Making EU Foreign Policy: National Preferences, European Norms and Common Policies (Palgrave Studies in International Relations Studies)

  • 48 832 3
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

Making EU Foreign Policy: National Preferences, European Norms and Common Policies (Palgrave Studies in International Relations Studies)

Palgrave Studies in International Relations Series General Editors: Knud Erik Jørgensen, Department of Political Science

2,273 1,030 849KB

Pages 253 Page size 407 x 648 pts

Report DMCA / Copyright


Recommend Papers

File loading please wait...
Citation preview

Palgrave Studies in International Relations Series General Editors: Knud Erik Jørgensen, Department of Political Science, University of Aarhus, Denmark Audie Klotz, Department of Political Science, Maxwell School of Citizenship and Public Affairs, Syracuse University, USA Palgrave Studies in International Relations, produced in association with the ECPR Standing Group for International Relations, will provide students and scholars with the best theoretically-informed scholarship on the global issues of our time. Edited by Knud Erik Jørgensen and Audie Klotz, this new book series will comprise cutting-edge monographs and edited collections which bridge schools of thought and cross the boundaries of conventional fields of study.

Titles include: Mathias Albert, Lars-Erik Cederman and Alexander Wendt (editors) NEW SYSTEMS THEORIES OF WORLD POLITICS Barry Buzan and Ana Gonzalez-Pelaez (editors) INTERNATIONAL SOCIETY AND THE MIDDLE EAST English School Theory at the Regional Level Geir Hønneland BORDERLAND RUSSIANS Identity, Narrative and International Relations Oliver Kessler, Rodney Bruce Hall, Cecelia Lynch and Nicholas G. Onuf (editors) ON RULES, POLITICS AND KNOWLEDGE Friedrich Kratochwil, International Relations, and Domestic Affairs Cornelia Navari (editor) THEORISING INTERNATIONAL SOCIETY English School Methods Dirk Peters CONSTRAINED BALANCING: THE EU’S SECURITY POLICY Simon F. Reich GLOBAL NORMS, AMERICAN SPONSORSHIP AND THE EMERGING PATTERNS OF WORLD POLITICS Robbie Shilliam GERMAN THOUGHT AND INTERNATIONAL RELATIONS The Rise and Fall of a Liberal Project Daniel C. Thomas (editor) MAKING EU FOREIGN POLICY National Preferences, European Norms and Common Policies

Rens van Munster SECURITIZING IMMIGRATION The Politics of Risk in the EU

Palgrave Studies in International Relations Studies Series Standing Order ISBN 978–0230–20063–0 (hardback) 978–0230–24115–2 (paperback) You can receive future titles in this series as they are published by placing a standing order. Please contact your bookseller or, in case of difficulty, write to us at the address below with your name and address, the title of the series and one of the ISBNs quoted above. Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire RG21 6XS, England

Making EU Foreign Policy National Preferences, European Norms and Common Policies Edited By Daniel C. Thomas School of Politics and International Relations and Dublin European Institute, University College Dublin, Ireland

© Editorial matter, selection and introduction Daniel C. Thomas 2011 All remaining material © respective authors 2011 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6–10 Kirby Street, London EC1N 8TS. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The authors have asserted their rights to be identified as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2011 by PALGRAVE MACMILLAN Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries ISBN 978-0-230-28072-4


This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Making EU foreign policy : national preferences, European norms and common policies / [edited by] Daniel C. Thomas. p. cm. Includes index. ISBN 978–0–230–28072–4 (hardback) 1. European Union countries–Foreign relations. 2. European Union countries–Foreign relations–Case studies. I. Thomas, Daniel C. (Daniel Charles) II. Title. JZ1570.8.M354 2011 341.242′2–dc22 2011004876 10 20

9 19

8 18

7 17

6 16

5 15

4 14

3 13

2 12

1 11

Printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne


List of Tables




Notes on the Contributors





The Challenge of EU Foreign Policy Daniel C. Thomas


Explaining EU Foreign Policy: Normative Institutionalism and Alternative Approaches Daniel C. Thomas

Case Studies

3 10



EU Policy on Ukraine During and Since the Orange Revolution: ‘A door neither closed nor open’ Richard Youngs



EU Policy on the International Criminal Court: Institutional Contexts and Policy Compromises Daniel C. Thomas



EU Policy on the Iraq War and its Aftermath: The Breakdown and Revival of Consensus-based Decision-making Jeffrey Lewis



EU Decision-making in CSDP: Consensus Building on Operation Artemis Simon Duke



EU Membership Negotiations with Turkey: Entrapped Again Frank Schimmelfennig



EU Policy on Economic Partnership Agreements: Trade… and Aid? Ole Elgström



vi Contents


EU Policy on Global Climate Change: The Negotiation of Burden-Sharing John Vogler





Normative Institutionalism and EU Foreign Policy in Comparative Perspective Frank Schimmelfennig and Daniel C. Thomas



Norms and All That: Progress in Research on EU Foreign Policy Thomas Risse



Interests, Power and the EU’s Role in International Security: A Sceptical Response to Normative Institutionalism Anand Menon



Norms, Institutions and EU Foreign Policy: Advancing the Research Programme Michael E. Smith




List of Tables 7.1 7.2 9.1 9.2 9.3 10.1 10.2

Member state preferences on Turkish accession Freedom House Index for selected EU membership applicants EU member state economies and emissions EU burden sharing 1997–98 ETS emissions and caps Negotiation characteristics and normative institutionalism Conditions of entrapment and cooperative bargaining


115 120 154 159 165 178 181

Acknowledgements This book is the result of a multi-year transnational collaborative research project that depended at every stage on support from individuals whose names do not appear in the Table of Contents. It all began with the Council on Foreign Relations’ decision to award me an International Affairs Fellowship – a unique and valuable programme that enables scholars of international relations to set teaching aside for a year and gain handson experience in a policy-making environment. I am grateful to the Council and especially to Elise Carlson Lewis for making this possible. In my case, the fellowship supported a sabbatical year working in Brussels at the European Commission’s Directorate General for External Relations, where Daniela Napoli courageously agreed to integrate an academic interloper into her team of dedicated professionals. This experience gave me invaluable insights into the dynamics of EU foreign policy-making that I quickly recognized were at odds with the conventional wisdom on how the EU overcomes differences in the policy preferences of its member states. Once back in academia, my new colleague Alberta Sbragia encouraged me to convert this experience into the scholarly project that eventually became this book. I am deeply grateful to Daniela and Alberta for their critical support. A stellar team of authors and discussants from across Europe and North America came together for conferences hosted in 2007 by the University of Pittsburgh’s European Union Centre of Excellence and in 2008 by University College Dublin’s College of Human Sciences and Dublin European Institute. Many thanks to Alberta Sbragia, Brigid Laffan, and Attracta Ingram for the grants that made these conferences possible, and to Andreas Dür, Reinhard Heinisch , Derek Hutcheson, David Long, Diana Panke, Eric Remacle, Tobias Theiler, and Ben Tonra for their helpful feedback on draft papers at the two conferences. Michael Cox then accepted the papers for publication as a special issue of International Politics – one of the few journals in our field that consistently makes space for collaborative projects of this sort. Finally, Knud Erik Jørgensen and Audie Klotz endorsed a revised version of the special issue for inclusion in their series Palgrave Studies in International Relations, and three noted colleagues agreed to provide critical commentaries for an extended conclusion to the collection. Palgrave editor Alexandra Webster, editorial assistant Liz Blackmore, viii

Acknowledgements ix

and production editor Shirley Tan made the final steps to publication a pleasure. To all these individuals, I owe a huge debt of gratitude. Daniel C. Thomas Dublin

Notes on the Contributors Simon Duke is Professor at the European Institute for Public Administration. He has published widely on European and transatlantic foreign and security issues, and is co-editor of the Journal of European Integration. Ole Elgström is Professor of Political Science at Lund University. He has published articles on negotiation in the EU and on the EU’s roles in international negotiations in a number of international journals and is the co-editor (with Christer Jönsson) of European Union Negotiations (2005) and (with Michael Smith) of The European Union’s Roles in International Politics (2006). Jeffrey Lewis is Associate Professor of Political Science at Cleveland State University. His current research focuses on European Union decisionmaking and has appeared in International Organization, Comparative Political Studies, and most recently, ‘How Institutional Environments Facilitate Cooperative Negotiation Styles in EU Decision Making’, Journal of European Public Policy (2010). Anand Menon is Professor of West European Politics at the University of Birmingham. He has written widely on many aspects of EU politics, including EU security policies. His latest book is Europe: The State of the Union (2009). Thomas Risse is Professor of International Politics at the Free University of Berlin and co-director of the Research College ‘The Transformative Power of Europe’. He is author of A Community of Europeans? Transnational Identities and Public Spheres (2010). Frank Schimmelfennig is Professor of European Politics at ETH Zurich, Center for Comparative and International Studies. He has published widely on European integration, EU enlargement, and EU democracy promotion. Michael E. Smith is Professor of International Relations at the University of Aberdeen. He has published numerous works on EU foreign/security policy, including Europe’s Foreign and Security Policy: The Institutionalization x

Notes on the Contributors xi

of Cooperation (2003) and Governing Europe’s Neighbourhood: Partners or Periphery? (2007). Daniel C. Thomas is Associate Professor in the School of Politics and International Relations and Director of the Dublin European Institute at University College Dublin. He is author of The Helsinki Effect (2001) and The Limits of Europe (forthcoming), among other works. John Vogler is Professor of International Relations at Keele University and a member of the ESRC Centre for Climate Economics and Policy. He has authored and edited books on the global commons and the international relations of the environment as well as (in collaboration with Charlotte Bretherton) The EU as a Global Actor (2006). Richard Youngs is Associate Professor at the University of Warwick and Director of FRIDE in Madrid. He is the author of five books on different aspects of EU foreign policy, the most recent of which is Europe’s Role in Global Politics: A Retreat from Liberal Internationalism (2010).

This page intentionally left blank


This page intentionally left blank

1 The Challenge of EU Foreign Policy Daniel C. Thomas

With American power receding and China still uncertain about its global role, the European Union clearly has the potential to exercise great influence in world affairs. Its combined GDP and its foreign investments are both comparable to those of the United States, its population is 50% larger, it has the largest total development aid budget in the world, and it has 27 votes in the United Nations, including two permanent seats on the Security Council. It has even demonstrated a capacity for united military action abroad. Some observers thus suggest that the EU has already achieved ‘superpower’ status (Cameron, 2007; Leonard, 2006; McCormick, 2006; Reid, 2005; Rifkin, 2004; Whitman, 1997). Yet others are deeply sceptical about foreign policy as a functional area of EU policy-making. They point, for example, to the EU’s failure to achieve a unified position on the US-led invasion of Iraq in 2003, the recognition of Kosovo in 2008, or the Libyan no-fly zone in 2011 as evidence of insuperable divergences in the member states’ foreign policy preferences. According to David Allen (1998: 47), the EU cannot have a real foreign policy unless it becomes a state: ‘A European foreign policy could only be achieved by creating central institutions within a European Union capable of identifying, selecting, and implementing a coherent set of objectives that could be legitimized as being in the European interest. But this could only be achieved by the establishment of a European state and hence a European government.’ Some hoped that the EU’s Lisbon Treaty, which came into effect in 2009, would solve this problem by creating a non-rotating President of the European Council and a newly-empowered High Representative for Foreign Affairs and Security Policy supported by a new External Action Service. But serious doubts persist, despite these institutional innovations. 3

4 The Challenge of EU Foreign Policy

Finnish Foreign Minister Alexander Stubb, himself a strong proponent of EU unity in world affairs, tells the following joke: ‘President Obama learns with interest that Europe now has a phone number. He’s told that, responding at last to Henry Kissinger’s famous jibe, the European Union has appointed a President named Herman Van Rompuy from Belgium and given him a 24/7 phone line. So, Obama decides to try out Europe’s phone number. Henry will be tickled. But the president forgets about the time difference and gets an answering machine: “Good Evening, you’ve reached the European Union, Herman Van Rompuy speaking. We are closed for tonight. Please select from the following options. Press one for the French view, two for the German view, three for the British view, four for the Polish view, five for the Italian view, six for the Romanian view. …” Obama hangs up in dismay’ (Cohen, 2010). This tale reflects the most fundamental challenge that the EU faces as it works to translate its vast resources into a common approach to foreign policy – decades of integration have transformed Europe’s ‘nation states’ into ‘member states’ but not into a unified super-state. So even though the EU now has more robust mechanisms for diplomacy, and the revised Treaty on European Union re-commits member states to pursue common policies, they still often struggle to reach agreement. Understanding the European Union’s actual role in world affairs thus requires, more than anything else, understanding the process by which member states pursue (and often achieve) agreement on common foreign policies. Hence the research question that drives this book: how do EU member states overcome their divergent preferences and reach agreement on common policies regarding issues and actors beyond their collective external border? Given the evident importance of this question, it is surprising how little attention it has received in the existing scholarly literature. Instead, most work on EU foreign policy has focused on explaining the evolution of the EU’s foreign policy competence and procedures (M.E. Smith, 2004), examining its impact on foreign policy formation and culture at the member state level (Tonra, 2001), and most often, on assessing the effectiveness or normative consistency of the EU’s international actions (Ginsberg, 2001; Manners, 2002; K. Smith, 2003; Telò, 2009; Whitman and Wolff, 2010).

Theory and methodology As a basis for answering this research question, the book proposes a Normative Institutionalist theory of EU foreign policy-making that

Daniel C. Thomas 5

views member states as thinly-socialized actors with quasi-autonomous preferences whose policy-making behaviour is tempered by responsiveness to the substantive and procedural norms of the Union. As a result, EU member states pursue their foreign policy preferences within an institutionalized setting that encourages certain negotiating practices and legitimates certain substantive outcomes while discouraging and delegitimating others. The theory thus expects the EU to adopt common foreign policies mid-way on the range of member state preferences and/or consistent with pre-existing EU commitments. These expectations contrast sharply with hypotheses derived from alternative theories of EU governance and policy-making – namely, the Intergovernmentalist assertion that member states rely on veto threats to pursue domestically-determined policy preferences, and Constructivist assertions that exchanges of information and moral arguments among member states cause their policy preferences to convergence (for further detail, see Thomas, ch.2, this volume). The book evaluates the three theories’ relative explanatory power through case studies of EU deliberations on contentious foreign policy issues. Each case study relies on detailed process-tracing of intra-EU negotiations to gather and assess evidence for and against the two research hypotheses and the three alternative hypotheses (Checkel, 2006; George and Bennett, 2004). To minimize bias in favour of ‘easy tests,’ the case studies cover a broad range of issue-areas, from traditional diplomatic and security issues to enlargement, trade and development, and environmental policy. These cases include some issues where supranational actors have little formal power and decisions rest on member state unanimity and others where the European Commission has an agenda-setting role and decisions are made by qualified majority voting. Wherever possible, the case studies are broken into distinct sub-cases, yielding a total of 14 cases for empirical analysis. By comparing sub-cases within and across issue-areas, the authors are able to assess the impact of other factors (‘scope conditions’) that may condition the dynamics proposed by Normative Institutionalism. To ensure relevance to the research question, all of the case studies chosen for this study exhibit considerable differences in the initial policy preferences of member states. This is not easy to determine, given that actors often seek to gain advantage in negotiations by concealing or miscommunicating their preferences (Frieden, 1999). However, the case studies minimize exposure to this risk by utilizing various combinations of evidence on preferences, including material interests, policy traditions in the relevant issue-area, policy proposals and official

6 The Challenge of EU Foreign Policy

statements on the issue at hand, and the political costs governments incur when pursuing their declared preferences. Having mapped the divergences in preferences that structured intra-EU deliberations on the issue in question, including the preferences of likely veto players, each case study then deduces the lowest common denominator outcome that would result from potential veto players refusing to make any meaningful compromise on their preferences. Each case study then traces the decision-making process to determine whether (and if so, how) various structural pressures and negotiation behaviours moved the member states toward (or away from) agreement on a common policy. Using evidence of the actual policymaking process and the content of the policy agreed by the member states, they then evaluate the relative explanatory power of the various hypotheses. When a case reveals evidence that multiple tactics were used in the course of intra-EU negotiations, evaluation of the hypotheses focuses on their distinct claims regarding the process that dominates movement toward agreement and the type of outcome to be expected. In so doing, the book aims to subject EU foreign policy analysis to the same theoretical and empirical rigour that is now taken for granted in studies of EU treaty-making or regulatory compliance, to name just two other areas of scholarship on EU governance.

Plan of the book Following this introduction, Daniel C. Thomas (ch.2) outlines the Normative Institutionalist theory of EU foreign policy-making, clarifying the assumptions underlying the theory, introducing two testable hypotheses derived from these assumptions, and identifying observable implications regarding causal mechanisms and the type of policy outcome to be expected. The theory’s ‘normative entrapment’ hypothesis suggests that once member states have committed themselves to a particular set of norms and/or policy course, they are likely to find themselves constrained to take further actions that do not reflect their original intentions and/or current preferences. The theory’s ‘cooperative bargaining’ hypothesis suggests that the member states’ increased identification with Europe, the consensus norm and consultation reflex undermine their veto rights and produce policy deliberations dominated by cooperative tactics rather than the competitive tactics that prevail in less-institutionalized settings. The chapter compares the causal mechanisms and observable implications of these hypotheses to those

Daniel C. Thomas 7

of a competitive bargaining and lowest-common denominator hypothesis derived from Intergovernmentalism as well as policy learning and normative suasion hypotheses associated with Constructivist theories of EU governance, and concludes with a brief final note on the empirical challenge of testing these various explanations. Within the traditional domain of foreign affairs and security policy, where member states most clearly dominate policy-making, the first case is Richard Youngs’ analysis (ch.3) of EU support for democratic reform in Ukraine, focused on policies during and after that country’s ‘Orange Revolution’. This study finds considerable evidence of normative entrapment in the first sub-case but an LCD-like outcome in the second sub-case, and highlights the changing external conditions that undermined the explanatory power of Normative Institutionalism. The second CFSP case is Daniel C. Thomas’ analysis (ch.4) of the EU’s response to US challenges to the International Criminal Court, which highlights the importance of institutional context in explaining intraEU negotiations by comparing an LCD outcome that emerged during deliberations among three EU member states in the UN Security Council and a mutual compromise that resulted from substantive and procedural norms when a very similar issue was raised within CFSP forums. Jeffrey Lewis’ study (ch.5) of EU policy on Iraq traces the movement from deadlock over the initial US-led invasion to the restoration of consensus through mutual concessions in favour of reconstruction assistance, which he finds was enabled by the member states’ sense of having endangered the procedural norms that traditionally guide Council deliberations and the entrapment effect of substantive EU norms on democracy, humanitarian assistance, and the value of joint action in foreign affairs. Simon Duke (ch.6) then examines how the EU achieved consensus in support of a limited military mission to the Democratic Republic of Congo, finding considerable evidence of normative entrapment regarding the authorization of the mission and far less regarding the commitment of troops and equipment. This reflects the unique character of the EU’s Common Security and Defence Policy, which allows each member state to decide the extent of its involvement in EU missions. Leading off the case studies within the broader domain of ‘external relations’, Frank Schimmelfennig (ch.7) explains why the EU has sometimes opened the door to Turkish accession while at other times it has moved toward closing the door. Comparing four stages or sub-cases in intra-EU negotiations regarding Turkish membership since the late

8 The Challenge of EU Foreign Policy

1990s, the paper argues that sceptical member states find themselves entrapped by the Union’s membership norms and prior commitments to Ankara when Turkey is moving toward implementation of fundamental EU norms, but free to act on their preferences when Turkey is contradicting EU norms. The final two case studies examine policy areas where the supranational European Commission is generally assumed to dominate EU policy-making – trade and environment – and find that not only do member states play a major role, but their negotiating behaviour and policy agreements generally fit the expectations of Normative Institutionalism. In his paper on Economic Partnership Agreements with developing countries, Ole Elgström (ch.8) finds that EU substantive norms of free trade and development assistance were hard for member states to disavow, but the tension between these norms left space for real negotiation. Nonetheless, the EU’s procedural norms of consensus-oriented negotiations facilitated a mutual-compromise outcome by limiting member states’ willingness to exploit this opportunity. Finally, John Vogler (ch.9) examines policy on climate change, where one might expect the material interests of member states to dictate hard bargaining and a lowest common denominator outcome. Instead, he argues, the Union’s high-profile role in the 1996–97 Kyoto negotiations created expectations regarding the EU as a global actor that entrapped the more cost-sensitive member states into accepting an EU burden-sharing deal for the post-2012 climate change regime that is far more stringent than what would have resulted from competitive bargaining in defence of national energy sectors. In conclusion, Frank Schimmelfennig and Daniel C. Thomas (ch.10) review the case studies’ findings with regard to the explanatory power of the various hypotheses and associated scope conditions, and discuss the research and policy implications of these findings. In order to further advance the research agenda on EU foreign policy-making, the volume then offers critical commentaries by three scholars in the field noted for their distinctive approaches to EU studies. Thomas Risse (ch.11) praises the volume for advancing the field of EU foreign policy studies by bringing competing hypotheses to bear on multiple cases of EU decision-making and for avoiding the debate over whether the EU is a ‘normative’ power, but questions the internal logic of Normative Institutionalism and its relationship to other institutionalist approaches to social science. Anand Menon (ch.12) focuses on EU policy-making regarding traditional diplomatic and security issues, critiquing Normative Institutionalism’s neglect of power resources and the behind-the-scenes play of national interests and questioning the normative consequences

Daniel C. Thomas 9

of policy-making by consensus. Finally, Michael E. Smith’s commentary (ch.13) addresses conceptual and methodological issues raised by Normative Institutionalist explanations of EU foreign policy, notes the strengths and weakness of the present study, and offers specific recommendations for future efforts of this kind. In total, these four concluding essays identify a diverse and challenging set of issues to be addressed by any scholar interested in advancing our understanding of the sources of EU foreign policy.

Works cited Allen, D. (1998) ‘Who Speaks for Europe? The Search for an Effective and Coherent External Policy’, in J. Peterson and H. Sjursen (eds) A Common Foreign Policy for Europe? Competing Visions of the CFSP. Oxford: Routledge, pp.41–58. Cameron, F. (2007) An Introduction to European Foreign Policy. London: Routledge. Checkel, J.T. (2006) ‘Tracing Causal Mechanisms’, International Studies Review, 8(2): 362–70. Cohen, R. (2010) ‘Gone, Solid Gone’, The New York Times, March 8, 2010. Frieden, J. (1999) ‘Actors and Preferences in International Relations’, in D.A. Lake and R. Powell (eds) Strategic Choice and International Relations. Princeton, NJ: Princeton University Press, pp.39–76. George, A.L. and Bennett, A. (2004) Case Studies and Theory Development in the Social Sciences. Cambridge, MA: MIT Press. Ginsberg, R.H. (2001) The European Union in International Politics: Baptism by Fire. New York: Rowman and Littlefield. Leonard, M. (2006) Why Europe Will Run the 21st Century. London: Fourth Estate. Manners, I. (2002) ‘Normative Power Europe: A Contradiction in Terms?’, Journal of Common Market Studies, 40: 235–58. McCormick, J. (2006) The European Superpower. Basingstoke: Palgrave Macmillan. Reid, T.R. (2005) The United States of Europe: The Superpower No-One Talks About. London: Penguin. Rifkin, J. (2004) The European Dream: How Europe’s Future is Quietly Eclipsing the American Dream. Cambridge: Polity Press. Smith, K.M. (2003) European Foreign Policy in a Changing World. Cambridge: Polity Press. Smith, M.E. (2004) Europe’s Foreign and Security Policy: The Institutionalization of Cooperation. Cambridge: Cambridge University Press. Telò, M. (2009) The European Union and Global Governance. London: Routledge. Tonra, B. (2001) Europeanisation of National Foreign Policy: Dutch, Danish and Irish Foreign Policy in the European Union. Aldershot: Ashgate Publishers. Whitman, R.G. (1997) From Civilian Power to Superpower? The International Identity of the European Union. Basingstoke: Palgrave Macmillan. Whitman, R.G. and Wolff, S. (ed.) (2010) The European Neighbourhood Policy Since 2003: Much Ado about Nothing? Basingstoke: Palgrave Macmillan.

