critique’, West European Politics, 18(3): 4–39. Wessels, Wolfgang (1996). ‘The Modern West European State and the European Union: Democratic Erosion or a New Kind of Polity?’, in Svein S. Andersen and Kjell A. Eliassen (eds.), The European Union. How Democratic Is It?. London: Sage. Wincott, Daniel (1998). ‘Does the European Union Pervert Democracy? Questions of Democracy in New Constitutionalist Thought on the Future of Europe’, European Law Journal, 4(4): 411–28. (2001). ‘Looking Forward or Harking Back? The Commission and Reform of Governance in the European Union’, Journal of Common Market Studies, 39(5): 897–911.

257

Bibliography Zielonka, Jan (2000). ‘Should the European Union Become a State?’, in Mark Leonard (ed.), The Future Shape of Europe. London: The Foreign Policy Centre, pp. 101–11. (2001). ‘How New Enlarged Borders Will Reshape the European Union’, Journal of Common Market Studies, 39(3): 307–36. (2006). Europe As Empire. Oxford: Oxford University Press.

258

Index

Aberle 151 Alesina, Alberto 22, 231 Amin, Ash 84, 85 Amsterdam, Treaty of 50 Aristotle 21 Austria 31, 162, 165, 185, 189, 194, 202, 203, 204, 213, 216 Belgium 161, 165, 185, 189, 194, 202, 203, 205 bipolarity 232 Bogaards, Matthijs 133 Brok, Elmar 29 Bulgaria 185, 187 Buzan, Barry 123 Chryssochoou, Dimitris N. 101–2, 144–5, 154 Cold War 232 confederal model 18, 93, 94–6, 130 consociational model 18, 93, 108–14, 131, 133, 242 Constitution (Constitutional Treaty) 4, 5, 10, 16, 17, 20, 49, 105, 132, 133, 243, 244 Copenhagen criteria 116 Council 36, 47, 48, 51, 53, 95, 101, 110, 112, 113, 132, 133, 217, 223, 224, 226, 234, 236, 238, 239, 245 Croatia 185, 192, 194 Cyprus 162, 172, 205, 213 Czech Republic 161, 192, 202, 204 Dahl, Robert 22, 103 Dahrendorf, Ralf 21 Dehousse, Renaud 36, 48 Delors, Jacques 17 demos 17, 19, 25, 36, 52, 61, 67, 68, 70–7, 79, 86, 88, 95, 97, 99, 103, 134, 136, 141, 144, 157,176, 236, 237, 240, 241, 242

Denmark 5, 25, 47, 149, 172, 181, 183, 185, 186, 192, 194, 197, 202, 203, 205, 213, 216, 217 Easton, David 7, 9, 10, 24, 34, 80–1, 138, 211 efficiency paradigm 17, 27, 45, 46, 55, 234 Estonia 162, 185, 189, 194, 202, 204 European Coal and Steel Community (ECSC) 1, 7, 116 European Commission 3, 12, 17, 37, 41, 42, 49, 50, 51, 95, 101, 104, 105, 110, 119, 132, 217, 219, 232, 238, 239, 243, 244 European Convention 73, 104, 107, 133, 216, 217 European Council 2, 4, 5, 95, 101, 105, 239, 244 European Court of Justice (ECJ) 51, 111 European Parliament 5, 36, 49, 50, 51, 53, 95, 101, 103, 104, 105, 108, 110, 113, 115, 119, 227, 132, 236, 238, 243 federal model 18, 93, 96–100, 130, 240 Finland 162, 175, 183, 187, 202, 203, 204, 205 Fischer, Joschka 115–16, 124, 125 France 5, 16, 30, 47, 149, 162, 183, 185, 194, 197, 198, 205, 243 Fukuyama, Francis 24 game theory 72 Germany 172, 173, 185, 194, 202, 205 Great Britain 177, 183, 185, 189, 194, 197 Greece 162, 166, 172, 175, 181, 197, 202, 205, 207, 213 Greenland 16, 25 guardianship 14, 37, 46–7, 234 Habermas, Jürgen 33, 87–8, 199, 237 Hix, Simon 8, 9, 38, 50, 66, 106, 238, 240