2 Explaining EU Foreign Policy: Normative Institutionalism and Alternative Approaches Daniel C. Thomas

This chapter analyses the making of EU foreign policy on discrete issues by deducing hypotheses from general theories of bargaining, negotiation and socialization in international relations and testing them with a standard empirical research methodology. To start, it introduces a Normative Institutionalist theory of EU foreign policy-making that highlights the effect of pre-existing EU norms and commitments on foreign policy negotiations among member states, specifying the observable expectations and scope conditions of two hypotheses derived from the theory. It then contrasts these hypotheses with three alternatives hypotheses highlighting the effects of veto threats, policy learning and normative suasion. Before proceeding, it is important to clarify that the paper defines EU foreign policy as the set of policies adopted by the Union’s member states to address issues and manage relationships beyond their collective external border. This includes policy areas dominated by the member states, namely the Common Foreign and Security Policy (CFSP) and Common Security and Defense Policy (CSDP),1 areas where policy-making competence is shared by member states and supranational institutions, such as enlargement, environment and anti-terrorism, and areas where supranational institutions play a leading role, such as trade, even though the EU classifies the latter two as ‘external relations’. Accordingly, the dependent variable for empirical analysis of EU foreign policy could be any of the following policy outputs, which we define as ‘common policies’: Council Conclusions related to world affairs; Common Strategies, Common Positions, Joint Actions and other CFSP/CSDP instruments; and the wide variety of positions adopted in other areas of external relations, such as mandates for international negotiations on trade, environment, or EU accession, decisions on development or humanitarian assistance, or the imposition of sanctions, to name just a few. The 10

Daniel C. Thomas 11

chapter takes as given (i.e., does not seek to explain) the policy preferences that member states bring to the start of EU-level deliberations, but the final two hypotheses emphasize the possibility that intra-EU contacts may cause these preferences to change.

The puzzle and research question EU member states have been extremely reluctant to transfer decisionmaking authority for foreign and security policy to supranational European institutions. In contrast to other policy areas, where the member states have accepted a prominent role for the European Commission and the European Court of Justice, they have insisted on a heavily intergovernmental process for decision-making on foreign and security policy (Koenig-Archibugi, 2004). As a result, the provisions of the CFSP grant considerable bargaining power to individual member states with respect to the adoption of particular policies (Decaux, 2002; Müller-BrandeckBocquet, 2002). With the exception of measures taken to implement pre-existing Common Strategies, all policies adopted under CFSP or CSDP require unanimous support and are thus subject to veto by a single member state. And since the EU has adopted relatively few Common Strategies (as opposed to the more frequent Common Positions and Joint Actions), the vast majority of policies are adopted by unanimous agreement of the member states. Even where the Treaty on European Union allows qualified majority voting (QMV), an individual member state may block it for ‘important and stated reasons of national policy’. The European Commission has greater authority on ‘external relations’ issues, and QMV is formally the rule here, but member states retain considerable powers to block the adoption of policies they do not like (Pocar, 2002). As such, the British government’s blunt assertion – ‘When we don’t agree, there is no common policy’ – could thus have been voiced by any member state in any area of EU foreign policy (Reynolds, 2004). It is thus noteworthy that the EU’s member states have reached agreement on more than 1,000 legally-binding common strategies, common positions, and joint actions under CFSP since 1993, involving high-profile diplomatic disputes, economic sanctions, and even the use of military force. Adoption of such common policies has increased considerably over time, from an average of just over 15 per year in 1993–95 to an average of 108 per year in 2004–08. Although the possibility of qualified majority voting was introduced in 1999, approximately three-quarters of all common policies are still adopted by unanimity.2 In addition, the EU Council

12 Explaining EU Foreign Policy

and Presidency issued 2,054 statements or declarations during 19952008 – an average of one every two and half days – reflecting member state consensus on particular developments in world affairs.3 Finally, EU member states generally act together in the United Nations Security Council, General Assembly, and other UN forums (Laitikainen and Smith, 2006; K. Smith, 2006; Telò, 2009). In short, the empirical record is inconsistent with the sceptics’ portrayal of CFSP as a policy-area plagued by non-cooperation. When we consider other foreign policy areas that the EU classifies as ‘external relations’, such as enlargement, development cooperation, global environmental policy, and especially trade, the record of policy agreement despite divergent preferences is even more striking. One could hypothesize, of course, that the EU achieves common policies when member states’ preferences are identical and fails to achieve common policies when preferences diverge. Yet easy harmony of preferences is rare and this hypothesis leads one to expect failure of agreement except for those few issues where the member states’ preferences are identical or where the stakes are so small that nobody cares about the outcome. Moreover, we know this to be false: member states’ representatives often argue intensively about issues that nobody considers inconsequential before they reach agreement on a common policy. Nor can we simply dismiss these debates as meaningless rhetorical exercises. The reason that such debates are so intense, it is reasonable to assume, is that even where some preference convergence has occurred, the quest for agreement on an EU policy involves the compromise of policy preferences and adjustment of policy behaviour by member states that otherwise would prefer to avoid such concessions. This focus on the challenge of overcoming the divergent policy preferences of the member states is not to say that the EU’s supranational bodies are irrelevant to the making of EU foreign policy: quite the opposite is true. Nor is it to say that EU member states always have radically different preferences. In fact, there is good reason to believe that frequent and intensive consultation between member states has weakened egoistic identities and accustomed national policy-makers to seeking out the views of their EU counterparts before determining a national position on a particular issue. Over time, this has produced a partial convergence in the strategic cultures and identities of the member states and thus in their foreign policy preferences. Yet even the most optimistic observer would admit that this Europeanization of the foreign policy-formation process is far from complete: the pursuit of common and community policies most often involves not implement-

Daniel C. Thomas 13

ing a single set of preferences shared by the member states, but negotiating the differences that persist between them with regard to policy goals and strategy. Thus we return to the research question introduced above: How and under what conditions do EU member states overcome their divergent preferences to reach agreement on common policies regarding issues and actors beyond their collective external border? This question draws attention to the behavioural choices of potential veto player(s) in intraEU negotiations – i.e., the member state(s) whose agreement is critical for the EU to achieve consensus. While all member states are formally capable of obstructing agreement, the veto card is most likely to be played on any given issue by the most conservative member state (seeking the smallest departure from the status quo), the most ambitious (seeking a great departure from the status quo and unwilling to accept less), the most committed (with the most intensive commitment to its preference), the most interested (with the most at stake in this policy area), the most powerful (whose choices will be closely watched by others), or some group thereof. The five hypotheses presented below differ not only in their general claims regarding the factors determining EU policy choice, but also with regard to the role that potential veto players are likely to play in intra-EU negotiations.

Explaining EU common policies This section of the paper presents the project’s core and alternative hypotheses regarding how EU member states overcome divergent preferences to reach agreement on common policies and what type of policy outcome is likely to result. The core hypotheses reflect a Normative Institutionalist approach to international relations and European governance, which posits that the EU’s substantive and procedural norms significantly shape the behaviour of its member states. The relative explanatory power of this approach can only be assessed, however, when compared to alternatives whose assumptions, observable expectations, and scope conditions have also been clearly and fairly presented. This presentation of Normative Institutionalism and associated hypotheses is thus followed by alternative hypotheses focused on the veto option, the exchange of policy-relevant information, and normative suasion. This theoretical exposition lays the groundwork for an assessment of the various hypotheses’ explanatory power through an examination of their correspondence with evidence from intra-EU negotiations and policy outcomes across a range of case studies.

14 Explaining EU Foreign Policy

Normative Institutionalism Normative Institutionalism is one strand of the new institutionalist theorizing current in political science and international relations since the late 1980s (March and Olsen, 1989; Peters, 1999). Institutionalist theories of European integration and governance treat the EU as a community of states whose rules and supranational organizations exert a significant impact on policy outcomes (Jupille and Caporaso, 1999; Stone Sweet, et al., 2001). While recognizing that member states dominate the creation of EU institutions, Institutionalism asserts that these institutions come to exert an independent effect on member states’ subsequent negotiating behaviour and policy choices. In short, ‘membership matters’ (Sandholtz, 1996). Some versions of Institutionalism emphasize the EU’s supranational organizations and transnational policy entrepreneurs (Armstrong and Bulmer, 1998; Sandholtz and Stone Sweet, 1998), while others emphasize formal rules such as the EU treaties’ distribution of policy-making competences (Tsebelis, 1994; Garrett and Tsebelis, 1996). In contrast, Normative Institutionalism emphasizes the behavioural impact of standards of appropriateness established by the community’s normative and policy commitments. Normative Institutionalism assumes that member states are committed to ensuring the Union’s political viability (albeit probably to varying degrees) and value being seen as acting in accordance with its normative and policy commitments, regardless of their divergent preferences with regard to particular policy issues. It asserts that EU decision-making on any given issue is shaped by the normative and policy commitments already made by member states in the course of creating the Union’s institutions, enlarging its borders, and adopting policies to govern its internal dynamics and external relationships. In fact, given its dense web of laws and regulations and the multitude of policies that it has adopted, the European Union is almost certainly the most normatively structured transnational space on Earth. So what are these normative and policy commitments? Through a combination of rhetorical commitment and treaty-making, EU member states have developed two meta norms as a general guide to their deliberations on foreign policy: joint action as an intrinsic value, including support for the functionality and credibility of the EU as a global actor; and consistency and coherence in EU policy-making across time and issue-areas. In addition to these meta-norms, EU member states have established a number of substantive (i.e., ends-oriented) and procedural (i.e., means-oriented) norms. In treaties and Council conclusions, member states have formally and repeatedly identified support for democracy

Daniel C. Thomas 15

and the rule of law, human rights, conflict prevention, the strengthening of multilateral institutions, free trade, and the promotion of development as the principal goals of EU foreign policy. In recent years, environmental protection has also emerged as a substantive EU norm. In addition, as discussed above, the EU has formally adopted nearly a thousand legal acts under CFSP (not to mention all those in other areas of external relations) to address issues and problems in world affairs. Some of these acts are politically insignificant, but the list includes hundreds of substantive policies that all EU member states are legally bound to support. Furthermore, at least three procedural norms have developed informally through member state practice over time: regular communication and consultation, including the expectation that member states will not publicize policy commitments before the Union has deliberated; confidentiality in the use of shared information; and decision-making by consensus, including the avoidance of qualified majority voting even where it is permitted by treaty (Jørgensen, 1997; Lewis, 1998, 2003; M.E. Smith, 2004). As a result, EU member states pursue their foreign policy preferences within an institutionalized setting that encourages certain negotiating practices and legitimates certain substantive outcomes while discouraging and delegitimating others. If Normative Institutionalism is correct, these norms and policy commitments should have a significant effect on how member states negotiate divergences in their policy preferences and on the type of policies they adopt at the EU level. The likelihood that the EU will adopt a common policy on a given issue, as well as the content of that policy, thus depend upon both the distribution of preferences among the member states (which varies from issue to issue) and how EU norms affect their choices in pursuit of those preferences. The following two hypotheses – complementary but not necessarily interdependent – specify this expectation in more detail. H-1: Entrapment Although the policy preferences of EU member states may diverge on particular issues, they value ‘coherence and consistency’ in EU foreign policy and value being seen as acting in accordance with the community’s normative and policy commitments. As a result, this hypothesis asserts, the policy-making behaviour of member states is shaped significantly by shared perceptions regarding which policy options are consistent or inconsistent with pre-existing EU norms and commitments. Those member states whose policy preferences are seen as

16 Explaining EU Foreign Policy

inconsistent with the EU’s substantive norms or policy commitments are less willing to insist on their preferences and more acquiescent to those with norm-consistent preferences. They thus compromise their preferences and ‘play along’ with the norm-consistent policy because they expect the social rewards for doing so to exceed the costs of the compromise. As a result, once member states have committed themselves to a particular set of norms and/or policy course, they are likely to find themselves entrapped, constrained to take further actions that do not reflect their original intentions and/or current preferences (Schimmelfennig, 2001, 2003, 2004). This dynamic affects all member states, including potential veto players. If this is correct, one would expect EU common and community policies to be consistent with pre-existing EU norms and policy commitments. Of course, the perception of normative (in)consistency is subject to deliberate acts of ‘framing’ that link issues and choices to pre-existing ideas and prior experiences (Benford and Snow, 2000; Payne, 2001). Successfully framing a possible common or community policy as consistent with the EU’s formal norms and prior policy commitments disempowers its opponents, while framing a possible policy as inconsistent with existing norms and commitments disempowers its supporters. All political actors – whether member state, supranational or nongovernmental organization – thus have a powerful incentive to frame EU policy choices in terms of pre-existing norms and commitments consistent with their policy preferences. For example, where one EU actor has set the negotiation agenda by highlighting an EU norm conducive to its preferences, others may seek to level the playing field by offering reasons why alternative norms should guide policy choice. Member states that fail to do so will find themselves trapped in support of EU policies that diverge from their preferences. Normative Institutionalism’s ‘entrapment’ explanation can thus be represented as follows: Divergent member state preferences + Rhetorical framing → Entrapment → Norm-consistent policy Furthermore, entrapment is most likely to occur when several conditions are met: C-1: Determinacy. Entrapment is more likely when the relevant substantive norms are determinate – that is, when actors, regardless of their preferences, have little doubt about which norm applies to the

Daniel C. Thomas 17

issue at hand, which policy behavior it condones and which it condemns (Franck, 2000). C-2: Precedent: Entrapment is more likely when the EU has already made policy commitments on the issue at hand – that is, already invested its resources and reputation on behalf of one principle or party involved in the issue at hand. C-3: Relevance: Entrapment is more likely when external conditions are consistent with the assumptions that underlay the existing EU norm or policy commitment. Where conditions are inconsistent, actors are disentrapped (or released) from their normative or policy commitments. For example, entrapment is unlikely if a member state is beholden to an international norm that all member states recognise as more obliging than the EU norm. Or, if an EU norm stipulates that a third party must act in a certain way in order to receive certain treatment from the Union, the third party’s failure to act this way releases the Union from its obligation. C-4: Forum: Entrapment is more likely when policy deliberation occurs within forums where EU norms and policy commitments are salient and thus exert strong compliance pull. Generally speaking, this means within the EU, though it may also pertain to deliberations within closely allied institutional forums such as the Council of Europe. When member states negotiate in a forum shielded (at least partly) from the compliance pull of EU norms, which is most likely outside the EU, then entrapment is less likely. EU member states are disentrapped when operating within institutional forums that assign them distinctly non-EU roles, such as mediator between third parties or the rotating presidency of an international organization. C-5: Publicity: Entrapment is more likely when the issue under discussion has received significant public attention, which increases the likelihood that non-compliance with existing EU norms or policy commitments will be noticed and subject to disapproval. However, many policy-making scenarios cannot be reduced to a single and determinate norm legitimating a single course of action. If two or more EU norms are equally relevant to the issue at hand, they may dictate contradictory policies and thus create a situation of mutual

18 Explaining EU Foreign Policy

entrapment. For example, member states that prefer policy X may find themselves entrapped by the norm dictating policy Y while member states that prefer policy Y are simultaneously entrapped by norm X. Or member states may agree on which EU norm is relevant to an issue, but if the norm is indeterminate, they may differ over its implications for the choice at hand. In either situation, the dynamics of entrapment discussed above will be under-determining with regard to policy choice. When faced with situations where entrapment is under-determining, member states may resort to threats in pursuit of their policy preferences, may exchange information that could influence each other’s factual understanding of the issue at hand, or they may seek to persuade each other that one course of action is normatively superior to others, as suggested by the alternative hypotheses introduced below. But as the Cooperative Bargaining hypothesis explains, they may also bargain in a manner consistent with the premises of Normative Institutionalism. H-2: Cooperative bargaining The growth and institutional stability of the EU have created the conditions that, according to negotiation scholars, foster a cooperative approach to negotiations: identification with common goals and values and trust in the dynamics of diffuse reciprocity (Elgström and Jönsson, 2000; Scharpf, 2006). These conditions encourage EU negotiators ‘to remember some identities and common ties, and to forget identities that tend to create cleavages and conflicts’ (March and Olsen, 1998: 961). Over time, these conditions have been reinforced by procedural norms that encourage policy-makers to consult each other before publicizing their preferences, to seek consensus, and to refrain from making veto threats, which are now deeply embedded in member state practice and expectations. Hence the tendency among scholars to refer to a ‘consultation reflex’ shared by foreign policy-makers in EU member states (Nuttall, 1992). In the words of a senior British official, ‘The foreign policy process has become Europeanised, in the sense that in every international issue, there is an exchange of information and an attempt to arrive at a common understanding and a common approach – compared to how things were in the past, where most issues were looked at in isolation without addressing the attitudes of other member states or a European dimension’ (cited in Aggestam, 2004: 81). The increased identification with Europe, the consensus norm and consultation reflex have not eliminated divergences in member state

Daniel C. Thomas 19

preferences, but they have significantly transformed the process by which member states deliberate collectively on foreign policy. Similarly, while these developments have not denied member states the ability to exercise a veto in pursuit of their policy preferences, they have made member states less willing to act on their veto rights and more inclined to discount such threats made by others. It is therefore reasonable to hypothesize that as member states deliberate over foreign policy issues at the EU level, movement toward common policies is dominated by cooperative tactics (also known as ‘integrative bargaining’ or ‘problemsolving’) rather than the competitive tactics that prevail in lessinstitutionalized settings (Lax and Sebenius, 1986; Hopmann, 1995; Elgström and Jönsson, 2000). And when the collective goal is to reach an agreement that comes as close as possible to satisfying the preferences of all member states – e.g., by seeking to ‘expand the pie’ rather than to divide it – a different type of policy outcome is sure to result.4 If this hypothesis were correct, we would expect intra-EU negotiations to be characterized by a great deal of give-and-take in the context of an intensive search for solutions that are acceptable (if not ideal) for the greatest number of member states. We would also expect that common policies resulting from this process would embody mutual compromises by all member states, including those with the formal ability to avoid compromise by defending their preferences with a veto threat.5 Yet since politicians are generally reluctant to publicize their concessions, we would expect EU and member state spokespeople to justify their negotiated agreements before European and foreign audiences by referring to the Union’s pre-existing substantive norms and policy commitments.6 In sum, Normative Institutionalism’s ‘cooperative bargaining’ explanation for EU policy outcomes can be represented as follows: Divergent member state preferences + Consensus norm/Consultation reflex → Cooperative bargaining → Mutual compromise policy Of the two conditions likely to promote cooperative bargaining, the first (forum) is identical to one of the conditions for entrapment and the second (secrecy) is the exact opposite: C-6: Forum. Cooperative bargaining is more likely to emerge when the issue in question is subject to collective deliberation within EU forums, where the EU’s procedural norms are most salient.

20 Explaining EU Foreign Policy

C-7: Publicity. Cooperative bargaining is more likely to emerge when deliberations occur in camera – that is, away from the media spotlight that raises the domestic political costs of compromise.7 The Normative Institutionalist theory of intra-EU negotiations thus emphasizes two explanatory hypotheses, each with its own scope conditions. In the end, it expects that member states with considerable bargaining power will nonetheless make concessions in order to achieve a common policy that is consistent with the EU’s existing commitments and/or that is partly responsive to the preferences of weaker member states. In fact, the theory suggests that the EU will sometimes agree on new common policies even when some of the member states prefer the status quo over all alternative policies on the table. Alternative hypotheses Looking beyond Normative Institutionalism, one could identify a vast number of alternative explanations for EU agreement on common foreign policies. In the interest of manageability, this project considers three such alternative hypotheses, each of which posits a different relationship between interests, institutions, interactions and policy outcomes and thus contests a different assumption of the Normative Institutionalist approach. The alternative hypotheses differ from Normative Institutionalism with regard to whether the likely character of intra-EU negotiations and the resulting policy are determined more by pre-existing EU norms or by the veto option (H-3), and with regard to whether interactions among EU member states affect just policy choices or underlying interests as well (H-4 & H-5). H-3: Competitive bargaining This hypothesis is close to the conventional wisdom on EU foreign policy-making. It derives from a pessimistic reading of the Intergovernmentalist theory of European integration and governance, which treats the EU as an international forum in which member states act strategically in pursuit of their interests and policy preferences on particular issues. Intergovernmentalist theory attributes these preferences to the interaction of international pressures and domestic political considerations, and assumes that they are unaffected by EU membership or participation in EU institutions. Furthermore, Intergovernmentalism assumes that divergences in the member states’ policy preferences are not significantly compensated by a shared commitment to common goals or values, and it emphasizes

Daniel C. Thomas 21

the fact that each member state wields a potential veto over policy proposals (Hoffmann, 1966; Moravcsik, 1999). This pessimistic reading of Intergovernmentalism expects member states to treat intra-EU negotiations as zero-sum games in which each seeks to maximize its own preferences. Member states may trade concessions on off-setting issues and thus shift their bottom-line in bargaining over policy choices, but such exercises in ‘specific reciprocity’ (Keohane, 1986) do not constitute a shift in their underlying preferences. These assumptions regarding the preferences and calculations of member states lead to the expectation of competitive bargaining (also known as ‘hard bargaining’) over policy alternatives (Scharpf, 1988). In particular, whichever member state is least receptive to change will dominate EU foreign policy by threatening to veto any proposal that is further from the status quo than its own ideal policy (Jupille, 1999). As a result, the EU will have difficulty acting decisively, if at all, in world affairs: as long as common policies require unanimous support, ‘the EU will be hampered… by the constant threat of having one of its numerous member states break from its ranks’ (Meunier, 2000: 132). Where member states’ policy preferences are mutually exclusive (such as the choice between cutting and expanding ties to a particular country), competitive bargaining will prevent the adoption of any common policy – an outcome known as deadlock. Where the disagreement of the member states is a matter of degree (such as how much aid should be given to a particular country), competitive bargaining will likely result in agreement on whatever policy is acceptable to all member states and closest to the ideal outcome of the one least receptive to change (i.e., the ‘lowest common denominator’ or LCD). In practice, this LCD policy may be the status quo (agreement not to change an existing policy) or a new policy that reflects the preferences of the veto player. And since EU enlargement increases the number of potential veto players, the likelihood of agreement on non-LCD policies is further reduced by the adhesion of every new member state (Tsebelis, 2002). The expectation of LCD outcomes is undeniable, of course, if one formally equates any non-status quo outcome with the preferences of the actor that is least open to policy change: ‘When each Member State possesses the power of veto, whether at the outset of a negotiation or at the ratification stage, the common position eventually reached is the lowest common denominator’ (Meunier, 2000: 109). However, this equation blinds us to the possibility that some EU-level aspect of the

22 Explaining EU Foreign Policy

policy-making process might lead member states to forego the pursuit of outcomes that they prefer with respect to the issue at hand. In fact, evidence of member states making veto threats is not sufficient evidence that the policy developed as expected by the competitive bargaining hypothesis – particularly if there is evidence that the threat was followed by a willingness to reach agreement through mutual concessions. The LCD outcome is thus better treated as an empirical possibility than as a theoretical assumption.8 Intergovernmentalism’s ‘competitive bargaining’ explanation for EU policy outcomes can thus be represented as follows: Divergent member state preferences + Veto option → Competitive Bargaining → Deadlock or Lowest Common Denominator policy In short, intra-EU negotiations will yield policy deadlock or lowest common denominator outcomes because the threat of a veto by the member state(s) with strong bargaining power will force the weak to compromise in favour of the policy preferred by the powerful. While this hypothesis is close to the conventional wisdom on EU foreign policy, other theories of international relations provide additional alternatives to Normative Institutionalism’s focus on the behavioral effects of EU norms and policy commitments. H-4: Policy learning This hypothesis derives from theories of how actors ‘learn’ new policy preferences from each other as they grapple with the complexity and interconnectedness of international issues (Goldstein and Keohane, 1993; Levy, 1994). EU member states face considerable uncertainty regarding the implications of various foreign policy options for their interests and values. This uncertainty is most common, and policy learning is thus most likely to occur, when highly-technical scientific or economic issues are under discussion. However, the frequent opportunities for consultation offered by EU institutions enable them to reduce uncertainty by exchanging policy-relevant ideas and information. Seen in this light, the principal value of intra-EU foreign policy consultation is its contribution to a learning process whereby member states whose policy preferences were originally divergent eventually converge around policy preferences indicated by particular bodies of information or causal ideas. Once preference convergence has occurred, agreement on common and community policies is not difficult.

Daniel C. Thomas 23

In sum, the ‘Policy Learning’ explanation for EU policy outcomes can be represented as follows: Divergent member state preferences + Exchanges of information → Policy learning → Preference convergence → Consensus policy Conceived in this manner, the learning process does not affect or transform the fundamental properties of the actors involved. However, as reflected in the next hypothesis, it is possible to imagine that ideational exchanges among member states may have a far more profound effect than a simple recalculation of which policy is most likely to realize a given set of interests. H-5: Normative suasion This hypothesis derives from Constructivist theories of international relations and European governance that posit an interactive relationship between the interactions of states and the understandings of self and self-interest that drive their behaviour (Adler, 2002; Risse, 2004).9 It portrays the EU as a densely-integrated political community whose existence inexorably transforms the policy preferences of its member states. In particular, the hypothesis emphasizes complex learning provoked by argumentation between EU and member state officials involved in collective deliberations over competing policy options (Risse, 2000; Checkel, 2003, 2005). But in contrast to the dynamics of policy learning described above, suasion is pursued through the communication of normative reasons why particular member states should reconceive their identities and fundamental interests and thus their views on why particular policies are more desirable or more appropriate, particularly with reference to the target’s identity as an EU member. They are likely to be exchanged through bilateral channels between member states and within various Council formations such as CFSP working groups, the 133 Committee on trade policy, or Coreper. Member states may engage in normative suasion under a variety of circumstances. For example, when policy preferences are polarized within Council formations, representatives of both positions may exchange arguments in an effort to persuade the other. ‘Old’ member states may direct their arguments at ‘new’ member states in an attempt to socialize them to EU norms, or ‘new’ members may seek to overturn a long-standing policy consensus. A majority grouping of member states (or pre-Lisbon Treaty, the member state holding the Council presidency) may direct their arguments at whichever member states are

24 Explaining EU Foreign Policy

blocking consensus. But whatever the circumstance, successful normative suasion produces a convergence of member state preferences that facilitates consensus on EU common policies. In sum, the ‘Normative Suasion’ explanation for EU policy outcomes can be represented as follows: Divergent member state preferences + Normative arguments → Persuasion → Preference convergence → Consensus policy Although attempts at normative suasion are probably inevitable whenever EU member states disagree, these attempts will not always be successful. Research on socialization in international institutions identifies a number of conditions under which efforts at normative suasion are most likely to be successful. First, the target of the suasion attempt ‘is in a novel and uncertain environment and thus cognitively motivated to analyze new information’. Second, the target ‘has few prior, ingrained beliefs that are inconsistent’ with the reasons offered for redefining its interests or policy preferences. Third, the actor seeking to persuade ‘is an authoritative member of the ingroup to which the target belongs or wants to belong [and/or] does not lecture or demand but, instead, acts out principles of serious deliberative argument’. And fourth, the interaction ‘occurs in less politicized and more insulated, in-camera settings’ (Checkel, 2005: 813). All of these are potentially relevant to the various actors and institutional contexts involved in making EU foreign policy.

A final note on methodology Depending upon the character of the policy outcomes they exhibit, the various case studies face different explanatory challenges. If the policy outcome is best characterized as norm-governed consensus (H-1) or mutual compromise (H-2), the challenge is to demonstrate that there was no significant convergence of preferences and to explain why the potential veto player(s) failed to threaten a veto or were unwilling to execute their threat. If the policy outcome is best characterized as lowest common denominator (H-3), the challenge is to demonstrate that there was no significant convergence of preferences and to explain how the veto player(s) were able to dominate the negotiations and avoid making significant concessions or revising their preferences. Finally, if the policy outcome is best characterized as convergence of preferences (H-4 & H-5), the challenge is to highlight the domestic

Daniel C. Thomas 25

changes or intergovernmental interactions through which governments’ preferences converged.

Notes 1 Before the Treaty of Lisbon came into effect, CSDP was known as the European Security and Defense Policy (ESDP). 2 Summary figures calculated from ‘Actes Juridiques PESC: Liste Thématique’, Bruxelles, 31 décembre 2008, ACTES_JURIDIQUES_PESC_200109.pdf, accessed 1 July 2010. 3 Summary figures calculated from ‘CFSP Statements’,, accessed 1 July 2010. 4 This expectation has never been systematically investigated in EU foreign policy and external relations, but there is considerable evidence of cooperative bargaining among member states on other issues (e.g., Elgström and Jönsson, 2005; da Conceição-Heldt, 2006). 5 A mutual compromise is likely to be far from a lowest common denominator outcome, but it is not necessarily the same as a median compromise, which implies an agreement at the mid-point of actors’ preferences. 6 As such, post-facto references to pre-existing norms are not necessarily reliable evidence that agreement was achieved simply through entrapment. 7 Jeffrey Checkel (2001: 563, 2005: 813) makes a similar claim with regard to the pre-conditions for persuasion as a means of achieving norm compliance. 8 Although most Intergovernmentalists working on EU foreign policy have treated competitive bargaining and LCD outcomes as highly likely – which is why this hypothesis approximates the conventional wisdom – the logic of Intergovernmentalism does not necessitate this outcome. For further discussion of this point, see Schimmelfennig and Thomas’ concluding essay in this volume. 9 While some scholars insist that Constructivism is not designed to generate testable or generalizable hypotheses and is thus incompatible with social science, others argue that social science should not be equated with utilitarianism or rationalism. For the latter view, which informs this paper, see Ruggie (1998) and Price and Reus-Smit (1998).

Works cited Adler, E. (2002) ‘Constructivism and International Relations’, in W. Carlsnaes, T. Risse and B. Simmons (eds) Handbook of International Relations. London: Sage, pp.95–118. Aggestam, L. (2004) ‘Role Identity and the Europeanisation of Foreign Policy: A Political-Cultural Approach’, in B. Tonra and T. Christiansen (eds) Rethinking European Union Foreign Policy. Manchester: Manchester University Press, pp.81–98. Armstrong, K. and Bulmer, S. (1998) The Governance of the Single European Market. Manchester: Manchester University Press.

26 Explaining EU Foreign Policy Benford, R.D. and Snow, D.A. (2000) ‘Framing Processes and Social Movements: An Overview and Assessment’, Annual Review of Sociology, 26(1): 611–39. Checkel, J.T. (2001) ‘Why Comply? Social Learning and European Identity Change’, International Organization, 55(3): 553–88. Checkel, J.T. (2003) ‘“Going Native” in Europe? Theorizing Social Interaction in European Institutions’, Comparative Political Studies, 36(1–2): 209–32. Checkel, J.T. (2005) ‘International Institutions and Socialization in Europe: Introduction and Framework’, International Organization, 59(4): 801–26. da Conceição-Heldt, E. (2006) ‘Integrative and Distributive Bargaining in the EU: What Difference Does it Make?’, Negotiation Journal, 22(2): 145–65. Decaux, E. (2002) ‘Le processus de decision de la PESC: Vers une politique étrangère européenne?’, in E. Cannizaro (ed.) The European Union as an Actor in International Relations. The Hague: Kluwer Law International, pp.17–50. Elgström, O. and Jönsson, C. (2000) ‘Negotiation in the European Union: Bargaining or Problem-Solving?’, Journal of European Public Policy, 7(5): 684–704. Elgström, O. and Jönsson, C. (2005) European Union Negotiations: Processes, Networks and Institutions. London: Routledge. Franck, T.M. (2000) The Power of Legitimacy among Nations. Oxford: Oxford University Press. Garrett, G. and Tsebelis, G. (1996) ‘An Institutional Critique of Intergovernmentalism’, International Organization, 50(2): 269–99. Goldstein, J. and Keohane, R.O. (eds) (1993) Ideas and Foreign Policy. Ithaca: Cornell University Press. Hoffmann, S. (1966) ‘Obstinate or Obsolete? The Fate of the Nation State and the Case of Western Europe’, Daedalus, 95(3): 862–915. Hopmann, P.T. (1995) ‘Two Paradigms of Negotiation: Bargaining and Problem Solving’, Annals of the American Academy of Political and Social Science, 542: 24–47. Jørgensen, K.E. (1997) ‘PoCo: The Diplomatic Republic of Europe’, in K.E. Jørgensen (ed.) Reflective Approaches to European Governance. Basingstoke: Macmillan, pp.167–80. Jupille, J. (1999) ‘The European Union and International Outcomes’, International Organization, 53(2): 409–25. Jupille, J. and Caporaso, J.A. (1999) ‘Institutionalism and the European Union: Beyond International Relations and Comparative Politics’, Annual Review of Political Science, 2: 429–44. Keohane, R.O. (1986) ‘Reciprocity in International Relations’, International Organization, 40(1): 1–27. Koenig-Archibugi, M. (2004) ‘Explaining Government Preferences for Institutional Change in EU Foreign and Security Policy’, International Organization, 58(1): 137–74. Laitikainen, K.V. and Smith, K.M. (eds) (2006) The European Union at the United Nations: Intersecting Multilateralisms. Houndmills: Palgrave Macmillan. Lax, D.A. and Sebenius, J.K. (1986) The Manager as Negotiator: Bargaining for Cooperation and Competitive Gain. New York and London: The Free Press. Levy, S.J. (1994) ‘Learning and Foreign Policy: Sweeping a Conceptual Minefield’, International Organization, 48(2): 279–312. Lewis, J. (1998) ‘Is the “Hard Bargaining” Image of the Council Misleading? The Committee of Permanent Representatives and the Local Elections Directive’, Journal of Common Market Studies, 36(4): 479–504.

Daniel C. Thomas 27 Lewis, J. (2003) ‘Informal Integration and the Supranational Construction of the Council’, Journal of European Public Policy, 10(6): 996–1019. March, J.G. and Olsen, J.P. (1989) Rediscovering Institutions: The Organizational Basis of Politics. New York: Free Press. March, J.G. and Olsen, J.P. (1998) ‘The Institutional Dynamics of International Political Orders’, International Organization, 52(4): 943–69. Moravcsik, A. (1999) The Choice for Europe. Ithaca, NY: Cornell University Press. Meunier, S. (2000) ‘What Single Voice? European Institutions and EU-U.S. Trade Negotiations’, International Organization, 54(1): 103–35. Müller-Brandeck-Bocquet, G. (2002) ‘The New CFSP and CSDP Decision-Making System of the European Union’, European Foreign Affairs Review, 7(3): 257–82. Nuttall, S. (1992) European Political Co-operation. Oxford: Clarendon Press. Payne, R.A. (2001) ‘Persuasion, Frames and Norm Construction’, European Journal of International Relations, 7(1): 37–61. Peters, B.G. (1999) Institutional Theory in Political Science: The ‘New Institutionalism’. London and New York: Continuum. Pocar, F. (2002) ‘The Decision-Making Processes of the European Community in External Relations’, in E. Cannizaro (ed.) The European Union as an Actor in International Relations. The Hague: Kluwer Law International, pp.3–16. Price, R. and Reus-Smit, C. (1998) ‘Dangerous Liaisons? Critical International Theory and Constructivism’, European Journal of International Relations, 4(3): 259–94. Reynolds, P. (2004) ‘Analysis: Quest for EU Identity’. BBC News Online, 27 October,, accessed 1 July 2010. Risse, T. (2000) ‘Let’s Argue! Communicative Action in World Politics’, International Organization, 54(1): 1–39. Risse, T. (2004) ‘Social Constructivism and European Integration’, in A. Wiener and T. Diez (eds) European Integration Theory. Oxford: Oxford University Press, pp.159–76. Ruggie, J.G. (1998) ‘What Makes the World Hang Together? Neo-Utilitarianism and the Social Constructivist Challenge’, International Organization, 52(4): 855–85. Sandholtz, W. (1996) ‘Membership Matters: Limits of the Functional Approach to European Institutions’, Journal of Common Market Studies, 34(3): 404–29. Sandholtz, W. and Stone Sweet, A. (eds) (1998) European Integration and Supranational Governance. Oxford: Oxford University Press. Scharpf, F. (1988) ‘The Joint-Decision Trap: Lessons from German Federalism and European Integration’, Public Administration, 66(3): 239–78. Scharpf, F. (2006) ‘The Joint-Decision Trap Revisited’, Journal of Common Market Studies, 44(4): 845–64. Schimmelfennig, F. (2001) ‘The Community Trap: Liberal Norms, Rhetorical Action, and the Eastern Enlargement of the European Union’, International Organization, 55(1): 47–80. Schimmelfennig, F. (2003) ‘Strategic Action in a Community Environment: The Decision to Enlarge the European Union to the East’, Comparative Political Studies, 36(1–2): 156–83. Schimmelfennig, F. (2004) The EU, NATO and the Integration of Europe. Cambridge: Cambridge University Press.

28 Explaining EU Foreign Policy Smith, K.M. (2006) ‘Speaking with One Voice? European Union Co-ordination on Human Rights Issues at the United Nations’, Journal of Common Market Studies, 44(1): 113–37. Smith, M.E. (2004) Europe’s Foreign and Security Policy: The Institutionalization of Cooperation. Cambridge: Cambridge University Press. Stone Sweet, A., Sandholtz, W. and Fligstein, N. (eds) (2001) The Institutionalization of Europe. Oxford: Oxford University Press. Telò, M. (2009) The European Union and Global Governance. Oxford: Routledge. Tsebelis, G. (1994) ‘The Power of the European Parliament as a Conditional Agenda Setter’, American Political Science Review, 88(1): 128–42. Tsebelis, G. (2002) Veto Players: How Political Institutions Work. Princeton, NJ: Princeton University Press.

Case Studies

This page intentionally left blank

3 EU Policy on Ukraine During and Since the Orange Revolution: ‘A door neither closed nor open’ Richard Youngs1

Policy related to Ukraine’s ‘Orange Revolution’ is widely cited as one of the most notable recent successes in European foreign policy, in which member state governments and EU institutions acted in a united and concerted fashion in favour of democratic transition. This chapter seeks to explain how such unity arose, when prior to the latter stages of the revolution significant differences had persisted within the EU over the appropriate role that Europe should play in relation to Ukraine’s political evolution. In line with this book’s theoretical focus (Thomas, ch.2), it is suggested that a combination of ‘normative entrapment’ and ‘cooperative bargaining’ help us explain policy agreement in the case of Ukraine. The dynamics of rhetorical entrapment were particularly potent in explaining why a number of member states initially reluctant to back the Orange Revolution did eventually do so. At the same time, it is pointed out that such dynamics were enabled by very specific circumstances and that unity was also facilitated by other exogenous factors; in this sense, policy also exhibited elements of cooperative bargaining and reaction to changing geostrategic factors. Since 2004 member states have disagreed strongly over whether to offer Ukraine the prospect of membership. These disagreements widened after 2004 as Ukraine was beset by successive political crises that culminated in defeat for ‘pro-Orange’ candidates in the 2010 presidential elections. The Ukraine case can thus be divided into two separate analytical tests: one instance (the moment of democratic breakthrough) where agreement emerged; and one instance (the post-transition period) where the lowest common denominator has prevailed. In the aftermath of Ukraine’s democratic breakthrough, a picture more akin to competitive bargaining has returned in relation to differences that persist between member states over whether Ukraine should be offered the prospect of EU accession. As Ukraine chartered serious political crisis through 2007–10, this case 31

32 EU Policy on Ukraine

study demonstrates both the validity of but also limits to normative institutionalism in explaining EU foreign policy.

Case 1 – EU policy toward the Orange Revolution Ukraine’s Orange Revolution redressed the country’s gradual slide into semi-authoritarianism. Under Leonid Kuchma, Ukraine had become an emblematic case of electoral authoritarianism. Kuchma agreed not to stand for an unconstitutional third term in 2004, but his regime mobilized to manipulate the election run-off to declare victory for the president’s pro-Russian placeman, Viktor Yanukovich. It was at this stage that half a million protesters took to the streets and eventually forced the presidential election to be held again on December 26. Viktor Yushchenko emerged victorious from this poll, in one of the most dramatic democratic breakthroughs of recent years. However, progress towards democratic consolidation has faltered since 2005. A number of government coalitions have fallen. The rivalry between Yushchenko and fellow Orange leader, Julia Timoshenko has allowed Viktor Yanukovich’s Regions of Ukraine party to regain influence. Politics remains personalistic and democratic institutions weak. Prior to the drama of late 2004, EU policy towards the Kuchma regime constituted a shifting balance between engagement and pressure for democratic reform. In formal terms, the EU as a whole committed itself to supporting Ukraine’s democratization. In practice, European efforts to press for political change were limited in the face of Kuchma’s tightening hold on power. At the same time some member states preferred not to prioritize any significant deepening of relations with Ukraine. For France and Germany this was primarily due to the priority attached to relations with Russia. For states such as Spain and Italy it was due to a preference for deepening relations with the southern Mediterranean and a general paucity of interest in Ukraine. The EU signed a Partnership and Cooperation Agreement (PCA) with Ukraine in 1994, which included a commitment to support the development of democratic norms. This commitment was reiterated and made more explicit in the EU’s Common Strategy on Ukraine, adopted in 1999 (European Council, 1999). A number of political reform projects were funded under the EU’s TACIS aid program and a ‘legislative approximation scoreboard’ guided a program of cooperation aimed at harmonizing a swathe of Ukrainian legislation to EU norms and standards. In 2002, Ukraine adopted a formal ‘national program of approximation’ with EU legislation.

Richard Youngs 33

Such reform initiatives pursued through the European Commission appeared instrumentally to leave scope for national diplomacy that was in some cases more oriented towards maintaining alliances with Kuchma. Whether correctly or not, Kuchma continued to be seen by several European governments as providing a useful bridge to Moscow (Kubicek, 2003: 155). The caution exhibited by these governments was reflected in the balanced nature of overall EU policy. A membership prospect was not offered to Ukraine at the crucial meeting of the European Council in Helsinki in December 1999, when other central and east European states were formally recognized as candidates. EU documents and statements from the early 1990s routinely suggested that Ukraine was making progress towards democratic consolidation, when events on the ground suggested that Kuchma’s commitment to reform was increasingly doubtful. Kuchma himself was scathing of the EU’s limited offer to Ukraine, as he wanted a Europe Agreement and FTA (Solonenko, 2006: 45). European impatience with Kuchma’s stalling of long-promised reforms did increase, especially after it emerged that Kuchma had been implicated in the murder of critical journalist Goergyi Gongadze. However, the EU did not support the popular demonstrations that erupted in 2000. Nor did it offer material support for the democratic opposition that took shape in organized and systematic fashion after 2001. Indeed, at this stage most European states actually encouraged Ukrainian reformists still to focus on trying to join the government and gain moderate change from within the parameters of the regime. Rather, changes in European policy were more nuanced. By 2001, EU declarations became more critical, expressing ‘profound concerns’ over the tightening of political space and the intimidation of journalists, while suggesting that progress on economic and political reform was ‘a prerequisite for a deeper relationship with the EU’ (Kubicek, 2003: 162). At the same time, European governments sought strategically to respond to shifting alliances and trends within Ukraine. The 2002 elections, in which the opposition recorded a strong showing, had a notable impact on European calculations. High-level visits were reduced: only Gerhard Schröder met with Kuchma in 2003, and by early 2004, contacts at the most senior level had dried up. While so much focus was later centred on the dramatic events surrounding the elections in autumn 2004, underlying political conditions had begun shifting earlier that year. And, it was here that European responses were slow and cautious. Pora, the influential student group that led the civic action in the autumn, was formed in March 2004, but

34 EU Policy on Ukraine

attracted no open European governmental support. Even as the crucial defections of Kuchma’s supporters and state and security service insiders accelerated, some in the EU remained wedded to a strategy of encouraging ‘reform from within’ – long after most informed observers in Ukraine thought this was at all likely. Western training programs remained nonpartisan, and the Yushchenko campaign was funded entirely by Ukrainians (Karatnycky, 2006: 40). Manipulation in a key mayoral election in March 2004 caused a growing stir, just at the moment when the EU seemed obliviously to be concluding its ENP Action Plan with Kuchma. Despite all the EU’s rhetorical commitments in favour of Ukrainian democracy, as events gathered pace in the autumn of 2004 the EU initially appeared indecisive and reluctant to intervene. The EU was largely silent in response to pre-election intimidation that included the episode of Yushchenko’s disfiguring poisoning and threats made to students that they would lose their accommodation if they voted for Yushchenko (Karatnycky, 2005). Indeed, it was at this moment, when Ukrainian democrats were emphasizing the motivating force of their European vocation, that European Commission President Romano Prodi suggested that Ukraine had as much chance of joining the EU as New Zealand. Some liberal reformers complained bitterly at Europe’s reluctance to intervene as tensions deepened early in the autumn of 2004. Even as democracy protests erupted and the scale of electoral manipulation was revealed, one diplomat acknowledged that several EU member states remained reluctant to be seen supporting reformists ‘for Russia-handling reasons’. Nevertheless, as indicated above, the EU did eventually throw its weight clearly behind Ukraine’s democratic transition, calling for a rerun of the elections and making clear that future relations depended on the Kuchma regime standing aside to allow the Orange coalition to assume power. So, how did agreement over such support come about? Internal EU dynamics: From contestation to agreement When the Orange Revolution occurred, Europe’s support was widely seen as crucial and firm. It seemed a fairly straightforward case of European governments backing democracy. In fact, the EU’s decision to back the Orange Revolution was not taken without considerable internal debate. A number of states held preferences that were at a minimum highly ambivalent, until a very late stage in Ukraine’s democratic transition. With some risk of simplification, a group of ‘maximalist’ states (Poland, Lithuania, other new member states, with some backing from the Nordics and UK)

Richard Youngs 35

fought for the EU to push harder for Ukrainian democratization against a group of ‘minimalist’ states (including France, Germany, Spain and Italy). Beyond a united European expression of concern over the direction of events in Ukraine, member states’ advocated very different responses. A number of member states argued that the EU should respond to the changing conditions by offering Ukraine a deeper and democracyconditioned partnership, as a means of bolstering reformists. In 2002, the United Kingdom and Sweden first proposed offering Ukraine a further reaching set of relations, through what became the European Neighborhood Policy (ENP). Indeed, the ENP was seen by these states as a way of dealing specifically with ‘the Ukraine problem’ (Smith, 2005: 768). At the bilateral level, in 2002 Lithuania signed a new Strategic Partnership with Ukraine, also with a focus on political reform. Poland pressed for the EU to change its ‘Russia first’ policy to a ‘Ukraine first’ policy. For these states the ENP was seen as a means of tying down Kuchma to reform commitments, as trust between the EU and Ukraine seemed increasingly brittle. By mid-2004, as the dynamics of the transition took shape, internal European differences became most evident. Some states wanted to ‘help give events a push’. Others cautioned that a more hands-off approach was desirable and, according to many diplomats, fretted that support for reform was still not widespread enough to off-set the risks of being perceived as ‘interfering’ in Ukraine’s internal politics. Slovakia advocated a stronger ‘push’ in favour of regime change, with Slovakian civil society organizations finding in Ukraine’s predicament an echo of their own experience with president Meciar in Slovakia’s 1998 elections. The Poles pushed even harder at this stage for firmer conditionality, linked as quid pro quo to more generous EU incentives for Ukraine if democratic reform were allowed. At the other end of the spectrum were the southern EU member states. Crucially, Germany also remained cautious over a deeper engagement aimed at pushing political reform issues in Ukraine (Emerson, 2005: 18). German officials asserted that on-the-ground funding initiatives for Ukrainian democrats were hampered by the high-level politics of the Berlin-Moscow relationship. It was reported that Poland worked hard at pushing a reluctant German government into agreeing to offer Ukraine a deeper range of engagement, and was frustrated at its lack of significant success (Gromadzki et al., 2005: 31–2). Formal EU positions continued to balance and accommodate these differing national perspectives. The Commission’s Neighborhood Policy

36 EU Policy on Ukraine

Country Report on Ukraine, adopted in May 2004, warned that given the irregularities witnessed in the 2002 elections, the EU would monitor the 2004 poll more closely; but it also asserted that Ukraine was making a ‘progressive transition to democracy’ (Commission, 2004: 6 and 9). As stated, the EU did conclude negotiations for a Neighborhood Action Plan with the Kuchma government. The text of the Action Plan alluded to the priority of ‘Ensuring the democratic conduct of presidential (2004) and parliamentary (2006) elections in Ukraine in accordance with OSCE standards.’ And immediate implementation of the plan was halted in the run up to the presidential elections. It was at this stage that the most prominent role was adopted by Poland and Lithuania. These two states pressed for a more positive signal to be given towards Ukraine’s potential membership to the EU and for a tougher line towards Russian influence in Ukraine. In the autumn of 2004, Lithuania took the lead in initiating Council discussions on offering stronger relations with Ukraine. It was backed by six other new central and eastern European member states, the Nordic countries and Austria (Emerson, 2005: 17). This group of member states met frequently on an ad hoc basis immediately prior to the elections. Already in early November the Polish foreign minister switched a planned visit to Kuchma and prime minister Yanukovich to meet Yushchenko instead (Sushko and Prystayko, 2006: 131). In the midst of Ukraine’s brewing crisis, reports from insiders in Brussels indicated that EU foreign policy representative, Javier Solana, was initially reluctant to get involved. The more activist states complained at Solana’s passivity; Solana’s team were concerned that they lacked a clear mandate supported by all EU governments. One civil society representative lamented that Solana focused on events in Ukraine only after being pushed hard by Poland and when he belatedly saw ‘history being written’. A triumvirate was eventually assembled of Solana, Aleksander Kwasniewski and Valdas Adamkus, the Polish and Lithuanian presidents, respectively. There was general agreement that it was Kwasniewski who served as the crucial interlocutor, based on a long-standing mutual confidence with Kuchma. Solana was generally recognized as having played a valuable mediating role, while maintaining a line of, ‘we do not meddle, or take sides’. He actually called for demonstrators not to impede the working of government ministries – a call that was unceremoniously ignored by the crowd! (Sushko and Prystayko, 2006: 140). It is now well known that the observation mission for the re-run second round of the poll in December was the largest assembled in history. But in the first round, the number of EU observers was limited, with

Richard Youngs 37

only Poland and Slovakia sending significant numbers. Even after the first round, the Poles complained of German resistance to the idea of the EU intervening against Russia’s overt backing for Yanukovich. Joschka Fischer was the only German cabinet member immediately to demand a re-run. The French foreign minister placed his stress on the risks of destabilizing change spreading through the region more than on celebrating the events that had led Ukraine to the cusp of democratic breakthrough (Youngs, 2006). Indeed, interviews uncovered that the French government was particularly ambivalent and tardy in backing protestors’ claims that the second round results were fraudulent (Guillemoles, 2005: 132). As one prominent Ukrainian activist railed: ‘Chirac was our worst enemy, worse even than Putin’. While the British, Dutch and Swedish governments joined the US in funding exit polls, party training (offered on a bipartisan basis but taken up only by the opposition) and some indirect logistical support in-kind for prodemocracy protestors, Germany, Spain and France eschewed directly political aid projects in the run up to or in the wake of the first round. The role of quasi-independent party foundations such as the Westminster Foundation for Democracy, the German Stiftungen or the Dutch Alfred Mözer Foundation represented the more notable aspect of European political assistance. Despite all these differences, the EU did hammer out a common position in defence of free and fair elections and an effective change of regime. Despite their misgivings, the EU’s reluctant member states did eventually support an EU call for the elections to be re-run and monitored, which effectively handed power to the Orange coalition and ousted the ancien regime. After the contested and blatantly manipulated second round the EU did, according to one account, ‘change to a stick approach’ and threaten ‘serious consequences’ (Sushko and Prystayko, 2006: 132). Ukrainian experts indeed contrast this evolution in the EU stance with what was a clearer US backing for reformers from far earlier on in the process of democratic transition. France and Germany did send observers to the OSCE mission that monitored the re-run of the election, and began to suggest that the future of Ukraine’s relations with the EU would depend on a democratic outcome to the impasse. One network of NGOs opined favourably that these moves ‘demonstrated that the European Union really is capable of formulating and implementing a common foreign policy’ (ICPS, 2005). The EU united behind support for a ‘pacted’ solution, based on Yushchenko agreeing to cede some presidential powers to the parliament in order to placate Kuchma’s allies who would thus retain influence.