259

Index horizontal democracy paradigm 17, 28, 49, 55, 236 Hungary 162, 181, 183, 204, 205, 213 identity 12–13, 17–18, 34–5, 38, 51–3, 85, 86–8, 100, 106, 119, 124–5, 134, 141, 144, 152, 154, 166, 181, 242 European 12–13, 52, 84, 108, 127–8, 151, 174–5, 173, 207, 228–9, 236–8, 243, 245 feeling 162–3 national 11, 15 political 20 transnational 12, 38 intergovernmentalism 7, 14 Ireland 162, 165, 166, 183, 185, 189, 190, 194, 202, 203, 205, 213 Israel 85 Italy 162, 183, 189, 194, 202, 205 Kuhn, Thomas S. 29, 54, 58, 232, 233 Laeken Declaration 2, 4 Latvia 162, 185, 189, 194, 203, 204 legitimacy 9, 10, 14, 15, 16, 17, 21, 22, 27, 33, 34, 41, 42, 43, 45, 48, 49, 51, 52, 53, 55, 61, 65, 78, 81, 82, 84, 100, 105, 130, 131, 136, 137, 149, 234, 238, 240, 243, 245 of the union 2, 4, 18, 29, 39, 42, 87, 88, 99, 106, 107, 108, 116, 123, 128, 129, 132, 198, 215 legal 31, 57 charismatic 57 input 47, 53, 74 output 6, 38, 40, 72, 149, 243, 245 gap 18 traditional 31, 57 and the European Parliament 36, 53 and the Council 37 and majorities 71 and networks 118, 121 of the integration process 38, 46 of interest groups 38 and civil society 18 and networks 57 Leonard, Mark 19, 241 Lijphart, Arend 84, 108–4, 133, 139 Lipset, Seymour Martin 9, 10 Lithuania 187, 189, 202, 204 Lord, Christopher 31, 33–5 Luxembourg 162, 185, 194, 202, 203, 205

260

Maastricht, Treaty of, see Treaty of the European Union Magnette, Paul 18 Majone, Giandomenico 16, 40, 52, 71, 77, 78, 99–100, 243 majoritarian democracy 19, 137–8 government 20, 70–1 institutions 135, 241 principle 74, 101–2, 108, 111, 130, 134, 241 Malta 162, 172, 183, 185, 189, 192, 194, 203, 205 Marquand, David 32–3 McCormick, John 96 Mény, Yves 41–2, 44 Moravcsik, Andrew 29, 40–1, 43–4, 48, 52, 54 Neo-functionalism 7 neo-medieval empire model 18, 93, 123–9, 131, 132, 135, 241 Netherlands 5, 47, 165, 172, 185, 186, 189, 192, 198, 202, 203, 205, 213, 217 network model 18, 19, 93, 117–23, 131, 132, 241, 243 Nice, Treaty of 49, 51 Northern Ireland 177, 183, 185, 189, 194 Norway 25 Offe, Claus 70, 77 parliamentary model 18, 93, 100–4, 130, 132 Plato 21 Poland 162, 183, 185, 187, 189, 194, 197, 202, 204 Portugal 175, 183, 185, 189, 205, 207 presidential model 18, 93, 105–8, 131 QMV (qualified majority voting) 36, 46, 112 Rasmussen, Poul Nyrup 149 Riker, William H. 96–7 Romania 185, 187, 189 Rome, Treaty of 7 Rumford 18 Sartori, Giovanni 14 Sbragia, Alberta 65 Scharpf, Fritz 36, 69, 234 Schmitter, Philippe 29, 38, 48 Siedentop, Larry 12, 237 Single European Act (SEA) 42

Index Slovakia 183, 185, 289, 202, 204 Slovenia 162, 183, 202, 204 socio-psychological paradigm 17, 28, 51, 55, 61, 233, 236, 241 solidarity 10, 17–18, 20, 38–9, 69, 72, 75–6, 79, 87, 125, 127, 141, 151–3, 157–8, 175–8, 180, 227, 237–8, 240, 242–3, 245 Soviet Union 232 Spain 162, 165, 175, 183, 185, 189, 194, 202, 203, 204, 205, 207 Spolaore, Enrico 22, 231 Sweden 30, 162, 181, 186, 190, 192, 194, 204, 205, 213 Switzerland 25 Taylor, Paul 237 Treaty of the European Union 13, 37, 73, 101, 149, 217, 228, 233

Tufte, Edward R. 22 Turkey 85, 187, 188, 189, 192, 194, 198 two-tier model 18, 93, 115–17, 124, 131, 132, 240, 241 Ukraine 85 United Kingdom 5, 16, 25, 31, 47, 162, 202, 203, 204, 205, 216, 217 vertical democracy paradigm 17, 47, 55, 235 Wæver, Ole 11, 76, 77, 86, 123, 237 Wallace, Helen 38 Weiler, Joseph 16, 34, 77, 151, 237, 241 Wincott, Daniel 33, 36, 48, 54 Zielonka, Jan 124–9

261