38 EU Policy on Ukraine

Somewhat contrary to subsequent impressions, European efforts focused more on such elite mediation rather than proactive support for the Orange coalition. The most tangible backing for democracy activists came not from Western official initiatives, but through links between Pora and its Serbian counterpart, Otpor (Pora was too high profile to receive either European or US funding) (Kuzio, 2005: 127). Explaining EU agreement on the Orange Revolution So how do we explain the fact that apparently firm differences gave way to agreement between member states at the moment of Ukraine’s democratic transition? The period running up to the Orange Revolution in 2004 provides much convincing evidence of normative entrapment. What was deemed to be appropriate behaviour had been conditioned by the EU’s series of pro-democracy commitments. The EU institutional setting itself militated in favour of eventually unified member state support for Ukraine’s democratic transition. At the same time, caveats and nuances to this strong case of normative institutionalism must not be overlooked. To some extent echoing policy deliberations in the case of Turkey, the discourse of democracy served as a convenient pretext: a number of European governments played up the importance of democracy-related criteria at a stage when these served to rebut Kuchma’s call for the EU to make clearer commitments towards Ukraine. This position was for the decade following Ukraine’s independence apparently able to reconcile a relatively non-specific aspiration in principle to assist political liberalization, with the strategic desire to play Ukraine as a bridge between Europe and Russia – as opposed to the country becoming a source of tension between Moscow and the West. For the EU the democracy commitment was both pull and push, serving to calibrate what was deemed the desirable mix between these competing dynamics. As the internal tensions of Ukraine’s ‘soft authoritarianism’ increasingly bubbled to the surface, however, the test became more actual of whether the EU would indeed proceed consistently with its many years of rhetorical commitment to Ukraine’s democratization. The impact of normative entrapment was made greater in the case of Ukraine due to the fact that the process of democratic transition here reached a very dramatic, ‘make or break’ denouement. With the Ukrainian constitution providing for free elections and other institutional aspects of democracy, the massive fraud that determined the officially proclaimed outcome of the initial run-off between Yanukovich and Yushchenko forced a clear moment of reckoning with the country’s own

Richard Youngs 39

formal constitutional guarantees. The very decisiveness of this moment de-legitimatized, and breached the limits of, the ‘reform from within’ line favoured by the more cautious EU member states, Javier Solana and some parts of the European Commission. It ensured that the dynamics of normative entrapment were strong enough to over-ride strategic preoccupations vis-à-vis relations with Russia. One illustrative example of this was seen in the case of German positions: Russian energy giant Gazprom was channeling huge amounts of funds into backing Yanukovich at the very moment when it was negotiating a gas deal with Berlin (Petrov and Ryabov, 2006: 150); if this sheds light on Germany’s initial reluctance unambiguously or actively to back Ukraine’s democrats, it also highlights that the Schröder government did eventually feel obliged to support democratic norms in a way that risked being uncomfortable for its short-term material interests. The competitive bargaining hypothesis would suggest that such factors would have pulled overall EU policy towards a lowest common denominator of studied ambivalence; yet, when Ukraine did reach its crisis point this did not happen. The importance of this context-dependency might be pushed a step further to argue that normative entrapment in this case only functioned against a background of shifting strategic calculation. Diplomats acknowledge that the crucial factor in the case of Ukraine in late 2004 was how fast and radically conditions changed on the ground. The unexpected scale and success of the Maidan protests left Western governments looking highly reactive in their policies, and clearly ran counter to the predictions of many EU member states. French and German positions only changed at the point when Russia itself concluded that defeat for the incumbent regime was inevitable, and when prudent self-interest required less hesitant backing for the likely next president, Viktor Yushchenko. Moreover, and in similar vein, even as normative entrapment appeared to have ‘done its work’, some degree of ‘diluting’ compromise was still necessary, in the form of the EU backing the mediated solution between Yushchenko and the Kuchma regime. Views on the deal struck with Kuchma/Yanukovich differed. Some saw it as both necessary and a means of guaranteeing against an over-bearing presidency in the future. But many civil society activists in Ukraine lamented that the EU ‘gave too much away’ in December 2004 to the Yanukovich camp, with reformists judging that it did so specifically in order to reach a negotiated position between France and Germany, on the one hand, and the new member states, the Nordics and the UK, on the other hand

40 EU Policy on Ukraine

(although even the latter group reasoned that some degree of ‘pacted exit’ was desirable). These negotiations provided a good example of cooperative bargaining: member states were not here negotiating ‘to the death’ over matters of direct and immediate self-interest, but did seek to have their own different positions reflected in the precise way in which Ukraine’s transition should be supported. A median line resulted that ensured Yuschenko assumed the presidency but under a quasi-pacted arrangement that some member states saw as necessary to ‘stabilize’ the process of democratization. All this suggests the need to add some precision to the theoretical framework offered in the opening chapter: in the case of EU policy towards the Orange Revolution normative entrapment was itself assisted by ‘strategic de-entrapment’. It did not flow simply from a conveyorbelt of incrementally accumulated EU commitments to democratic reform, but rather found expression when calculations of a more strategic nature began at least to some extent themselves to point in a similar direction. It is this latter element that cannot be satisfactorily captured solely by what are now seen as the mainstream dynamics in studies of CFSP, that is of constructivist identity-formation, communicative deliberation, incremental Europeanization, policy isomorphism etc (Wong, 2005: 151). The case of Ukraine reveals that while these perspectives provide necessary parts of the explanatory equation, it is also important to be attentive to how diplomacy evolves ‘in the real world’ against specific, shifting political backdrops. A dynamic applied of what might be termed strategically-conditioned normativism (Youngs, 2004).

Case 2 – Ukraine’s accession perspective Debates since the Orange Revolution have been dominated by the question of whether the EU should offer Ukraine the perspective of eventual membership to the European Union. No conversation with Ukrainian officials or civil society representatives takes place without the over-riding importance of this question being stressed. For members of the Orange movement, the possibility of EU membership was both their motivation and after 2004 represented the necessary anchor for Ukraine’s new, fragile democracy. However, on the question of Ukraine’s membership there has been no convergence of views within the EU; indeed, differences between member states have, if anything, widened since the Orange Revolution.

Richard Youngs 41

A similar line-up of maximalists and minimalists took shape. Poland, Hungary, Slovakia and Lithuania advocated that offer of membership be made immediately after December 2004, and argued that this was vital to underwrite Ukraine’s new democracy. A sizeable majority in the European Parliament also expressed itself in favour of offering Ukraine a membership prospect. In contrast, Germany, Spain, France, Belgium and the Netherlands opposed offering Ukraine even a distant prospect of membership (Stefan Batory Foundation, 2005: 13). Interestingly, the EP’s Spanish president Josep Borrell chose to reflect his country’s negative position rather than his institution’s advocacy of Ukrainian membership (Hansen, 2006: 28). Formal EU policy awkwardly bridged these contrasting positions. The Commission warned Yushchenko against lodging a formal application in the short term, so as to avoid a likely rebuff that might prejudice Ukraine’s longer-term membership chances. The standard line became that ‘the door is neither closed nor open’ – the logical impossibility of this metaphor reflecting the extent of internal EU divergence. Ukraine’s Neighbourhood Action Plan also offered a master class in opacity, to the extent that it ‘acknowledges Ukraine’s European aspirations and welcomes Ukraine’s European choice’, without specifying whether this constituted a ‘yes’ or a ‘no’ to the question of whether the offer of eventual membership would be put on the negotiating table. In view of sharp differences over the membership question, the European Commission presented itself a neutral arbiter and focused on means of strengthening the partnership with Ukraine, short of a membership offer. Aid was increased, progress made towards ‘deep free trade’ and cooperation deepened in a wide range of sectors. In 2008 the EU offered Ukraine an Association Agreement. The Ukrainian government complained that such measures were no substitute for an accession perspective. It argued that the range of EU cooperation now on offer was still not significantly greater than that which had been offered to Kuchma. From their perspective the European Neighborhood Policy was more of an affront than a fillip to democratic consolidation. Ukrainians came routinely to ask, ‘Why, as a European state and thus eligible to apply for EU membership, have we been lumped together with non-European states that have neither the possibility nor the vocation to seek accession?’ A commonly heard comparison was, ‘In terms of stability and democratic development we are well ahead of the Balkan states, and yet it is they and not us who have been offered membership.’ Government spokesmen summarized their puzzlement: ‘We are in Europe, not a neighbour to it.’

42 EU Policy on Ukraine

The rejection of the draft EU constitution by French and Dutch voters was attributed by many European politicians – in fact, without incontrovertible evidence and in some cases self-servingly – to a popular hostility to any further enlargement. After 2007, the Netherlands, a potential swing state between the two poles of internal EU opinion, no longer saw Ukraine’s potential membership as being on the agenda. The December 2006 European Council confirmed that stricter conditions would be imposed on future candidates. These new conditions left the EU able to refuse future applicants on grounds of concerns over its own (ill-defined) ‘absorption capacity’. In addition, member states had very different views regarding how the EU should respond to Ukraine’s increasingly faltering democratic consolidation. In the run-up to and in the aftermath of the March 2006 presidential elections, responses differed to the rise of Yanukovich’s Party of the Regions. Yanukovich’s return to power triggered firmer and more outspoken advocacy from eastern European member states for the EU now to offer membership. These states argued that the travails of Ukraine’s democratic consolidation had been aggravated by the weakness of EU policy since the end of 2004. Crucially, the UK also began to press in far stronger fashion in favour of the membership option – although in terms of an ‘open door’ being offered to Ukraine rather than a firm timetable being set for accession. One EU diplomat defined this as ‘a crucial shift’, bringing the first ‘old member state’ firmly into a leading role in pressing for a membership offer. In direct contrast, the German government of Angela Merkel expressed a firmer opposition to Ukraine’s membership. Instead, Germany sought to deflect pressure for further enlargement by devising a new ‘Neighbourhood Policy plus’ framework of intensified cooperation for Ukraine and other eastern European states. German foreign minister, Frank-Walter Steinmeier, explained: ‘In the EU we need attractive and credible offers for our neighbours’ (EU Observer, 2006). The new proposal was based on Ukraine being offered some EU acquis, under the banner of a ‘Partnership of Modernisation’. This was presented as a ‘New Ostpolitik’, that would crucially involve Russia as a partner in managing these eastern states, rather than as a ‘target’ state as proposed under the original ENP. It would preclude EU accession, now seen by Germany as, for all intents and purposes, off the agenda. This represented a middle position between the advocates of Ukrainian membership, on the one hand, and southern European states, on the other hand. The latter, led in particular by France and Spain, initially opposed any significantly enhanced partnership for Ukraine, arguing that

Richard Youngs 43

the Arab states of the southern Mediterranean merited greater priority for European foreign policy. Conversely, Germany joined with France and Spain to reject pressure from Poland and Lithuania for a new European Foundation for Democracy to focus specifically on supporting democrats in Ukraine and Belarus. Lithuania complained that even beyond the question of accession, other member states were increasingly cautious in offering Ukraine a deeper political partnership, in direct contrast to Russia’s concerted re-engagement with the country’s political elite. The Commission, insisting that it lacked any defined political objective of its own, continued to try to circumvent these internal differences by focusing on concrete forms of current cooperation. On 12 September 2006 the Commission proposed negotiating directives for a new ‘enhanced agreement’ with Ukraine that would include free trade and a deepening of thematic cooperation; it also accelerated work on a visa facilitation agreement. 494 million euros of aid were promised from the Commission for 2007–10, an average of 123 million per year compared to 70 million a year during 2004–06. But Ukrainian civil society groups complained that the EU had lost an opportunity to raise the prospect of membership prior to the March 2006 elections in a way that would have undercut support for the anti-Orange parties. The EU’s prevarication was used to good effect by Yanukovich in the elections, as he stressed a message to the electorate that: ‘You trusted the EU and they let you down.’ Germany, France and Spain even expressed a degree of relief in Yanukovich’s victory. Yanukovich returned Ukraine to the ‘two vectors’ foreign policy of the Kuchma era, in which relations with Europe would be balanced by a renewed strengthening of relations with Russia. The new prime minister argued that Ukraine must move from ‘Euro-romanticism’ to ‘Euro-pragmatism’. This would not involve a complete abandoning of Ukraine’s aim to join the EU (the Party of the Regions including a relatively pro-European strand of business oligarchs), but would reduce pressure for this option in the short to medium term. One diplomat revealed that these positions caused a palpable sense of relief amongst those states who wanted the membership debate ‘kicked into the long grass’. The change in perspectives was most succinctly epitomized in the statements of Commission president, José Manuel Barroso: in October 2005, Barroso had told the Ukrainian prime minister that ‘our door remains open. The future of Ukraine is in Europe’ (Hansen, 2006: 124) one year later, at the annual EU-Ukraine meeting in Finland on 27 October 2006, he dealt Yushchenko his most explicit rebuff to

44 EU Policy on Ukraine

date, insisting that ‘Ukraine is not ready, and we are not ready [for discussions on enlargement]’ (EU Observer, 27 October 2005). One French diplomat admitted that France even diluted the extent of free trade on offer in 2006 and that Paris was instrumental in preventing more ENPI funds going to Ukraine, as opposed to North Africa (Lefebvre, 2006: 18–22). Arguing a diametrically opposed logic, Poland pushed for better balancing between south and east in terms of ENPI allocations and was critical that the EU did not even offer the status of ‘association’, leaving the impression that after 2004 Ukraine was still ranked even behind the Arab states (Buras and Pomorska, 2006: 39). Polish diplomats admit that they failed in their attempt to boost EU offers to Ukraine and were angrily disappointed with other member states, this linking in to what became a broader clash between an assertive new Poland and some of the existing member states. Internal differences deepened as Ukraine’s political crisis became more acute and dramatic during 2007 and 2008. Yuschenko’s decision to recall parliament in April 2007 unleashed a bitter struggle with Yanukovich. The actions of both sides demonstrated the persisting limits to Ukrainian democracy, with key institutions such as the Supreme Court and parliament being ‘colonized’ by the two sides (Valasek, 2007). This struggle reached its latest stage in 2010 when Yanukovich won presidential elections, and moved quickly to centralize political control and sign a raft of new deals with Russia. In the midst of the 2007 crisis, Barroso warned Yushchenko in Brussels that the new enhanced agreement could now be delayed (EU Observer, 17 April 2007). But no firm responses were delineated for different possible outcomes to the crisis, due to divergence between member states. General admonishments were made only, exhorting compromise and solutions mutually acceptable to both sides. In early 2008 the French government proposed a ‘privileged partnership’ for Ukraine, which it presented as a more positive turn in its approach, but which was widely interpreted as a more definitive alternative to accession. At the NATO summit in Bucharest in April 2008 France and Germany blocked the offer of a Membership Action Plan to Ukraine; conversely the fact that they did agree to language stating that ‘Ukraine will become a member of NATO’ threw into sharper relief their refusal to accept the same in relation to EU membership. It was at this stage that the EU decided to ratchet up to the status of the new accord to that of an Association Agreement and also craft a new Eastern Partnership that would include Ukraine and five other states on the EU’s eastern borders. By 2009 the question of accession had been

Richard Youngs 45

kicked firmly into the long grass. And the EU’s uncertainty was further compounded by Yanukovich’s victory in the 2010 presidential elections. Explaining disunity on accession In the light of this disunity, the Ukraine demonstrates that rhetorical entrapment can be expected to function in relation to very tightly delineated issues and in very specific circumstances. Agreement between national preferences is more likely to occur where commitments have a clear normative imperative in terms of policy outcome; and where policy implications have previously been clearly spelled out. In the case of Ukraine’s membership prospect, these conditions did not pertain. Unlike in relation to other prospective candidates, the EU had since 1991 avoided making any concrete promise of membership, regardless of Ukraine’s future political evolution. And the promise to support the country’s democratization was sufficiently non-specific for France, Germany, Spain and other sceptical states to argue that this pledge was not necessarily betrayed by withholding the offer of accession. It was also an issue where there was no fashioning of convergence-facilitating trade offs – something that might in particular have bought Spain’s acquiescence to Ukrainian accession. As such imprecision weakened the foundations for rhetorical entrapment on this issue, policy remained determined by calculations of a more strategic nature. In particular, in the aftermath of the Orange Revolution considerations relating to Russia reasserted themselves as the prominent influence over European policies. One EU spokesman recognized that policy towards Ukraine was increasingly ‘caught up in debates over the best way to deal with Russia’. Most obviously, France and Germany were increasingly unwilling to collide with Vladimir Putin, keen to cement their Iraq-related alliance and with energy security considerations increasingly of concern. One Ukrainian civil society activist complained that some states in the EU ‘still see Ukraine as a kind of extension of Russia’ and this ‘has contributed to the recent negative developments in the country’. One Commission director admitted that concern over Russian energy supplies was the key reason why several member states were ‘slowing down’ on relations with Ukraine. The perception was widespread and strong amongst civil society groups that Germany was blocking a membership offer to Ukraine in the wake of its signing a bilateral deal with Russia for the development of the North European gas pipeline. One expert even argued that the setback of the March 2006 elections could prove positive if it taught Western

46 EU Policy on Ukraine

states to approach Ukraine anew, with Russia more firmly in mind (Lieven, 2006). Another analyst suggested that, compounding energy considerations, some member states feared that if the EU admitted Ukraine it would be harder to refuse Russia eventual entry (should Moscow one day seek this) (Larrabee, 2006). It was striking that amidst Ukraine’s 2007 crisis, many in the EU favoured a ‘negotiated’ solution between the three principals. French prime-minister Francois Fillon explained his government’s position at the April 2008 NATO summit pointing out that Ukraine’s NATO membership would disturb the balance of power between Russia and Europe (CEPS, 2008: 1). Other member states were equally exercised with Russia, but argued that such concerns had exactly the opposite implication for policy towards Ukraine. The positions of the new central and east European member states might have reflected an ideational association with Ukraine’s democrats (Hansen, 2006: 125); however, while these states deployed the ‘values’ argument they also insisted in increasingly strategic terms that the EU needed to prize Ukraine away from Russia’s sphere of influence (Dannreuther, 2006). To the extent that this had not unequivocally happened solely as a result of the Orange Revolution, they argued, the offer of EU membership was increasingly desirable and urgent. Indeed, for new member states the choice firmly to back the Orange Revolution was as much about ‘standing up to Russia’ as it was about Ukrainian democracy itself (Sushko and Prystayko, 2006: 130). Indeed, some experts detected that such strategic impulses behind the advocacy of Ukrainian membership were set to become increasingly strong (Garton Ash, 2005: 2). This strand of reasoning was nourished by Yushchenko intimating that a membership perspective would come hand in hand with firmer Ukrainian backing for independent (i.e. from Russia) gas pipelines from Central Asia into Poland and Austria. Eastern European member states pointed to the inconsistency in French positions: Paris had insisted on a January 2007 entry date for its client, Romania, regardless of the latter’s record on reform; this revealed how France’s new keenness to focus on Ukraine’s democratic imperfections as a barrier to accession was more to do with geopolitics than ‘constructed’ normative purity. Lastly, if strategically-rooted discordance appeared to be deepening by 2007, post-Orange Revolution Ukraine also revealed the potentially divergent impact of internal EU dynamics. The above account highlights the extent to which the crisis over the European constitution conditioned debates on policy towards enlargement, including the specific case of Ukraine. One EU diplomat acknowledged, in a senti-

Richard Youngs 47

ment shared widely amongst policy-makers: ‘If the Orange revolution had occurred five or six years earlier, Ukraine would be a member by now’. Polls conducted after the French and Dutch referendums showed that in France a strong majority existed against Ukrainian accession, while in many other member states a clear majority favoured the EU making an accession offer to Kyiv (Pew, 2006). One diplomat suggested that the issue of accession was ‘symbolically’ important for the ‘sceptics’ and ‘enthusiasts’ in apparently irreconcilable terms: for eastern European member states and Ukraine the importance was in the symbolism of a membership offer being made, as a spur to Ukrainian democrats; for the sceptical states, the symbolic importance vis-à-vis their own publics was in not making further accession promises. In short, debates over this specific case were increasingly caught up in the broader, thorny relationship between the EU’s internal and external legitimacy. Diplomats agreed that by 2007–08 internal conditioners were as potent as external factors.

Conclusion In short, the case of Ukraine’s democratic transition offers a mixed picture in terms of the hypotheses that this volume has set out to explore. The two analytical test cases that have been separated out here – the moment of transition versus the post-transition period – offer different conclusions. Together they suggest that the pertinence of normative institutionalism can vary even within a single area of policy and depends on the nature of the decision(s) being contemplated. In the moment of transition much normative entrapment was in evidence, mixed with a dose of cooperative bargaining over the details of exactly how Europe would support the difficult process of regime change. This was normative entrapment, as explained, in the sense that many member states had constantly made a rhetorical commitment to support Ukraine’s democratization without ever contemplating that this would involve proactive EU action in a moment of dramatic revolutionary change. The nature of the decision was a dramatic one, and not easily fudged. In the wake of the Orange Revolution, outcomes tending more towards the lowest common denominators associated with more competitive inter-state bargaining have been present in EU policy towards Ukraine, mainly in relation to the issue of Ukraine’s membership prospects. The normative dimensions of the accession question are far more open to contestation as to what would really be a ‘normative’ EU response, and thus so far reluctant member states have avoided being ‘entrapped’.

48 EU Policy on Ukraine

In addition, this chapter shows that both at and after the moment of democratic transition broader geostrategic factors provided additional variables that set the context within which the dynamics of normative institutionalism were – or were not – played out. Ukraine might be summed up as a case towards which normative institutionalism can explain much, but not the entirety of internal EU bargaining and negotiation.

Note 1 The author wishes to thank Jos Boonstra, Natalya Shapovalova, Balazs Jarabik and colleagues at the preparatory workshops to this volume for their input, along with the diplomats interviewed in Kyiv and Brussels over various trips during 2005, 2006, 2007 and 2008.

Works cited Buras, P. and Pomorska, K. (2006) ‘Poland and the European Neighbourhood Policy’, in Foreign Policy Dialogue 7/19, July ( CEPS (2008) Neighbourhood Watch, Issue 37, April. Dannreuther, R. (2006) ‘Developing the Alternative to Enlargement: The European Neighbourhood Policy’, European Foreign Affairs Review, 11/2, pp.183–201. Emerson, M. (2005) ‘The Reluctant Debutante’, in M. Emerson (ed.) Democratisation in the European Neighbourhood. Brussels: Centre for European Policy Studies. European Council (1999) ‘Common Strategy of December 11, 1999 on Ukraine’, 1999/877/CFSP, Official Journal of the European Communities, L331, no. 1, December 23, 1999. Garton Ash, T. (2005) Free World. London: Penguin. Gromadzki, G., Lopata, R. and Raik, K. (2005) ‘Friends of Family? Finnish, Lithuanian and Polish Perspectives on the EU’s Policy Towards Ukraine, Belarus and Moldova’, FIIA Report 12. Guillemoles, A. (2005) Même la Neige etait Orange: La révolution ukrainienne. Paris: les Petits Matins. Hansen, F. (2006) ‘The EU and Ukraine: Rhetorical Entrapment’, European Security, 15/2. ICPS Newsletter, ‘European Union Should Open Door to Ukraine Membership, says PASOS’, no. 1(260), 17 January 2005. Karatnycky, A. (2005) ‘Ukraine’s Orange Revolution’, Foreign Affairs, 84(2). Karatnycky, A. (2006) ‘The Fall and Rise of Ukraine’s Political Opposition: From Kuchmagate to the Orange Revolution’, in A. Aslund and M. McFaul (eds) Revolution in Orange. Washington: Carnegie Endowment for International Peace. Kubicek, P. (2003) ‘The European Union and Ukraine: Real Partners or Relationship of Convenience’, in P. Kubicek (ed.) The European Union and Democratization. London: Routledge. Kuzio, T. (2005) ‘The Opposition’s Road to Success’, Journal of Democracy, 16(2). Larrabee, F.S. (2006) ‘Ukraine and the West’, Survival, 48/1, 93–110.

Richard Youngs 49 Lefebvre, M. (2006) ‘France and the European Neighbourhood Policy’, in Foreign Policy Dialogue, 7/19, July ( Lieven, A. (2006) ‘Failure of Orange Revolution is a Historic Opportunity’, Financial Times, 25 July. Petrov, N. and Ryabov, A. (2006) ‘Russia’s Role in the Orange Revolution’, in Aslund and McFaul, op. cit. Pew Global Attitudes Survey, July 2006. Valasek, T. (2007) ‘Ukraine’s Real Problem’, CER Bulletin, April/May, Issue 53. Smith, K. (2005) ‘The Outsiders: The European Neighbourhood Policy’, International Affairs, 81(4). Solonenko, I. (2006) ‘European Neighbourhood Policy – The Perception of Ukraine’, in Foreign Policy Dialogue, 7/19, July ( Stefan Batory Foundation (2005) ‘Will the Orange Revolution Bear Fruit? EUUkraine Relations in 2005 and the Beginning of 2006’. Warsaw: Stefan Batory Foundation. Sushko, O. and Prystayko, O. (2006) ‘Western Influence’, in Aslund and McFaul, op. cit. Wong, R. (2005) ‘The Europeanisation of Foreign Policy’, in C. Hill and M. Smith (eds) International Relations and the European Union. Oxford: Oxford University Press Youngs, R. (2004) ‘Normative Dynamics and Strategic Interests in Europe’s External Identity’, Journal of Common Market Studies. Youngs, R. (ed.) (2006) Survey of European Democracy Promotion Policies 2000–2006. Madrid: FRIDE.

4 EU Policy on the International Criminal Court: Institutional Contexts and Policy Compromises Daniel C. Thomas1

One of the most politically sensitive issues on the European Union’s foreign policy agenda in recent years has been the question of how to respond to the United States’ campaign for immunity from the International Criminal Court (ICC). When the US government set out in early 2002 to shield its officials, citizens and other employees from the jurisdiction of the new Court, the European Union (EU) and its member states were forced to weigh their oft-stated commitment to an international order based upon multilateral institutions, the rule of law, and respect for human rights, which underlay their commitment to the ICC, against their commitment to maintaining good relations with world’s only superpower, with which they share many common interests, similar political values, and a long history of cooperation. Given these stakes, it is not surprising that EU member states did not see eye-to-eye on how to respond to the US effort: some thought it deserved a cooperative response while others considered it an unacceptable assault on the integrity of the Court. These differences were evident in negotiations among member states in both forums where they encountered the US campaign for ICC immunity – within the United Nations (UN) and in bilateral relations with Washington. This paper examines the making of EU policy toward the ICC in the context of transatlantic relations as a test of competing explanations for how member states negotiate divergent preferences in order to reach agreement on common policies. The conventional wisdom, also known as Intergovernmentalism, asserts that EU foreign policy emerges from the no-holds-barred pursuit of national policy preferences by member states. Common policies are thus made (or denied) through a process of competitive bargaining that can yield only deadlock or lowest common denominator outcomes. In contrast, Normative Institutionalism 50

Daniel C. Thomas 51

(Thomas, ch.2, this volume) asserts that member states pursue their policy preferences within an institutionalized setting that encourages certain negotiating practices and legitimates certain substantive outcomes while discouraging and delegitimating others. Accordingly, the theory leads us to expect either rhetorical entrapment resulting in an outcome consistent with prior normative or policy commitments or cooperative bargaining resulting in mutual compromise. The paper also considers the possibility that agreement is achieved when member state preferences converge, whether due to normative suasion or the exchange of information. To this end, the paper employs two methodologies. First, in order to determine which theory is more consistent with the character and the outcome of intra-EU negotiations, it utilizes the optimal method for assessing the reasoning and behaviour of deliberate actors within socially-constructed settings – process tracing (Checkel, 2006). It also utilizes the method of controlled or structured-focused comparison (George, 1979), taking advantage of the fact that member states encountered the US campaign for ICC immunity on two fronts and thus negotiated among themselves in two distinct settings: beyond the institutions of the Union in the first case, but within them in the second case. Although the two cases are sequential and thus not fully independent, the latter method is justified by similarities in the external challenge faced by the member states, by similarities in the pre-existing policy and normative commitments at the EU level, and by the fact that member states ultimately responded in unison to the US campaign in both cases, despite significant differences in their policy preferences. Comparing the two cases thus allows us to assess Normative Institutionalism’s expectation that the EU’s substantive and procedural norms are more likely to shape negotiations among member states that occur within EU forums than those that occur outside it (see also Puetter and Wiener, 2007: 1072). Comparing evidence of the policy preferences of the various member states to the common policies that they agreed in both cases, all in the light of considerable evidence regarding the negotiation processes, we can draw even better informed conclusions about when and how the EU manages to avoid succumbing to the dynamics of the lowest common denominator (LCD). If the Normative Institutionalist theory is strong, we would at a minimum expect to observe cooperative bargaining and mutual compromise in the second case, which involved policy-making within the various formal and informal institutions of the Common Foreign and Security Policy (CFSP). If we observe competitive bargaining

52 EU Policy on the International Criminal Court

and a LCD outcome in both cases, then the explanatory power of Normative Institutionalism would be greatly undermined. If we observe competitive bargaining and LCD only in the first case, we could reasonably attribute this to the fact that it involved cooperation among member states outside the context of EU institutions, where the dynamics and outcomes expected by the Normative Institutionalist theory are less likely to be found. Significant convergence of member state preferences during the course of the negotiations would challenge both of these theories. The remainder of this paper is organized in four parts. The first part outlines the origins of the EU-US dispute over the International Criminal Court. The second and third parts present case studies of how EU member states responded to the American campaign for ICC immunity in different institutional settings: the first examines decision-making with regard to US initiatives within the UN Security Council, and the second examines the EU’s collective response to the US quest for bilateral agreements. Finally, the fourth part evaluates the findings of the two case studies in light of the project’s theoretical framework.

Background The ICC is the world’s first permanent court designed to ensure individual accountability for the most serious crimes under international law: genocide, war crimes, and crimes against humanity. It was established by the Rome Statute of 1998, which entered into force on 1 July 2002. In addition to the list of crimes, the ICC’s jurisdiction is limited in a number of ways, including inter alia: the ICC may act only when national courts are ‘unwilling or unable genuinely to carry out the investigation or prosecution’ (the principle of complementary jurisdiction); and the UN Security Council, acting under Chapter VII of the Charter, may defer cases for 12 months, renewable indefinitely. The European Union has long supported the creation and functioning of the ICC, which is widely seen in Europe as a desirable (if limited) solution to the problem of impunity for international crimes. A number of EU member states played a central role in the ‘group of like-minded states’ that pushed for a strong and effective ICC during the Rome negotiations (Benedetti and Washburn, 1999). Every member state then signed and ratified the Rome Statute, and most quickly adopted the necessary implementing legislation, generally with strong support from both governing and opposition parties. During the time period covered by this paper, eight of the 15 member states contributed funds to the nongovernmental Coalition for the ICC (Patten, 2002).

Daniel C. Thomas 53

Echoing this sentiment, EU Commissioner for External Relations Chris Patten declared in 2002, ‘In the twenty-first century, potential tyrants and mass murderers will know in advance that the international community is prepared to hold them accountable for massive violations of human life and dignity. It is our belief and our hope that this awareness will help reduce the frequency and the severity of such crimes. But when it does not, and the relevant national legal authorities are unwilling or unable to act, the international community will have in place a complementary system of criminal justice that is fair, transparent and effective.’ Between 1995 and 2003, the European Commission’s European Initiative for Democracy and Human Rights provided approximately €13 million to the Advance Team setting up the ICC and to the NGO Coalition for the ICC and other nongovernmental initiatives promoting the Court (Patten, 2002). In contrast, the United States has never embraced the ICC. Although American negotiators achieved many of their objectives during the Rome negotiations (including the principle of complementary jurisdiction and the right of the Security Council to defer cases), the US was one of seven countries to vote against the Statute in 1998. The following year, the US Congress passed a law prohibiting the use of government funds to support the ICC, denying legal effect to ICC jurisdiction within the United States unless the US becomes a party to the Rome Statute, and barring extraditions of US citizens to third countries unless they agree not to surrender them to the ICC. President Clinton signed the Rome Statute three weeks before leaving office, but told the Senate that he did not recommend ratification until further changes were made (Clinton, 2001). With the transition to the Bush Administration, US ambivalence toward the Court was replaced by rejectionism. In May 2001, Representative Tom Delay introduced in Congress the American Servicemembers’ Protection Act (ASPA), which would prohibit cooperation with the ICC by any US agency or official, require the US to pursue ICC exemption for US participants in UN peacekeeping missions, prevent the transfer of US intelligence information to the ICC, and prohibit US military aid to any states party to the ICC. The ASPA was seen in Europe as a legislative assault on the good functioning of the ICC; its authorization of ‘all necessary means’ to obtain the release of any American held by the ICC led European critics to call it the ‘Hague Invasion Act’. Faced with such a clear challenge to an institution that it supported, and absent any significant dissension among member states, the EU

54 EU Policy on the International Criminal Court

had little trouble achieving a unanimous response. In June 2001, the EU adopted its first Common Position on the ICC, which began, ‘The principles of the Rome Statute of the International Criminal Court, as well as those governing its functions, are fully in line with the principles and objectives of the Union.’ The Common Position also expressed the EU’s full support for the early establishment of the Court, indicated that the EU and its member states would encourage other states and international organizations to support the ICC, and encouraged the US to cooperate with the ICC as well.2 In the spring of 2002, as it became apparent that the 60 ratifications necessary for the Rome Statute to enter into force would soon be achieved, the Bush Administration launched a diplomatic campaign to ensure that the ICC would not affect Americans or the US government. On 11 April, the US boycotted a special ceremony at the UN for the ten countries whose ratification of the Statute would make the ICC a reality.3 Several weeks later, the US government announced that it did not intend to ratify the Statute and thus did not consider itself bound to comply with its provisions. In addition, senior administration officials expressed a strong commitment to ensuring that US officials and military personnel would never be subject to the new Court.4 In response, EU member states agreed unanimously within CFSP for a declaration that criticized Washington’s repudiation of the Rome Statute as unwarranted and damaging to international law and reiterated the EU’s commitment to ending impunity for individuals who commit the most serious international crimes.5 As the ASPA legislation moved through the US Congress, Europeans expressed growing concern about its authorization of ‘all necessary means’ to obtain the release of Americans held by the ICC. A London newspaper asked if Bush would invade England.6 The Dutch parliament passed a resolution expressing its concern about the ASPA bill, which it considered ‘detrimental to transatlantic relations’.7 This concern was soon echoed at the EU level. On 17 June, the EU Council adopted Conclusions criticizing the ASPA legislation. Three days later, the EU adopted an Action Plan detailing its initiatives in support of the ICC, as well as a revised version of its Common Position on the ICC emphasizing the diplomatic and technical assistance it would provide to support the Court. On 23 June, the EU Presidency and the Commission expressed concern about the APSA in joint letters delivered to the US Senate and House of Representatives. Less than two weeks later, the European Parliament added its voice to this chorus of disapproval.8

Daniel C. Thomas 55

Up to this point, the course of the EU-US dispute over the ICC reflects the EU’s easy adoption of common foreign policies when member states’ preferences are similar or identical. However, the real challenge for EU foreign policy cooperation, and thus the more interesting test for the various theories, is policy-making in the face of clearly divergent member state preferences. The following analyses of the next two stages in the EU-US dispute over the ICC constitute case studies of this sort. Case 1 – ICC immunity for peacekeepers The EU maintained its unity when Washington first took its anti-ICC campaign to the UN in May 2002. Barely a week after renouncing its signature of the Rome Statute, the US threatened to veto renewal of the UN mission to Timor Leste if the Security Council did not grant ICC immunity to international military and civilian personnel serving there. Although the US presence in Timor Leste was only three military observers and about 80 police officers, a veto of the entire UN mission would have jeopardized stability on the newly-independent island. Just as important for this analysis, the US veto threat forced EU member states to weigh their collective commitment to the ICC against their collective commitment to support the United Nations and multilateral solutions to global problems. As such, the relevant normative framework for the behaviour of EU member states at the UN was not clear. France and the United Kingdom lobbied against the US proposal. After failing for several days to gain the support of any other Security Council member, Washington set aside its threatened veto.9 Six weeks later, though, on the same day that the Rome Statute entered into force, the US again signaled its opposition to the ICC by simultaneously withdrawing its three soldiers from Timor Leste and vetoing an extension of the mandate for the UN Mission in Bosnia-Herzegovina. In Europe, the American moves were seen as a challenge to the EU’s commitment to defend the integrity of the Rome Statute, and to its concrete interest in the security and stability of the Balkans.10 The US then proposed two alternative resolutions: one would have extended immunity only to peacekeepers deployed in Bosnia, while the second would have covered all peacekeepers involved in operations authorized or mandated by the UN. France’s UN ambassador responded that while the Statute allowed the Security Council to defer ICC investigations on a case-by-case basis when they would interfere with efforts to negotiate a peace agreement, it was never designed to provide sweeping immunity from prosecution.11 France, the UK, and Ireland (then a

56 EU Policy on the International Criminal Court

rotating member of the Security Council) rejected both American proposals. Meanwhile, fearing that the Security Council standoff could threaten the continuity of the UN’s mission in Bosnia, EU officials began preparing to replace it with EU forces, while Britain’s UN ambassador drafted a resolution that would authorize the hand-over.12 EU Commissioner Chris Patten again pleaded publicly for the US to reconsider its policy, which he said threatened international security and contradicted longstanding American support for human rights.13 This time, though, the Bush Administration did not relent in its quest for a UN Security Council resolution that would grant ICC immunity to UN peacekeepers. The difficulty for the EU was that while all its member states opposed the US campaign, they did not all agree on how to respond. Ireland (whose population had just endorsed the Rome Statute by 2:1 in a national referendum) favoured an unequivocal rejection of Washington’s demand, which it believed would compromise both the UN Charter and the Rome Statute.14 On the other hand, while Tony Blair’s government had always been a strong supporter of the ICC, it began to reinterpret that position to accommodate its ‘special relationship’ with Washington. The US initiative also placed France in an awkward position: despite having signed and ratified the Rome Statute, France was the sole EU member state to have taken advantage of Article 124, which permits signatory states to exempt their nationals from the ICC’s jurisdiction over war crimes (but not genocide or crimes against humanity) for a seven-year period beginning with the Statute’s entry into force. Given that this stage of the dispute concerned votes in the Security Council, where no other EU members had seats, consultations between France, Ireland and the UK did not lead to discussion of a common EU policy within CFSP. As a result, supranational actors were marginalized and NGOs had little opportunity to mobilize. With the UK expressing sympathy for the US’s legal claim that as a non-Party to the ICC it should not be subject to the Court’s jurisdiction, its position clearly diverged from that of France and Ireland. (Although not on the Security Council, Germany was outspoken in its opposition to US proposals.15) When Denmark’s representative addressed an open meeting of the Security Council on behalf of the EU presidency, he could do little more than express regret regarding the risk to UN peacekeeping and echo the EU’s general support for the ICC that had been unanimously agreed the previous year.16 EU coherence and consensus norms that might have led the three to adopt a policy premised upon those pre-existing EU common positions

Daniel C. Thomas 57

or at least promoted mutual compromise, were rendered virtually irrelevant by the fact that the UK had special consensus-building responsibilities as president of the Security Council.17 As a result, the UK tabled an alternative to the US resolution that would limit ICC immunity for UN peacekeepers to one year and to the armed forces of States that were not Party to the Rome Statute.18 This conceded more to the US than France and Ireland would have preferred, but it fell far short of the broad and permanent immunity sought by Washington.19 Anxious to protect UN peacekeeping, France and finally Ireland agreed not to oppose the British proposal, the de facto lowest common denominator among the EU 3. Faced with European unity in opposition to its full demands, Washington accepted the British proposal as well. On 12 July, the Security Council voted 12–0 in favour of Resolution 1422, giving peacekeepers from non-ICC party states immunity from the Court through 30 June 2003. Case 2 – Bilateral agreements on ICC immunity While the negotiations over the Security Council resolution continued in June 2002, US Secretary of Defense Donald Rumsfeld announced that Washington also intended to pursue bilateral agreements with governments around the world to protect US citizens, government officials, military personnel and other employees from surrender to the ICC.20 Speaking off the record, members of the administration conceded that their real goal was to protect senior US officials, whom they considered to be most at risk.21 Such agreements, US officials argued, would be consistent with Article 98 of the Rome Statute, which addresses potential inconsistencies between obligations created by the Statute and those created by other international agreements. US officials thus typically referred to them as ‘Article 98 agreements’ or ‘non-surrender agreements’. In contrast, most Europeans and most ICC experts (including David Scheffer, the chief American negotiator of the Rome Statute) argued that Article 98 had been crafted to accommodate Status of Forces agreements and extradition treaties that were already in effect when the Statute entered into force, but not to facilitate new agreements to limit the jurisdiction of the new Court (Scheffer, 2002). They began referring to the US pursuit of ‘immunity agreements’ or even ‘impunity agreements’. Most EU leaders saw the proposed bilateral agreements as the latest US challenge to the integrity of the Rome Statute and thus to the good functioning of the ICC itself. Unlike the UN resolution, these agreements fell clearly within the policy-making scope of the CFSP. However,

58 EU Policy on the International Criminal Court

more was at stake than simply the EU’s commitment to the ICC. Denying the US request could provoke a further confrontation with Europe’s most important ally and the world’s only superpower. In addition, there was real tension between the EU’s specific commitment to support the ICC and its general commitment to international law, which traditionally limits the jurisdiction of international treaties to states that have accepted them. Finally, the issue engaged the long-standing division between the EU’s more ‘Atlanticist’ and ‘Europeanist’ member states. As the weeks passed, the ICC immunity issue began to overlap with the controversial diplomatic and then military build-up to the invasion of Iraq, over which the EU was increasingly divided. A common EU policy on ICC immunity agreements for the United States in mid– late 2002 was thus no easy matter for CFSP. In response to direct overtures from Washington, Berlin announced on July 23 that it could not consider US requests for bilateral agreements until the entire EU had deliberated on a common policy. This announcement, which effectively set the CFSP process in motion, was a major setback for the US, which had hoped that a positive response from Germany would set an example across Europe.22 The EU’s Political and Security Committee (PSC) discussed the matter three days later and agreed to inform Washington that the EU remained committed to its Common Position to support the ICC and to preserve the integrity of the Rome Statute and that it could not reply further to the US request until it completed an in-depth analysis.23 This procedural agreement reinforced the salience of the existing Common Position as precedent for the issue of bilateral agreements. In early August, the European Commission got involved. In its role as the EU’s negotiator on enlargement, the Commission had been asked by candidate member states whether the EU would consider a bilateral agreement with Washington to be consistent with the requirements of the Rome Statute. Upon the request of Commission President Romano Prodi’s office, the Commission’s Legal Service prepared a written opinion on 13 August stating that such agreements were inconsistent with the Rome Statute and with the commitment to support the effective functioning of the ICC that the EU had repeatedly agreed and expressed: ‘The Common Position is an expression of the member states’ strong commitment to ensure full effectiveness of the Court. In light of the above analysis it is inconceivable to reconcile that commitment with any attempt to give positive consideration to the U.S. proposal.’ Based on this opinion, which was welcomed by Prodi’s office, the Commission strongly advised candidate states that they should refuse to sign a bilateral agreement with the US.24

Daniel C. Thomas 59

The Commission opinion was also circulated to member states through the EU’s confidential Coreu message system. All recognized that the opinion was potentially of great significance. If the EU were to translate its logic directly into a common policy, none of the 15 member states – all of whom had ratified the Statute – would have been able to reach bilateral agreements with the US. This in turn would have undermined Washington’s efforts to convince other states around the world to sign such agreements. Many of the EU member states were already sympathetic to the position expressed by the Commission, but others appreciated the clarity of its argument on this emerging issue. Another factor was the automatic sympathy that the smaller member states have for the Commission’s Legal Service, which they tend to regard as protective of their interests through its commitment to European and international law. However, the member states were not unanimous in this view. Perhaps because some were fearful of the opinion’s potential consequences for the issue at hand, or perhaps simply because they resented the Commission’s assertiveness on a CFSP issue, they told the Commission in no uncertain terms that it had overstepped its competence. Although the Commission responded by keeping a low profile during the CFSP deliberations over the next six weeks, its Legal Service’s opinion remained influential.25 The US government also learned of the Commission opinion, which contributed to its growing awareness that opposition to the quest for bilateral agreements was mounting in Europe. Three days after the Commission opinion was released, US Secretary of State Colin Powell wrote to a number of his European counterparts recommending that they not wait for a common EU position and threatening that failure to sign bilateral agreements could have negative consequences for the United States’ role in European security.26 Powell’s demarche was seen in EU capitals as heavy-handed interference, and stiffened the resolve of some member states to resist American overtures.27 On the other hand, Washington’s anti-ICC campaign succeeded in its goal of forcing EU member states to balance their commitment to support the Rome Statute against their desire to avoid a major transAtlantic dispute. As one EU diplomat explained, ‘We are coming under huge pressure from Washington. All it takes is one member state to give in and it undermines the credibility of the ICC.’28 Given that the EU had just updated its Common Position on the ICC two months earlier, such a failure of EU unity would undermine the credibility of the CFSP as well. By late August, only Israel, Romania, Timor Leste and Tajikistan had signed bilateral agreements with the US. But given the global spread of

60 EU Policy on the International Criminal Court

Washington’s campaign on this issue, it seemed to be only a matter of time before other governments signed as well. The formerly Communist states of Central and Eastern Europe, which wanted to become members of both NATO and the EU, found themselves caught between American and European preferences. ‘I can’t remember anything they put so much weight or interest into’, Romania’s foreign minister said of the US campaign.29 In an effort to forestall this eventuality, the EU presidency instructed all EU candidate states (including Romania, whose Parliament had not yet ratified the government’s signature) that it disapproved of signing bilateral accords in the absence of an agreed EU policy. The existence and content of the Commission’s legal opinion leaked to the press at the end of August, just two days before the EU foreign ministers were scheduled to discuss the issue.30 With a Financial Times headline declaring ‘Bilateral pacts with US a violation, says EU’, the foreign ministers were no longer able to assume that their deliberations would go unnoticed by national parliaments and publics who supported the ICC. The fact that the legal opinion had been drafted by the European Commission, and not the EU Council, would make little difference to most readers. Human rights NGOs stepped up their pressure on the EU for a negative answer to Washington. A flexible EU policy on bilateral agreements was thus framed in public discourse as illegal before the member states even began to negotiate. Among the member states, Germany remained outspoken in its opposition to the US request, and was supported in this position (to varying degrees) by Austria, Belgium, Finland, France, Greece, Ireland, Luxembourg, the Netherlands, and Sweden. Their opposition was motivated both by a principled rejection of American exceptionalism with regard to international law and a practical concern that granting special immunity to the US would defeat the purpose of the ICC by inviting other states around the world whose judicial systems were far less protective of human rights to seek immunity for themselves as well. If EU member states were to sign immunity agreements with the US, they feared, it would be politically difficult, if not impossible, to refuse similar requests from others.31 On the other side of the debate, Italy and the United Kingdom openly favoured a more flexible response to Washington that would permit bilateral agreements, and were supported (to varying degrees) by Denmark, Portugal and Spain. As they saw the issue, American concerns about the ICC had to be addressed because the US plays a unique role in maintaining international peace and security.32 As the foreign ministers prepared for their meeting at Elsinore on 31 August, the question was whether the minority of member states that

Daniel C. Thomas 61

favoured a flexible response to Washington could trump the majority of member states, supported by the European Commission, the European Parliament, and an attentive public, which all favoured a negative response. After all, the ‘flexible’ member states knew that all those favouring a negative response also valued EU unity on the issue. For example, while Swedish Foreign Minister Anna Lindh feared that accommodating the US would undermine the Court, she openly concluded that ‘the most important thing is that the EU stick together’.33 Under these conditions, according to the Competitive Bargaining hypothesis, Italy and the UK should have been able to dominate EU deliberations by threatening to veto any policy that did not meet their preferences. And they certainly seemed willing to try: the Italian and British foreign ministers told their EU colleagues at Elsinore that unless the EU found a way to accommodate US concerns in its common policy, Rome and London would break ranks and sign agreements with Washington.34 In short, the lowest common denominator policy would have been an EU declaration that repeated general support for the ICC while permitting member states to sign bilateral agreements proposed by Washington if they so wished. It is unlikely that this exercise in hard bargaining was simply tactical, rather than a reflection of underlying preferences. Both the British and Italian governments had already alienated large segments of their populations, as well as many of their EU counterparts, with their support of US policy on Iraq. Had they not been truly sympathetic to the US requests, they thus had strong incentives to use the high-profile ICC issue to muster domestic support and to comply with the EU’s written and unwritten rules that encourage member states to reach consensus under CFSP. The fact that they didn’t do so suggests that Rome and London’s threats were driven by real policy preferences and intentions on the issue at hand. But contrary to the logic of the Competitive Bargaining hypothesis, the Italian-British veto threat failed to control the process. At the end of the Elsinore meeting, the foreign ministers announced that they had agreed to pursue a common response to Washington that would maintain the integrity of the Rome Statute and respond constructively to US concerns within that framework.35 Amnesty International and other NGOs quickly criticized Italy and the UK for taking positions that threatened the Rome Statute, but the CFSP process had just begun.36 The member states began preparing for an upcoming meeting of COJUR, the Council’s working group on public international law, which would start the process of translating the awkward Elsinore compromise into a common policy that they all could accept.

62 EU Policy on the International Criminal Court

Although COJUR is normally composed of the foreign ministers’ chief legal advisors, this meeting involved COJUR’s ICC sub-group, which had been created earlier to handle the growing volume of ICC-related work in CFSP. The members of the sub-group knew each other well and communicated on a regular basis. Many had been their governments’ chief negotiator of the Rome Statute, and all were experts on the ICC. As a result, every member was personally committed to the ICC: regardless of their governments’ position on bilateral agreements, most considered the US quest for immunity to be an assault on the Court.37 Yet even if COJUR was ideally conducive to the sort of preference convergence dynamics expected by the theories of Policy Learning and Normative Suasion, there is no evidence that such convergence occurred. Before COJUR convened, the Danish foreign ministry, which then held the EU’s rotating presidency, consulted intensively with its 14 counterparts regarding what they could and could not support. Despite the Elsinore compromise and the presidency’s consultations, the COJUR meeting was contentious, with the UK and Germany leading the arguments for and against a flexible response to Washington’s request.38 In the end, COJUR agreed on the basic elements of a common EU policy that they would recommend to higher levels in the CFSP hierarchy. As Germany was insisting, they agreed that the draft agreement proposed by Washington was incompatible with the obligations imposed by the Rome Statute, which would effectively prohibit any EU member state from signing the text that US officials were circulating. On the other hand, the COJUR framework included four points that were closer to the conditions preferred by the UK: new bilateral agreements were permissible under Article 98; such agreements could not allow impunity for those exempted from ICC jurisdiction; they could only cover military personnel and officials; they could not cover citizens of a State Party to the Statute.39 But the British government insisted on keeping its options open: less than a week after it had accepted the COJUR text, the UK blocked agreement in the PSC on a proposal that the EU Presidency would speak on the behalf of the member states on this issue.40 While the Council was deliberating, other European actors continued to frame the issue in manner designed to constrain those member states sympathetic to the US position. The Council of Europe’s Parliamentary Assembly passed a resolution declaring that the agreements proposed by Washington were incompatible with the Rome Statute and calling upon Council of Europe members (which included all EU member states) not to sign such agreements. The following day, the European Parliament adopted a resolution expressing its view that signature of an agreement

Daniel C. Thomas 63

that undermines the effective implementation of the Rome Statute was incompatible with EU membership.41 The White House’s release of its new National Security Strategy, which included an explicit rejection of the ICC, only strengthened the convictions of the anti-flexibility camp.42 Even after COJUR, there was a great deal of give-and-take in the member states’ deliberations, led by British and German attempts to shift the emerging compromise in their respective directions. The Danish EU presidency prepared an elaborated version of the COJUR recommendations, which were debated and revised repeatedly at multiple meetings of PSC and COREPER in late September. Although Germany had initiated the process of developing a common EU policy on the bilateral agreements, its government hesitated as the policy took shape: Foreign Minister Joschka Fischer was deeply committed to the ICC and reluctant to endorse a policy that could be seen to authorize actions that he feared would weaken the Court. Germany’s representative to COJUR’s ICC sub-group, Hans-Peter Kaul, then spoke with Fischer. The draft EU deal was closer to Germany’s position than Fischer may have realized, Kaul indicated, and if Fischer insisted on his position, London and/or Rome were likely to refuse any further concessions. Recognizing that the resulting collapse of EU unity in the face of Washington’s global campaign would be worse for the ICC than a compromise agreement, Fischer relented and deadlock was avoided.43 In its meeting on 30 September, the General Affairs Council (GAC) issued the EU’s policy on bilateral agreements in the form of Conclusions that reaffirmed the EU’s support for the effective functioning of the ICC and expressed its belief that the Rome Statute provided all the necessary safeguards against the use of the Court for politically motivated purposes. In addition, the Conclusions signalled the readiness of EU member states to discuss with the US how its concerns could be accommodated within existing agreements. Finally, the Conclusions included a set of EU Guiding Principles to guide any State Party to the Rome Statute (including but not necessarily limited to EU member states) that wishes to reach an arrangement regarding the conditions of surrender of persons to the ICC.44 The Guiding Principles were drafted to ensure that any bilateral arrangement would be maximally consistent with the obligation of States Parties to cooperate fully with the ICC. In particular, they stipulate that (1) existing international agreements (such as extradition treaties and SOFAs) should be taken into account; (2) the draft agreements proposed by the US are inconsistent with the Statute; (3) any arrangement must ensure that persons who

64 EU Policy on the International Criminal Court

have committed crimes covered by the Statute are investigated and punished; (4) no arrangement for non-surrender can apply to nationals of an ICC State Party; and (5) any such arrangement can only cover government officials or military personnel. The Council Conclusions were thus clearly filled with compromises, notwithstanding Danish Foreign Minister Per Stig Moller’s insistence that there was ‘no concession… no undermining of the ICC’.45 The real question is who made concessions to whom in order to achieve the compromise. In an opinion piece written for an American audience, Moller referred diplomatically to ‘a good compromise’ between US concerns about the ICC and the EU’s commitment to the integrity of Rome Statute.46 In fact, the Conclusions were less a compromise between the EU and the US than a compromise between the member states with most divergent preferences, the UK and Germany, as expected by Normative Institutionalism. In that the Conclusions did not expressly forbid EU member states from signing a bilateral agreement with the United States under Article 98 of the Rome Statute, they reflect the British and Italian government’s insistence on their right (and intent) to do so. As soon as the Conclusions were issued, a British diplomat confirmed that London was considering whether to sign an agreement with Washington.47 In mid-October, an Under-Secretary from the British Foreign Office told the House of Lords that the government was ‘beginning discussions with the United States on the possibility of a bilateral agreement’.48 However, the preferences of the majority of EU member states are clearly reflected in the rejection of the text proposed by Washington and in the strict requirements that any new agreement would have to meet. In order to comply with the Guiding Principles, the United States would have to stop seeking immunity for US citizens who were not sent abroad on government business, stop seeking immunity for citizens of states party to the Rome Statute who were working for the US government (most likely as private contractors), demonstrate that US courts have jurisdiction over all crimes within the jurisdiction of the ICC as defined in the Statute (which is not presently the case), and commit to investigate in good faith all credible accusations of such crimes. Given the deep suspicions of international law in general and the ICC in particular that prevailed in Washington during this period, it was highly unlikely that the US government would satisfy these conditions. Emerging from the GAC, Joschka Fischer described the Conclusions as a virtual rejection of bilateral agreements: ‘We are against the con-

Daniel C. Thomas 65

clusion of special agreements and we will not conclude such an agreement… We would have wished a clear rejection of the agreements. Because of the Principles we are very close to such a position.’49 By quickly expressing its displeasure with the EU’s guidelines, the US State Department seemed to agree with Fischer’s analysis, thus confirming that the veto threatened by Italy and the UK in order to ensure an agreement satisfactory to Washington had failed to dominate negotiations among the member states.50 Just in case, Germany’s Foreign Office soon released a detailed analysis of the Council Conclusions designed to bolster its view that they constituted a de facto (if not de jure) prohibition on signing bilateral immunity agreements.51 The non-governmental Coalition for the ICC agreed with the German memorandum (which it posted on the CICC website), and mobilized its lobbying efforts accordingly. At the next COJUR meeting, the UK expressed its strong displeasure with the German memorandum, which led others to conclude that London was still leaning toward signing an agreement.52 Yet despite these multiple indications that the British government intended to sign a bilateral agreement, and that it believed this would be consistent with the Council Conclusions, the strict requirements imposed by the Guiding Principles, combined with pressure from EU and transnational channels, apparently made it difficult for London to do so. A special US ambassador travelled to London, Madrid, Rome and Vienna for discussion of the issue in late 2002, but failed to achieve any agreements. The Foreign Ministry of Portugal, which had tacitly supported the British and Italians during the negotiation of the Council Conclusions, rejected a bilateral agreement proposed by Washington after receiving a negative legal opinion from the General Attorney’s office, and decided to ‘freeze’ further discussion of the issue.53 In the end, no EU member state signed a bilateral agreement on ICC immunity with the US after the Council Conclusions were agreed, and the US subsequently abandoned its effort to obtain such agreements.

Conclusions These two case studies have a number of interesting implications. First, although there is no evidence of preference convergence among member states during the negotiations, the two cases suggest that both Intergovernmentalism and Normative institutionalism have real explanatory power with regard to EU foreign policy. In the first case, as the Competitive Bargaining hypothesis would expect, the UK made clear

66 EU Policy on the International Criminal Court

to France and Ireland that if they wanted EU unity within the Security Council, they had to accept the UK’s proposal. By refusing to compromise, they dominated the process. Two months later, the British-Italian threat to veto any EU prohibition on bilateral immunity agreements forced the majority of EU foreign ministers to make real concessions to the UK and Italy at their Elsinore and COJUR meetings. However, contrary to the expectations of Intergovernmentalism, the British-Italian threat ultimately failed to dominate the negotiations in the second case. The German government’s preference for a firm rejection of American demands regarding ICC immunity, as well as the exercises in public framing and rhetorical pressure by non-state actors, all exerted significant influence on the EU’s ultimate position. Although the Commission’s Legal Service had little statutory role in CFSP, the distribution and then media leak of its opinion on ICC bilateral agreements shaped public expectations and steered the ensuing debate within the Council. In the end, despite their veto threat, London and Rome made significant concessions in order to achieve an EU policy that was consistent with prior EU commitments and that satisfied Germany as much (or more) as it did them. Similarly, once a compromise policy was adopted, the UK and Italy found their freedom of maneuver constrained by rhetorical pressure from actors with different preferences, including other member states, national parliaments, the European Parliament, and NGOs. As a result, while both the UK and Italy insisted on the flexibility to sign bilateral agreements, and even threatened EU unity in order to guarantee this flexibility, neither did so in the end. In sum, while the first case seems to support Intergovernmentalism, the entrapment and mutual compromise in the second case can only be explained by Normative Institutionalism. Finally, but no less important, the contrast between the LCD outcome in the case of the UN resolution and the compromise outcome in the case of bilateral agreements indicates that the institutional setting in which EU foreign policy cooperation occurs matters a great deal. In particular, it supports Normative Institutionalism’s assertion that cooperative bargaining is most likely to emerge where the EU’s procedural norms are most salient (that is, when the issue in question is subject to collective deliberation within EU forums) and less likely when member states negotiate in a forum (most likely outside the EU) that is shielded from the compliance pull of EU norms. The two-case comparison thus provides powerful evidence that while the EU’s foreign policy-making process may fall victim to LCD dynamics, it is not condemned to produce such processes or outcomes.

Daniel C. Thomas 67

Notes 1 Research on this paper was supported by the Council on Foreign Relations, the German Marshall Fund of the United States, and the University of Pittsburgh’s European Union Center. I thank Anand Menon and William Wallace, as well as the participants and commentators at the project workshops in Pittsburgh and Dublin, for valuable comments. 2 For all CFSP acts regarding the ICC, see cms3_fo/showPage.asp?id=404&lang=EN&mode=g, accessed 1 July 2010. 3 ‘War Crimes Court Created Over Fierce U.S. Objections’, Washington Post, April 12, 2002. Compare Javier Solana, ‘International Court Signals a New Era’, International Herald Tribune, 11 April 2002. 4 US Department of Defense News Release No. 233-02, 6 May 2002. 5 Declaration by the EU on the Position of the US towards the International Criminal Court, Brussels and Madrid, 13 May 2002. Twelve EU candidate countries plus Norway formally aligned themselves with the CFSP declaration. 6 ‘Will Bush Invade Cambridgeshire?’, The Daily Telegraph, 9 May 2002. 7 For text, see 8 For the full text of these documents, see reaction.html, accessed 1 July 2010. 9 ‘U.S. Seeks Court Immunity for E. Timor Peacekeepers’, Washington Post, 16 May 2002. 10 ‘Les Etats-Unis Surs de Leurs Bon Droit’, Libération, 18 juin 2002. 11 ‘U.S. Drops Demand for War Court Immunity’, Washington Post, 11 July 2002. 12 ‘EU to Speed Action on Bosnia Police’, Financial Times, 4 July 2002; ‘Action on Bosnia Forces Delayed’, Washington Post, 4 July 2002. 13 Chris Patten, ‘Why Does America Fear This Court?’, The Washington Post, 9 July 2002. 14 Author’s interview with European official, June 2008. 15 ‘Germany Steps Up Criticism of US Over International Court’, Agence FrancePresse, 7 July 2002. 16 United Nations, Press Release SC/7445/Rev.1, 10 July 2002, Security Council 4568th Meeting. 17 Author’s interview with European official, June 2008. 18 ‘UK Accused of Preparing a Deal with America over Criminal Court’, The Independent, 10 July 2002; ‘British Back US Plan to Avoid War Crimes Court’, The Times (London), 11 July 2002. 19 Author’s interview with European official, June 2008. 20, accessed 18 February 2009. 21 ‘On World Court, US Focus Shifts to Shielding Officials’, The New York Times, 7 September 2002. 22 ‘Germany Gives Cool Response to U.S. Request for Soldiers’ Immunity’, Handelsblaat (English on-line edition), 23 July 2002. 23 Statement by Danish Foreign Minister Per Stig Moller on the International Criminal Court at the European Parliament, Strasbourg, 25 September 2002. 24 Author’s interviews with European officials and NGO experts, November 2004. The Commission’s full opinion was later published in Human Rights Law Journal, 23 (September 2002), 158–9.

68 EU Policy on the International Criminal Court 25 Author’s interviews with European officials and NGO experts, November 2004. 26 ‘US Issues Warning to Europeans in Dispute Over New Court’, The New York Times, 26 August 2002. 27 Author’s interviews with European officials and NGO experts, November 2004. 28 ‘Bilateral Pacts with US a Violation, says EU’, Financial Times, 28 August 2002. 29 ‘East Europeans Torn on the Rack by International Court Row’, The Guardian, 17 August 2002. 30 ‘Bilateral Pacts with US a Violation, says EU’, Financial Times, 28 August 2002; ‘EU Legal Experts Say Nations Have No Right to Exempt US Military from War Crimes Trials’, Associated Press, 28 August 2002. 31 Author’s interviews with European officials and NGO experts, November 2004. 32 ‘UK to Back US on War Crimes Court’, Financial Times, 31 August 2002; ‘UK, Italy Eye U.S. Court Deal’,, 31 August 2002. 33 Quoted in ‘UK, Italy Eye U.S. Court Deal’,, 31 August 2002. 34 ‘Italian Premier’s Stance on International Court Drops Bombshell at EU Talks’, BBC Monitoring Service, 1 September 2002. 35 ‘Brussels Notebook: A Battle for the ICC’, Financial Times, 1 September 2002; ‘Straw Urges War Court Retreat’, The Guardian, 2 September 2002. 36 ‘US-EU Deals Over Criminal Court Rapped’, Bahrain Tribune, 3 September 2002. 37 Author’s interviews with European officials, November 2004. 38 Author’s interviews with European officials, November 2004. 39 ‘Britain Reveals Plans for Compromise on ICC’, Financial Times, 3 September 2002; ‘EU Nears Deal on New Court’, BBC News Online, 4 September 2002. 40 See timeline at 41 Council of Europe Parliamentary Assembly Resolution 1300, 25 September 2002; European Parliament Resolution P5_TA-PROV (2002)0449, 26 September 2002. 42 National Security Strategy of the United States of America, Washington, DC, September 2002, p.31; interviews with European officials, November 2004. 43 Author’s interviews with European officials and NGO experts, November 2004. 44 EU Council Conclusions, 30 September 2002,, accessed 1 July 2010. 45 ‘EU Gives US Troops Immunity’, BBC News Online, 30 September 2002. 46 Per Stig Moller, ‘A Good Compromise on the ICC’, Wall Street Journal, 2 October 2002. 47 ‘EU Deal to Exempt US from New World Court’, The Scotsman, 1 October 2002. 48 Lords Hansard, 14 October 2002, Volume 639, Part 186. 49 Associated Press, 30 September 2002; ‘Despite EU Accord, Germany Won’t Exempt U.S. From ICC’, The Wall Street Journal Europe, 1 October 2002. 50 ‘US Rejects EU Criminal Court Guidelines’, BBC News Online, 2 October 2002. 51 ‘ICC – Supportive Interpretation of and Commentary on the EU General Affairs Conclusions on the International Criminal Court of 30 September

Daniel C. Thomas 69 2002’, 24 October 2002, 24Oct02.pdf, accessed 1 July 2010. 52 Author’s interviews with European officials and NGO experts, November 2004. 53 ‘Portugal “freezes” US immunity deal’, BBC Monitoring Service, 12 April 2003.

Works cited Benedetti, F. and Washburn, J.L. (1999) ‘Drafting the International Criminal Court Treaty: Two Years to Rome and an Afterword on the Rome Diplomatic Conference’, Global Governance, 5(1): 1–38. Checkel, J.T. (2006) ‘Tracing Causal Mechanisms’, International Studies Review, 8(2): 362–70. Clinton, W.J. (2001) ‘Statement on the Rome Treaty on the International Criminal Court’, 31 December 2000. Weekly Compilation of Presidential Documents, 37(1): 4. George, A. (1979) ‘Case Studies and Theory Development: The Method of Structured, Focused Comparison’, in P.G. Lauren, Diplomacy: New Approaches in History, Theory, and Policy. New York: Free Press, pp.43–68. Patten, C. (2002) ‘The International Criminal Court’. Address to the European Parliament, 25 September 2002. Puetter, U. and Wiener, A. (2007) ‘Accommodating Normative Divergence in European Foreign Policy Coordination: The Example of the Iraq Crisis’, Journal of Common Market Studies, 45(5): 1065–88. Scheffer, D.J. (2002) ‘Staying the Course with the International Criminal Court’, Cornell International Law Journal, 35: 47–100.

5 EU Policy on the Iraq War and its Aftermath: The Breakdown and Revival of Consensus-based Decision-making Jeffrey Lewis1

At first glance, the deep divisions in Europe over the March 2003 US-led invasion of Iraq hardly represent a successfully chapter in the evolution of Common Foreign Security Policy (CFSP) and the EU’s long-term effort at becoming a coherent, influential international actor. The run up to the Iraq war split open fissures amongst EU member states rarely displayed in such naked clarity: ‘new’ versus ‘old’, ‘Atlanticist’ versus ‘Europeanist’, and within the big state ‘triumvirate’ of Britain, France, and Germany. And when a majority of member states publicly broke ranks with a tenuously reached common position, sceptics argued that the EU’s consultative and consensus-based process of foreign policy-making was either fictitious or irrevocably broken. According to David Calleo, ‘internal divergences over Iraq mocked the geopolitical vision of European unity’ (2004: 32). But what is striking about this case is not the failure of EU members to reach a common policy on Iraq. Indeed, as John Peterson noted soon after the war began, ‘Iraq had been a bitterly divisive issue in both transatlantic and inter-European relations for at least ten years’ (2004a: 11).2 What is striking, rather, is the normative reframing of policy for reconstruction assistance to Iraq which followed in the wake of such open member state divergence. Contrary to the conventional wisdom on Europe’s divided response to the Iraq crisis, deliberations among EU foreign ministers in the first half of 2003 showcase the durability of member states’ shared commitment to make foreign policy decisions in camera and by consensus, which tends to produce median compromises rather than lowest common denominator (LCD) outcomes. How and why Europe overcame discord on Iraq is the focus of this chapter, with the overall goal of explaining how the arguments for ‘no EU action’ and a ‘let the 70

Jeffrey Lewis 71

UN do it’ approach to Iraq’s reconstruction became disempowered. To track the evolution of EU policy, the chapter is divided into two cases with distinct outcomes. The dependent variable for both is the nature and extent of the EU’s common policy towards Iraq. The potential LCD outcome for both cases is no EU-level action and reliance on a ‘UN only’ approach to Iraqi assistance. As will be shown in the first case (November 2002–March 2003) open discord over military action in Iraq worked to deactivate the consensus-based and norm-driven CFSP decision-making process. As case one examines, the way this context became deactivated led member states into a disengaged dialogue, galvanized uncompromising national positions, and even triggered an internecine public letter campaign which seemed to confirm the worst fears of a divide-and-rule logic in transatlantic relations. As a result, case one concludes with an the LCD outcome, evidenced by the European Council decision that the UN ‘Oil For Food’ Program should handle Iraq’s reconstruction. During this period, the most recalcitrant ‘no EU action’ veto player was Germany, bolstered by a larger subset of member states with strong ‘let the UN do it’ preferences, including France, Belgium, and Luxembourg. In other words, the initial policy divergence was between a small but insistent group of member states who were loathe to use EU-level action to lend any legitimacy to the Iraq war, hence the ‘UN only’ viewpoint. Despite the outcome of case one, a new common policy for EU reconstruction assistance to Iraq slowly gained currency and is documented in the second case below (March 2003–June 2003). Using the Normative Institutionalist (NI) approach developed by Daniel Thomas (ch.2, this volume), the analysis attempts to explain how and why such a new common policy was possible in the second case given the deep rifts exposed in the first. Going well beyond the LCD status quo of the first case, the EU compiled a multi-part economic, humanitarian, and technical assistance package to aid in the post-war reconstruction of Iraq. Observables include sustained EU reconstruction and humanitarian assistance to Iraq, in total about one billion euros since 2003 and an integrated ‘rule of law’ mission (EUJUST LEX) which has provided training to over 3,200 senior police, penitentiary officials, judges, and prosecutors since July 2005.3 Less observable, but as significant for Europe’s foreign policy aspirations is the revived legitimacy attached to consensus-based CFSP and the shared understanding that the Iraq crisis violated a number of informal norms. In particular, the way some members (and accession candidates) publicly broke ranks with a tenuously reached common position by the foreign ministers in January 2003 was seen as a violation of

72 EU Policy on Iraq

CFSP consultation norms specifically, and the Council’s ‘code of conduct’ generally.4 Europe’s reframing efforts over Iraqi reconstruction offer confirming evidence for both NI hypotheses developed by Thomas: entrapment (H1) and cooperative bargaining (H2). First, it will be shown that as the dust settled from a very public display of disagreement, the EU’s normative commitments to promote democracy and the rule of law worldwide entrapped those member states with strong ‘no EU action’ and/or ‘let the US/UN do it’ preferences into accepting EU-coordinated reconstruction assistance (H1). It certainly helped that normative commitment arguments were inseparable from the view that EU reconstruction aid was important to rehabilitate transatlantic relations and safeguard against being treated as a junior partner subject to divide-and-rule tactics. Substantive policy commitments for Europe’s long-term role in the Greater Middle East further entrapped those with ‘UN-legitimated action only’ and/or ‘no EU action period’ preferences. During a series of formal and informal meetings among the EU foreign ministers in the first half of 2003, they gradually reached a new understanding that to do otherwise would fly in the face of long-term European commitments to promoting stability in the Middle East. Second, the Iraq crisis did not obviate the EU’s norm-based process of foreign policy-making; instead, it helped renew the principled commitment among member states for such cooperative bargaining procedures as consultation and in camera deliberation (H2). The most directly observable instance of this, discussed in Case 2 below, is the informal foreign ministers’ cruise in the Greek isles that legitimated a consensus on EU-coordinated aid and gave a ‘political green light’ to CFSP High Representative Javier Solana to draw up a new comprehensive European Security Strategy (ESS). The lessons learned from the Iraq crisis thus include a new shared understanding among the EU foreign ministers on principled, substantive, and procedural grounds for concerted EU assistance to postinvasion Iraq. While ‘getting out the checkbook’ may code to some observers as a minimalist LCD outcome, there are two reasons to question this interpretation. First, as Case 1 shows, the LCD status quo was to let UN ‘Oil for Food’ coordinate reconstruction efforts and what the EU eventually endorses goes significantly beyond this. Second, the meaning attached to a new consensus-based approach for EU reconstruction assistance was inextricably linked to intangible process interests in how EU foreign policy is made. The Iraq crisis is an instructive case to view how procedural norms for cooperative bargaining can be deactivated and reactivated and the interplay between the two dynamics. There is

Jeffrey Lewis 73

also confirming evidence from close participants in the CFSP process that the Iraq crisis, and especially the public letter campaign examined in case one below, violated unwritten norms and procedures which should in the future proscribe similar behaviour and may serve as guidance for a new ‘code of conduct’ in CFSP. The remainder of the chapter is organized as follows. Section 2 documents the first case where the divergence among EU members during the Iraq crisis from November 2002 to March 2003 led to an atmosphere of uncompromising positions and a breakdown of informal norms of behaviour once a tenuous common position was reached on the eve of war. Section 3 focuses on the second case from March 2003 to June 2003, where the EU’s discord over Iraq was reframed around the issue of reconstruction assistance and assesses how and why this policy was rhetorically empowered in the face of several member states who held contrary ‘no EU action’ preferences. Section 4 concludes by assessing possible longer-term implications for this case on the EU’s ability to make effective common foreign-policy decisions.

Case 1 – Europe’s position on Iraq goes from ambiguous to open divergence (November 2002–March 2003) On 8 November 2002, after months of tough negotiations over the wording of conditions that Iraq needed to meet in order to avoid war, UN Security Council Resolution 1441 was adopted. For Europe, the ‘Iraq crisis’ was about to live up to is name. During the Fall of 2002, intra-European divergence on policy towards Iraq became more public and more pronounced. Prior to this time the lack of agreement among member states had been ‘successfully clouded in ambiguity’ (Toje, 2005: 119).5 But during the Fall of 2002, the growing divide between Britain and the anti-war governments in France and Germany became regular headline news. Coupled with a close domestic election race, by September 2002 the Schröder government had announced ‘a very outspoken antiwar stance’ that rejected German support for military intervention even if legitimated by the UN (Stahl et al., 2004: 422; Pond, 2003). By November, in EU forums, the foreign ministers were barely discussing Iraq at all, and for all practical purposes the Iraq crisis ceased to be an active agenda item for the General Affairs and External Relations Council (GAERC), the Union’s premier foreign policy decisionmaking body. At the heart of this blockage was Franco-German opposition. As John Peterson recounts: ‘Chirac and Schröder, together with very small groups of close advisers, crafted uncompromising

74 EU Policy on Iraq

anti-war positions on Iraq and thus made the bitterest of clashes with the Bush administration, and much of the rest of the EU, inevitable’ (2004a: 15). This uncompromising atmosphere at the highest political levels, effectively deactivated the EU’s norm-based process for consultative and consensus-based foreign policy-making. The public posturing which took place by both those who supported and those who opposed military intervention short-circuited the institutional context within which the EU’s CFSP normally operates. One former EU ambassador had this reflection five years later: ‘I remain convinced that if politicians had allowed professional diplomats to try and find an acceptable form of words enabling each participant to do what he wanted to do anyhow, but without creating a major transatlantic and intra-European crisis, that could have been done in good Coreper fashion. But with Chirac, Rumsfeld et alia, this was not [in] the cards.’6 In the words of another high-ranking EU official, ‘On Iraq the three big leaders came out on very different positions. There is no [CFSP] mechanism to overcome that.’7 This applies with particular clarity to the sidelining of the EU’s CFSP High Representative, Javier Solana. Allen and Smith (2004: 95) note that ‘Solana’s attempts to preserve unity by brokering innocuous common statements were undermined by the determination of larger Member States to pursue their own policies regardless of their impact on an EU common position’. By January 2003 the EU’s internal divisions over Iraq, which already seemed insurmountable, were about to amplify. On 23 January, US Defense Secretary Rumsfeld dropped his caustic ‘old’ Europe remark when asked by a reporter to comment on the lack of European support. A common position on Iraq was reached by the foreign ministers on 27 January but it was an LCD status quo (endorsing a UN solution to the crisis) that fell apart within days. Fueled by ‘old Europe, new Europe’ headlines, the minimalist common position was circumvented as a group of member states bandwagoned to sign public letters of support for the military intervention option favoured by the US.8 First, on 30 January, the leaders of eight states (led by Spain and Britain) published an open letter in the Wall Street Journal Europe and 12 other European newspapers condemning Iraq as a ‘clear threat’ to international security.9 It was not difficult to read between the lines to comprehend this was an open endorsement for the US position. Even worse, the so-called ‘Letter of 8’ was published on the same day that the European Parliament adopted a joint resolution (by a vote of 287–209, with 26 abstentions) that the current material breaches of

Jeffrey Lewis 75

UNSCR 1441 did not justify military intervention. Second, on 5 February, a new US support letter from other regions of ‘new’ Europe was published. This letter became known as the ‘Vilnius 10’ as a reflection of the Central and Eastern European who signed on (Albania, Bulgaria, Croatia, Estonia, Latvia, Lithuania, Macedonia, Romania, Slovakia, and Slovenia). Immediately, the ‘Vilnius 10’ letter prompted a harsh rebuke from the French President that for newcomers and would-be club members, ‘It is not really responsible behavior, it is not well broughtup behavior. They missed a good opportunity to keep quiet.’10 The two letters had the effect of trashing the common position on Iraq which the EU15 foreign ministers had inked only days before and after lengthy wrangling over the wording. The public letters were clear violations of the long-standing custom of consultation that was cultivated over time within the European Political Cooperation (EPC) process (Dimitrakopoulos and Passas, 2004: 44; Smith, 2004a). According to Michael Smith, ‘the most fundamental principle of European foreign policy cooperation is that EU member states must avoid taking fixed positions on important foreign policy questions without prior consultation with their partners’ (Smith, 2004b: 101). The Financial Times claimed, ‘to some diplomats, the ‘declaration of the eight’ was regarded as a betrayal of all the attempts by Javier Solana, the EU’s foreign policy chief, to keep the Europeans together’.11 Solana apparently only learned of the letter’s publication from the radio.12 Nor was the Greek presidency consulted or even forewarned (Toje, 2005: 119), another explicit violation of the Council’s culture and procedural code. Some reports note that Greek authorities were not told beforehand, even though Prime Minister Simitis spoke ‘a few hours prior to the publication of the open letter’ to both Prime Ministers’ Blair and Berlusconi.13 It is also significant that drafts of the letter bypassed the EU’s institutions completely, routed instead through embassies to the individual countries.14 While the media typically interpreted French President Chirac’s chastisement of newcomers (i.e. – to ‘shut up’) as an ‘extraordinary outburst’,15 it is politically symbolic of the violation of unspoken rules in an institutionalized community setting. A clear if more subtle reference to this violation can also be detected in a speech by the Dutch European Affairs Minister at Romania’s European Institute shortly following the ‘Letter of 10’, where he notes, ‘If we learn of partners’ initiatives only from the press, something is clearly wrong. You do not always need to agree, but you do need to understand each other.’16 If the ‘Vilnius 10’ letter by unsocialized newcomers prompted such a

76 EU Policy on Iraq

public reaction, the ‘Letter of 8’ was a more flagrant abuse of community standards of appropriateness amongst members. Granted, the common position reached by the EU15 foreign ministers on January 27 was nothing more than an LCD statement reaffirming the lead role of the UN in resolving the crisis, but it still represented a common position reached within the EU’s institutional context after weeks of mutual responsiveness over semantics. There was a clear intention by the letter’s organizers (especially Britain and Spain) to not inform or consult antiwar member states. Citing an interview with a British foreign office official, Toje notes that a conscious decision was made to not approach France and Germany (2005: 119). The select countries that were asked to sign the letter kept the entire matter secret, and there was apparently a ‘marketing’ effort to give the letter only to select newspapers in countries where the prime minister signed on, along with the Wall Street Journal Europe.17 There are even reports that Prime Minister Blair spoke by telephone to President Chirac on 29 January and did not mention the letter’s ensuing publication slated for the next day.18 A clear confirmation for the existence of appropriateness standards here can be seen in the Dutch invitation but refusal to sign on. The Netherlands was asked to sign the ‘Letter of 8’ but rebuffed the invitation under the reasoning that ‘no purpose is served by accentuating differences between member states on the Iraqi question’.19 In one EU official’s words, ‘the Dutch didn’t disagree with the position [in the letter] but they did disagree with the procedure and they felt this was not the right way to do it’.20 Perhaps because of their closeness to the French position on Iraq, Belgium was not invited to sign, as publicly noted by Foreign Minister Louis Michel who added that ‘Belgium has no wish to take part in this dividing up of the international community over the Iraqi dossier’.21 In short, as Anand Menon nicely puts it, the letter’s ‘divisive impact stemmed from bitterness about the process that spawned its appearance’ (2004: 638). Luxembourg responded to the letter’s publication with perhaps the most public outrage, ‘regretting the scandalous behavior and lack of solidarity of the eight European countries’.22 The letters of the ‘8’ and ‘10’ certainly shredded the EU’s finely balanced if minimalist common position. But beyond the immediate damage to the thinly glued common position, was the perception by some that there was a ‘tangible breach of some “unwritten rules” of EU functioning’ and that the ‘lack of solidarity’ should ‘in the future be avoided by a “code of conduct” for EU governments’.23 A high-ranking official characterized the public letter campaign as a ‘very disruptive way of conducting busi-

Jeffrey Lewis 77

ness’. He went on to add: ‘[In the EU] we have a very elaborate way of talking with one another. We do not settle our problems this way. You don’t go to the press and write open public letters.’24 In light of this, the Dutch non-signature offers an insightful illustration at how pronorm behaviour in EU foreign policy-making relies on self-restraint and internalized standards of appropriateness (rather than external sanctioning and formalized rules of conduct). Noting that it would have been ‘unsurprising’ for the Netherlands to have signed the public letter given their strong ‘Atlanticist’ and NATO credentials, Menon and Lipkin (2003: 20) hypothesize several reasons why they may have held back support: ‘Concern about the potential effects on European relations (particularly relations with Germany) of such a public EU division, together with uncertainty associated with the absence of a new government after the general election, were key factors in persuading the caretaker Dutch government not to sign.’ But the former rather than the latter seems to weigh decisively in the decision not to sign the letter, especially given the outcome of the general election on 22 January 2003 which extended the existing Prime Minister Bakenende’s CDA (Christian Democratic Appeal) seats in the parliament from 43 to 44 (with 28.3% of the vote). While it was several months before a new coalition government was sworn in, both the Prime Minister and a number of key foreign ministry personnel remained. Given the strong 22 January outcome for the CDA it is unlikely that withholding the Netherlands’ signature from the letter was based on electoral calculations or new coalition-formation dynamics. Rather, it is more significant that the new Dutch Minister of Foreign Affairs (from 3 December 2007) was Bernard Bot, a former long-standing former EU Permanent Representative who was well versed in the Brussels-based culture of decision-making by consensus and cooperative bargaining.25 Likewise, the outgoing Foreign Minister (and soon to become Secretary-General of NATO), Jaap de Hoop Scheffer was another top foreign policy official with strong credentials in the culture of European foreign policy cooperation. We can see indirect reference to the rhetoric of this culture in his speech before the Parliamentary Commission on Foreign Affairs in Paris at the inauguration of the Franco-Dutch Cooperation Council shortly after the letters’ publication. Specifically, he refers to the foregoing Iraq crisis which ‘laid bare a number of differences… within the EU itself,’ by noting that ‘over the years we have come to share an “esprit Européen” which transcends our occasional differences of opinion’.26

78 EU Policy on Iraq

The depth of Europe’s divergence over Iraq is best seen in the conclusions of the ill-fated 17 February European Council in Brussels (dubbed the ‘war summit’ in the media). The February summit was organized by the Greek presidency as an ‘extraordinary’ meeting with Iraq as the sole agenda item. Britain allegedly did not want the meeting to happen, given the rigidity of national positions at this stage, but failed to persuade the Greek presidency to call it off.27 Four days before the scheduled meeting, it remained unclear whether a joint statement would even be possible to produce (Ludlow, 2003: 1). Two days before, on 15 February, organized war protests across Europe saw mass demonstrations, estimated in the millions, against military intervention. France and Germany refused to send their Political Directors who were assigned a ‘pre-summit’ meeting the night before.28 The summit did produce a terse joint statement, less than two pages, conceding ‘primary responsibility for dealing with Iraqi disarmament lies with the [UN] Security Council’.29 The debate, held during a working dinner that ended in less than two hours, has been described by one inside account as a ‘dialogue des sourdes’ where ‘all we heard, one participant observed, was yesterday’s speeches’ (Ludlow, 2003: 8). In summary, the highly public disagreement over Iraq led to uncompromising national positions and deactivated the norm-laden institutional context of consensus- and consultation-based CFSP. As the war commenced, the EU heads of state and government met in Brussels on the evening of March 20. At this point, one might expect EU humanitarian assistance as a minimalist LCD option for the European Council to endorse. The EU already had a decade of experience with multilateral humanitarian aid to Iraq, coordinated by the Commission’s Humanitarian Office (ECHO).30 Media accounts of the March summit note that EU leaders were expected to offer immediate aid to the Iraqis, but this did not happen. Richard Youngs (2004a: 2) notes, ‘the Europeans’ preference was for the UN to assume full control’. This was most strongly advocated by the French, but with key support from Germany as well. According to Youngs (2006: 55): ‘The German development ministry was eager to channel aid through the UN as a means of depoliticizing its potential contribution in Iraq by sheltering it from German public opinion.’ Across Europe, in general, as reported by The Independent, ‘The issue of reconstruction aid is seen as too sensitive since many countries believe that their taxpayers should not be asked to repair damage from a war they oppose.’31 Thus, the outcome of the first case is best characterized as a minimalist LCD outcome to endorse the status quo: any European assistance must occur

Jeffrey Lewis 79

through the UN not the EU. The 20 March European Council Presidency Conclusions simply state, ‘We support the UN Secretary General’s proposal that the humanitarian needs of the Iraqi people can continue to be met through the “Oil for Food” programme.’32 However, as the second case will document below, within two months of the March summit, the EU would reframe discord over the Iraqi war into a reconstruction program steadily increasing in scope and commitments.33

Case 2 – How normative entrapment and cooperative bargaining led to the EU’s reconstruction assistance program (March–June 2003) Between the European Council summit on 20 March and the 19–20 June Thessaloniki summit, the EU overcame discord with a new common strategy for the multilateral reconstruction efforts that would be needed in post-Saddam Iraq. The foreign ministers also initiated the development of a new comprehensive ‘European Security Strategy’ by Javier Solana’s team that was adopted in December 2003. An initial, small package of aid to Iraq was approved by the foreign ministers on 22 April 2003 for ‘urgent medical needs’, but this was only a modicum of what was to come over the next several months and beyond. Overcoming divergence on Iraq took place slowly over several months, with many bilateral, informal, and ‘back channel’ discussions leading the way rather than group deliberations within the framework of Councilbased CFSP settings such as the GAERC. Complicating any analysis of the EU’s post-invasion policy towards Iraq is the lack of a clear turningpoint moment, single decision, or smoking-gun argument that wins the day. The way divergence over war in Iraq was reframed is thus not unproblematic to interpret since there was simply not a clear winning argument that persuaded those with ‘do nothing’ and/or ‘let the UN do it’ preferences to change positions. In general, the period of April–June 2003 represents a cooling of tempers and principled effort by member state officials at a range of levels (foreign ministers and their close advisors, EU Political Directors, the Political and Security Committee Ambassadors) to reactivate the cooperative bargaining style of norm-governed and consensus-based foreign policy-making, to cope with one of the worst transatlantic crises in post-war history, and to further Europe’s substantive long-term goal of promoting peace and stability in the Greater Middle East. During this period, overcoming EU divergence on Iraq is best characterized by the emergence of a compelling overall logic that the

80 EU Policy on Iraq

opportunity costs both in transatlantic relations and Europe’s influence in the Greater Middle East and beyond (globally) were too high a toll to pay and forced a more-or-less grudging admission for a new common policy endorsing EU-based reconstruction assistance. Those with ‘no EU action’ preferences gradually became entrapped by the EU’s longer-term normative commitments to become a global actor. At first, around the time of the February and March summits, a group of member states opposed to military intervention were flatly against any EU-directed material support since it could be viewed as legitimating a conflict they opposed. We see direct evidence of this view in the ‘let the UN Oil for Food program do it’ position advocated in the 20 March EU Presidency Conclusions. Iraq reconstruction assistance as a legitimate EU foreign policy was incrementally empowered between March and June 2003 because it fit several key themes in the type of global actor the EU aspires towards: as an advocate of UN-centred multilateralism and as a worldwide promoter of democracy, human rights, and the rule of law. In short, the debacle over preemptive military action soon became repackaged as an opportunity to showcase Europe as a regional source of ‘soft power’. All of these arguments were on the table during the foreign ministers’ deliberations during case two between March–June 2003. Characterizing the overall tone of these meetings, one EU official emphasized there was an effort to ‘look at the situation as it was, to leave the background [over military intervention] aside. Emerging from this was a rationale that [reconstruction assistance] was the right thing to do… This was a gradual coalescing and positions were slowly evolving.’34 The ‘no EU action’ arguments were slowly losing force in the face of arguments that reconstruction aid was consistent with the normative commitments listed above. The same official more or less described this process in his own words, by adding, ‘There was not a lot of backand-forth to be honest, it was more a common realization of what are we to do now, where can we be useful. It was a subtle process.’35 At this point we are able to focus more directly on the two NI hypotheses developed in Daniel Thomas’s theoretical framework (ch.2, this volume). Hypothesis 1: Entrapment First, the EU’s normative commitments played a direct role entrapping those with ‘no EU action’ preferences into accepting the idea of EUcoordinated reconstruction assistance to Iraq. After March 2003, the irreconcilable and open disagreements over whether military force in

Jeffrey Lewis 81

Iraq was justified faded into the background of discussions centred on what the EU could and should do next. An important background condition for the normative entrapment process to disempower the ‘no EU action’ viewpoint was the increased involvement of the UN in Iraq. As the UN became more involved in post-invasion Iraq, the ‘no EU action, period’ view (most strongly expressed by Germany) lost its clarity and this rhetoric fades from official policy pronouncements. Scepticism remained (especially among the ‘chocolate summit’ member states) but from March onwards, officials close to the discussions confirm that there was little conviction that nothing should be done. But for the original ‘veto players’ to EU action, obtaining UN legitimacy was an absolute prerequisite to EU involvement. According to one official, EU reconstruction assistance ‘was in a way legitimized by the UN as the UN became more involved’.36 Another participant was even more specific on the timing: ‘At the heart of the problem were the issues of legitimacy and impartiality. Hence the continual insistence on the UN’s role. That legitimacy was not in fact provided until UNSC Resolution 1483 was approved at the end of May37, and that in turn formed the basis for the Thessaloniki conclusions. It was also a point that was made very clear at Rhodes/Kastellorizo [an informal meeting among the EU foreign ministers].’38 Reconstruction assistance also resonated strongly with Europe’s shared values in promoting liberal democracy and the rule of law and this made the ‘no EU action’ views more difficult to maintain.39 As early as March 2003, the Political and Security Committee (PSC) ambassadors all ‘saw a clear responsibility of the EU, and an opportunity for humanitarian assistance, reconstruction, institution building and assisting the establishment of a more democratic order and civil society in postSaddam Iraq’.40 The Iraq reconstruction policy fit Europe’s ongoing humanitarian and economic assistance program, and anticipating this, Commission officials had begun preparing an Iraqi aid strategy back in December 2002, even before the military conflict broke out.41 In general, as Pace describes citing an interview with a Commission official in the External Relations DG, ‘EU reconstruction and aid projects are construed by EU actors as vital for peace to stabilize and civil society to flourish in devastated, conflict areas’ (Pace, 2007: 1046). By May 2003, those who viewed Iraq as the United States’ war and the United States’ problem were seeing their argumentative power losing force.42 A representative sample of this view is former French Foreign Minister Hubert Vedrine’s public comment that it is ‘hard to take responsibility for a war we thought was wrong, now that the

82 EU Policy on Iraq

United States finds itself in a traditional colonial trap’.43 And Richard Youngs cites an EU diplomat who claimed there were real concerns ‘not to be associated with a failure’ (2004a: 5). But over time the compatibility of reconstruction aid to the EU’s shared values empowered an EU-directed assistance program. In this sense, the EU utilized a policy failure to creatively recast Europe’s commitment to post-conflict reconstruction efforts. The EUJUST LEX integrated rule of law mission is clear evidence of this thinking, notably in the three extensions of this program’s mandate since 2005, which will now run through 30 June 2012 and increasingly experiment with ‘in country’ training events.44 EU member state participation rates in hosting EUJUST LEX training courses, seminars, and secondments has also become more inclusive from seven in the first year to now 20 of the EU27.45 Furthermore, reconstruction aid was rhetorically framed as a strategic investment to repair transatlantic relations. In the words of one highranking official, ‘Transatlantic relations were very bad and this would improve them globally.’46 Strategically, the motives here include both cost and benefit calculations. The costs for continued dissension are nicely summarized by John Peterson: ‘Iraq showed the EU to be both weak and divided as a collective… From a US perspective the cherrypicking of European allies was a remarkable success, despite overwhelming European public opposition to the Iraq war’ (2004b: 614). Normative reframing would enable the EU to dispel the image of an internally divided Union easily manipulated by the US. The benefits of reframing EU-Iraq policy were equally evident. That is, the EU’s concerted reconstruction program would help rebrand Europe’s credentials at projecting power (and especially ‘soft power’) internationally. Reconstruction helped reframe the issue with Iraq away from negative transatlantic disputes to positive and proactive EU involvement. Among EU member states, this new approach had tactical advantages which ‘expressly focused on the long-term structure of EU-Iraq relations to avoid short-term controversies’ (Youngs, 2004a: 3). Both the ongoing EUJUST LEX program to promote an integrated criminal justice system and a new Partnership and Cooperation Agreement (PCA) with the EU are encouraging signs of a longerterm perspective on Iraq and one that engages the US on the multilateral and rule of law principles so strongly articulated in EU policy.47 Broadly stated, the aim is to rebalance the imbalance in transatlantic relations by strengthening the European voice, and EU-led reconstruction assistance fits into that normative ideal. Finally, Iraq reconstruction assistance dovetails with Europe’s longterm strategic interests in promoting a stable, prosperous, and demo-

Jeffrey Lewis 83

cratic Middle East. In this view, as Peterson explains, Europe has real potential for ‘genuine partnership with the United States in terms of actual policy: that is, stabilizing, pacifying and eventually democratizing the Greater Middle East’ (2004b: 619). To the extent conflict resolution in the Middle East has ‘long been one of the EU’s top foreign policy priorities’ (Pace, 2007: 1042), EU reconstruction assistance was a substantive policy highly resonant with those normative goals. And likewise, those who advocated ‘No EU action’ positions found themselves increasingly entrapped by the logic of such a substantive policy connection between Iraq and long-term efforts at conflict resolution in the Middle East.48 Hypothesis 2: Cooperative bargaining A second enabling condition for EU reconstruction assistance is seen in the calculated effort by foreign ministers to renew a principled commitment for CFSP decision-making made in camera and by consensus. Following the unusually public violations of the procedural norms embedded in EU foreign policy-making (especially over the public letter campaign) EU reconstruction aid presented an opportunity for foreign ministers to reactivate the established institutional context for collective deliberations and ringfence the set of ‘bad practices’ that had emerged during the former months. After several weeks for ‘cooling off,’ a number of meetings were slated with the intent to lubricate the mechanisms for CFSP decision-making without per say attempting resolve divergence over Iraq or to decide any EU position on the post-invasion situation. Framing the sensibility of such a collective rapprochement on consensus-based foreign policy was the exit threat posed by the so-called ‘chocolate summit’ on 29 April between France, Germany, Belgium and Luxembourg to discuss the viability of an European-level military command centre under the general rubric of ‘enhanced cooperation’. Triggered by Iraq, and in large part a reaction to US ‘divide-and-rule’ tactics (‘old’ Europe quips, promotion of public letter writing campaigns), the praline summit broadcast another important message as well. Namely, that options of ‘exit’ among a core group of states should not be ignored if Europe’s military and security policy-making could not be improved. Of particular note in the renewal of CFSP norms in the wake of the Iraq crisis is the informal meeting of the foreign ministers on 2–3 May aboard a luxury yacht which travelled between the Greek islands of Rhodes and Kastellorizo. The informal meeting was a welcome relief to months of hostile exchanges and the ministers agreed to put their differences over military intervention in Iraq behind them. Emerging

84 EU Policy on Iraq

from the five-hour cruise on the second day of talks, EU foreign ministers had reached a common understanding on the need for systematic EU assistance in post-conflict Iraq as well as a broader green light for Javier Solana to formulate a new ‘EU Security Strategy’. According to one EU official, ‘It was the relationship with the US and NATO, rather than assistance to Iraq, which was at the center of the debate at Rhodes-Kastellorizo.’49 The same official went on to elaborate: ‘Kastellorizo did not, then, solve this issue but in approving the concept of a European Security Strategy it did set the basis for a more independent, more coherent, and more specifically European response to global threats. In short it adroitly succeeded in defusing the problem by gaining time, and in papering over the cracks by, effectively, changing the subject.’50 Another EU official confirms this general view: ‘Six months later we have our first Security Strategy, a direct outcome to this divergence. It served a double objective: overcome division over Iraq, and at the same time, deal with the US who says you are hard strategy wimps.’51 The foreign ministers also discussed the issue of Iraq reconstruction at a restricted lunch session of the 19 May GEARC.52 Before the 19–20 June Thessaloniki summit, the foreign ministers met on 16 June and requested Commission options for reconstruction assistance. And at the Thessaloniki summit, EU leaders endorsed a wide-ranging package of assistance based on these recommendations, declaring: ‘the [EU] stands ready to participate in the reconstruction of Iraq… [and] invites the Commission and High Representative to submit proposals for an EU contribution’.53 Procedurally, a new common policy on Iraqi reconstruction assistance was instrumental in reviving the shared commitment to normgoverned and consensus-based foreign policy. Close participants describe a ‘never again’ attitude among the foreign ministers and key advisors in the period of March–June 2003 and this intangible component to overcoming discord on Iraq may have the most significant long-term effects. A similar conclusion is reached by Puetter and Wiener who describe the Iraq crisis as less the result of ‘fundamentally diverging policy agendas’ than a coordination failure in the ‘framework for collective processes of norm contestation and interpretation’ (2007: 1080, 1084). It was precisely this framework which the discord over UNSCR 1441 in November 2002 deactivated and then later publicly scorned with the open letters. Irrespective of the substantive and normative reasons discussed above, a common policy on Iraq’s reconstruction was important to reach given the process-level stakes in recommitting member states to the EU’s decision-making culture.

Jeffrey Lewis 85

Conclusion: Europe’s norm- and consensus-based promotion of common foreign policies As numerous commentators have argued, agreement on a common position gives the EU greater influence internationally and makes it more difficult to ignore Europe’s voice (Peterson, 2004b; Crowe, 2003: 546). By this yardstick, the Iraq crisis was a resounding failure for the EU’s foreign policy aspirations. Particularly in the way that the thinly veneered common position agreed in January 2003 came almost instantly unglued with a public letter campaign, one could conclude there is a paucity of norms to guide and inform behaviour in EU external relations. But as case two above documents, the deliberative processes that led to a new consensus on EU reconstruction assistance suggests such a judgment may be premature or even wrong. Between March and June 2003, utilizing substantive and normative consistency reasoning, Europe’s foreign ministries quietly forged a new shared understanding that EU-coordinated reconstruction aid was important to rehabilitate transatlantic relations, ensure Europe’s longer-term role in the Greater Middle East and global affairs, and support Europe’s high standards for liberal democracy, human rights, and the rule of law. The key scope condition for a new mutual compromise on EU-level reconstruction aid was greater UN involvement, and hence, legitimacy. Without this, the potential veto players – Germany, France, Belgium, and Luxembourg – would have insisted on the status quo LCD reached in March 2003. Procedurally, the new common policy on EU reconstruction aid helped renew the long-established cooperative bargaining style of making foreign policy. The damage done to the EU style of cooperative bargaining in the first case raised the stakes of a status quo LCD outcome of ‘no EU-level action’ in the second. The trauma of the disagreement over UNSCR 1441 and the public letter campaign shaped the choices of the March–June 2003 period and contributed to a shared understanding among the foreign ministers that a new EU common action on post-invasion Iraq could help repair the institutional environment of norm-governed and consensus-based policy in deliberative, in camera settings.54 Europe’s security and defence policy has emerged from the Iraq crisis stronger and more amenable to consensus-based decision-making (Menon, 2004). The Iraq crisis ‘strengthened general support for a Europe able to look after its own security’ (Calleo, 2004: 35).55 Doing so will require updating the informal rules of consensus-based foreign policy decisionmaking. In most areas of EU policy-making, constructive abstention is

86 EU Policy on Iraq

a ‘normal’ way for dissenters to accept and live with collectively legitimated decisions – external policies, especially those with military and security implications, are still fairly new for ‘enhanced cooperation’ but there is good reason to hypothesize that the EU’s institutionalized culture of consensus can take hold here as well. Future foreign policy differences among EU member states will now take place in the shadow of the Iraq crisis, and member states who contemplate uncompromising foreign policy positions or public dissent from collectively legitimated positions will likely calculate their costs and benefits differently. In this way, how EU policy-makers overcame divisions on Iraq and utilized EU-coordinated reconstruction assistance as a means to revive consensusbased decision-making offers supporting evidence of the general pattern identified by Daniel Thomas in this volume’s theoretical chapter – namely, that ‘frequent and intensive consultation between member states has weakened egoistic identities and accustomed national policymakers to seeking out the views of their EU counterparts before determining a national position on a particular issue’. But the above analysis is not intended to paint too glossy a portrait of Europe’s Iraq policy. The findings offer an interpretation that is perhaps a touch too optimistic to this author’s eye, and it needs balancing with a conclusion that is upfront in noting that the EU still faces a ‘crisis of purpose over Iraq’ (Peterson, 2004b: 621). Below the rhetoric of reconstruction are substantive differences over the means and ends of what a long-term role for the EU in Iraq’s rebuilding effort should be. This case is particularly noteworthy for the high degree of normative coherence with a markedly lower degree of policy coherence. In terms of advanced liberal democracies promoting the rule of law and the values of civil and political liberties internationally, the lessons are unclear. Strong normative commitments to democracy and the rule of law do not necessarily translate into coherent policies promoting them. It does not help in this case that some EU member states retain what Richard Youngs calls a ‘legacy of opposition’ to a conflict they opposed to begin with, which works against more systematic EU involvement in Iraq (2004a: 1). While still premature to gauge whether EU reconstruction efforts will be a ‘success’, the embarrassment of the Iraq crisis to Europe’s common foreign policy aspirations may well be a critical juncture remembered more for a renewed commitment to consensus-based foreign policy decision-making and the collective legitimacy behind promoting UNcentred multilateralism, liberal democratic values and the rule of law. Just as the Balkans wars of the 1990s helped trigger the development of

Jeffrey Lewis 87

the CFSP pillar and ESDP, the Iraq crisis has the hallmarks of a similar long-term comprehensive rethink for the Union’s foreign policy aspirations and principled commitments to becoming a more coherent global actor.

Notes 1 For comments on earlier versions I thank the conference participants and especially Daniel Thomas, Anand Menon, and Simon Duke. 2 See also Crowe (2003: 534–5). 3 For current data on cumulative EU-Iraq reconstruction spending see, European Commission-External Relations, ‘EU Joint Strategy Paper For Iraq 2011–13 Concept Note’, available at: 2013_concept_note_en.pdf; Transcript of European Parliament Debate on 19 January 2010 in Strasbourg with Catherine Ashton, High Representative of the Union for Foreign Affairs and Security Policy/Vice-President of the European Commission, available at: do?type=CRE&reference=20100119&secondRef=ITEM-007& format=XML&language=EN; and European Parliament Answer to a Written Question on the Reconstruction of Iraq, 23 March 2010, E-0657/2010, available at: For current information on the EUJUST LEX mission, see: 4 The pioneering observations on the Council’s ‘code’ were made by neofunctionalist scholars, including Haas (1958: especially chapter 13); Lindberg (1970: 81–2); and Lindberg and Scheingold (1970: 95–8). For more recent analysis, see Hayes-Renshaw and Wallace (2006: 330–2); Westlake and Galloway (2004: inter alia pp.12–13, 54–5, 100–1, 256–62); and Lewis (2010: 656–7). See Ludlow (2006) for an insightful analysis of how the Council’s ‘code’ performed during the 1965–66 Empty Chair Crisis. For an analysis that covers the substantive area of foreign policy-making, see Smith (2004a: especially chapters 4–5). 5 See also Crowe (2003), Youngs (2006: 33). 6 Personal correspondence, former EU permanent representative, 12 February 2008. 7 Interview by telephone with author, 16 January 2008. 8 Nominally, this includes 5 of the EU15 and 15 of the EU27. 9 The ‘Letter of Eight’ was signed by the Heads of Government of Britain, Spain, Italy, Denmark, Portugal, Poland, Hungary, and the Czech Republic. Agence France Presse, 30 January 2003, accessed via Lexus/Nexus. 10 Associated Press, 17 February 2003, accessed via Lexus/Nexus. 11 The Financial Times, 18 February 2003, UK edition. 12 European Report, 1 February 2003, No. 2747, accessed via Lexus/Nexus. 13 European Report, 1 February 2003, No. 2747, accessed via Lexus/Nexus. 14 Global News Wire, 1 February 2003, No. A2003020855-3BB6-GNW, accessed via Lexus/Nexus. 15 Associated Press, 17 February 2003, accessed via Lexus/Nexus.

88 EU Policy on Iraq 16 ‘Europe: Making It Work’. Speech by Atzo Nicolaï, State Secretary for European Affairs, European Institute, Bucharest, Romania, 17 February 2003. Accessed online at: making_it_work.html. 17 Financial Times Information, Global News Wire, 1 February 2003, accessed via Lexus/Nexus. Part of the motivation for the later ‘Vilnius 10’ letter may have been that some of these countries would have signed on to the ‘Letter of Eight’ had they been asked. On 31 January 2003, Agence France Presse claimed that Slovenia, Slovakia, and Lativa have declared their postpublication support for the ‘Letter of Eight’ (accessed via Lexus/Nexus). 18 For a discussion of this point see de Schoutheete (2003). 19 European Report, 1 February 2003, No. 2747, accessed via Lexus/Nexus. 20 Interview by telephone with author, 16 January 2008. 21 European Report, 1 February 2003, No. 2747, accessed via Lexus/Nexus. 22 European Report, 1 February 2003, No. 2747, accessed via Lexus/Nexus. 23 European Commission, DG External Relations, Unit of the European Correspondent, ‘Note for the File’, Seminar on ‘CFSP Confronting Iraq’, at the EU Institute for Security Studies, Paris, 3 March 2003, dated 7 March 2003, p.2. 24 Interview by telephone with author, 16 January 2008. 25 Bernard Bot served as the Netherlands EU Permanent Representative from 1992–2002 and was Minister of Foreign Affairs from 3 December 2003 to 22 February 2007. His first posting to the Dutch permrep in Brussels was from 1964–70. 26 ‘Making Europe Stronger: Our Common Task’. Speech by Jaap de Hoop Scheffer, Paris, 14 May 2003. Speech available online at: static/actuell/toespraken/2003/05/making_europe_strongerx_our_common_task. html. 27 The Financial Times, 18 February 2003, UK edition. 28 The Financial Times, 18 February 2003, UK edition. 29 Presidency Conclusions, Extraordinary European Council, Brussels, 17 February 2003, 6466/03. 30 ECHO contributions between 1992 and 2003 were approximately 157 million euros (Iraq Assistance Program, 2004. COM (2004) 667, 4/3/04). 31 The Independent, 21 March 2003, p.10. 32 Council of the European Union, Presidency Conclusions, Brussels European Council, 20–21 March 2003, 8410/03/32. 33 Communication From the Commission to the Council and the European Parliament, Recommendations for Renewed European Union Engagement with Iraq, COM(2006) 283 Final, 7.6.2006. 34 Interview by telephone with author, 16 January 2008. 35 Ibid. 36 Ibid. 37 UNSC Resolution 1483 was adopted on 22 May 2003. 38 Personal correspondence with EU official, 5 October 2007. 39 Around this time, the European Council issued a statement on 16 April 2003 to celebrate the accession of the new ten member states. Known as the ‘Athens Declaration’, it states ‘our commitment to democracy’ is ‘the fundamental value underpinning the Union’.

Jeffrey Lewis 89 40 European Commission, DG External Relations, Unit of the European Correspondent, ‘Note for the File’, Seminar on ‘CFSP Confronting Iraq’, at the EU Institute for Security Studies, Paris, 3 March 2003, dated 7 March 2003, p.2. 41 I thank Daniel Thomas for suggesting this point. A parallel ‘prewar’ planning body for reconstruction assistance was initiated at the UN by the SecretaryGeneral ‘but was careful to keep the plan “secret” lest it suggest a view on the inevitability of war’ (Mac Ginty, 2003: 608). 42 Although such views did not become extinct. A year later, new Commission President Barroso found it necessary to state in an interview, ‘Some people in Europe may think that it is good that things are going badly for the US in Iraq. I really think that is an irrational and a bad policy’ (As quoted by the Associated Press, 20 August 2004, accessed via Lexus/Nexus). 43 National Journal, 20 December 2003, p.3833, cited in Peterson (2004a: 20–1). 44 Council Decision, 2010/330/CFSP, 14 June 2010. 45 EUJUST LEX Course Table, dated 31 December 2009, available at: http:// %20TABLE.pdf. 46 Interview by telephone with author, 16 January 2008. 47 Negotiations for the EU-Iraq Partnership and Cooperation Agreement (PCA) began in November 2006 and were concluded in November 2009. The PCA is scheduled to be signed during 2010. For more information, see http://ec. 48 On this reading, there are interesting similarities (and possible ‘learning effects’) to the logic of EU Palestinian assistance. See Pace (2007: 1046–7). 49 Personal correspondence with EU official, 5 October 2007. 50 Ibid. 51 Interview by telephone with author, 16 January 2008. 52 On the importance of restricted sessions and especially Council luncheons for CFSP decision-making, see Hayes-Renshaw and Wallace (2006: 60–1). On the general role of in camera meetings of the Council in facilitating deliberation and consensus-based decisions, see Stasavage (2004) and Lewis (2005). 53 Council of the European Union, Presidency Conclusions, Thessaloniki European Council, 19–20 June 2003, 11638/03/25. 54 For a discussion of how perceptions of CFSP ‘failures’ can lead to positive feedback effects, see Pace (2007: 1047). 55 Similar effects on public support for EU security and defence are attributed to the post-Cold War Balkans crises. See Youngs (2004b: 418) for a discussion.

Works cited Allen, D. and Smith, M. (2004) ‘External Policy Developments’, Journal of Common Market Studies, Vol. 42, Annual Review, pp.95–112. Calleo, D. (2004) ‘The Broken West’, Survival, 46(3): 29–38. Crowe, B. (2003) ‘A Common European Foreign Policy After Iraq?’, International Affairs, 79(3): 533–46.

90 EU Policy on Iraq Dimitrakopoulos, D. and Passas, A. (2004) ‘The Greek Presidency: In the Shadow of War’, Journal of Common Market Studies, 42: 43–6. Haas, E.B. (1958) The Uniting of Europe: Political, Social, and Economic Forces, 1950–1957. Stanford, CA: Stanford University Press. Hayes-Renshaw, F. and Wallace, H. (2006) The Council of Ministers. Second Edition. New York: Palgrave Macmillan. Lewis, J. (2005) ‘The Janus Face of Brussels: Socialization and Everyday Decision Making in the European Union’, International Organization, Fall 2005, 59(4): 937–71. Lewis, J. (2010) ‘How Institutional Environments Facilitate Cooperative Negotiation Styles in EU Decision Making’, Journal of European Public Policy, 17(5): 650–66. Lindberg, L. (1970) ‘Political Integration as a Multidimensional Phenomenon Requiring Multivariate Measurement’, International Organization, 24(4): 45–127. Lindberg, L. and Scheingold, S. (1970) Europe’s Would-Be Polity: Patterns of Change in the European Community. Englewood Cliffs, NJ: Prentice-Hall. Ludlow, N.P. (2006) ‘De-Commissioning the Empty Chair Crisis: The Community Institutions and the Crisis of 1965–66’, in J-M. Palayret, H. Wallace and P. Winand (eds) Visions, Votes and Vetoes: The Empty Chair Crisis and the Luxembourg Compromise Forty Years On. Brussels: P.I.E. Peter Lang, pp.79–96. Ludlow, P. (2003) ‘The European Council and Iraq: Beyond the Conclusions’, EuroComment Briefing Note, No. 13, 21 February. Mac Ginty, R. (2003) ‘The Pre-War Reconstruction of Post-War Iraq’, Third World Quarterly, 24(4): 601–17. Menon, A. (2004) ‘From Crisis to Catharsis: ESDP After Iraq’, International Affairs, 80(4): 631–48. Menon, A. and Lipkin, J. (2003) ‘European Attitudes Towards Transatlantic Relations 2000–2003: An Analytical Survey’. The University of Birmingham, European Research Institute. Prepared for the informal meeting of EU Foreign Ministers, Rhodes and Kastellorizo, 2–3 May 2003. Pace, M. (2007) ‘The Construction of EU Normative Power’, Journal of Common Market Studies, 45(5): 1041–64. Peterson, J. (2004a) ‘Keynote Article: Europe, America, Iraq: Worst Ever, Ever Worsening?’, Journal of Common Market Studies, Vol. 42, Annual Review, pp.9–26. Peterson, J. (2004b) ‘America as a European Power: The End of Empire by Integration?’, International Affairs, 80(4): 613–29. Pond, E. (2003) Friendly Fire: The Near-Death of the Transatlantic Alliance. Washington, DC: Brookings. Puetter, U. and Wiener, A. (2007) ‘Accommodating Normative Divergence in European Foreign Policy Coordination: The Example of the Iraq Crisis’, Journal of Common Market Studies, 45(5): 1065–88. de Schoutheete, P. (2003) ‘Du bon usage de la diplomatie’, La Libre Belgique, 29–30 March 2003, p.11. Smith, M.E. (2004a) Europe’s Foreign and Security Policy: The Institutions of Cooperation. Cambridge: Cambridge University Press. Smith, M.E. (2004b) ‘Institutionalization, Policy Adaptation, and European Political Cooperation’, European Journal of International Relations, 10(1): 95–136.

Jeffrey Lewis 91 Stahl, B., Boekle, H., Nadoll, J. and Jóhannesdóttir, A. (2004) ‘Understanding the Atlanticist-Europeanist Divide in CFSP: Comparing Denmark, France, Germany and the Netherlands’, European Foreign Affairs Review, 9: 417–41. Stasavage, D. (2004) ‘Open-Door or Closed-Door? Transparency in Domestic and International Bargaining’, International Organization, 58(2): 667–703. Toje, A. (2005) ‘The 2003 European Union Security Strategy: A Critical Appraisal’, European Foreign Affairs Review, 10: 117–33. Westlake, M. and Galloway, D. (2004) The Council of the European Union, Third Edition. London: John Harper. Youngs, R. (2004a) ‘Europe and Iraq: From Stand-off to Engagement?’, FRIDE Policy Paper, Working Paper 45, June. Youngs, R. (2004b) ‘Normative Dynamics and Strategic Interests in the EU’s External Identity’, Journal of Common Market Studies, 42(2): 415–32. Youngs, R. (2006) Europe and the Middle East: In the Shadow of September 11. Boulder, CO: Lynne Rienner.

6 EU Decision-making in CSDP: Consensus Building on Operation Artemis Simon Duke

This case study is designed to examine the central normative institutionalist assertion that decision outcomes can be shaped by institutions which may not be formal structures per se but ‘collections of interrelated rules and routines that define appropriate action in terms of relations between roles and situations’ (Peters, 2001: 28). Under normative institutionalism individual actors do of course make choices, but these are influenced by the prevailing normative values and the collective understanding of ‘appropriate’ action. With the passage of time, actors begin to associate more closely with the institutions and the values they embody, rather than the normal individual interest maximization that is typically associated with rational choice approaches. Normative institutionalism therefore posits that the constant shaping of norms and rules by the institutions will shape the decisions made by the actors within them. Applying this logic to the making of EU foreign policy, Thomas (ch.2, this volume) hypothesizes that negotiations among member states will be shaped by normative entrapment and cooperative bargaining. In the specific arena of the EU’s Common Security and Defence Policy (CSDP), normative institutionalism begs the question of what is understood by ‘the institutions’. Since it is a highly intergovernmental area, the member states have a particularly important role to play and it is therefore assumed that as they meet at different levels (working groups, Coreper, the Political and Security Committee, the General Affairs and External Relations Council) the appropriateness of a given action is defined. The ‘rules’ are also extremely important since, in the context of the Treaty on European Union or the principles of the UN Charter, the standards of appropriateness are established as a benchmark for the member states. Given the highly intergovernmental 92

Simon Duke 93

nature of CSDP, the institutions are relatively weak compared to the more communautaire areas of external relations. It is the combination of the ‘rules’ and the interaction through institutions that decisively shape the options selected by the individual actors. The case investigation, Operation Artemis conducted in the Democratic Republic of the Congo (DRC) in mid-2003, suggests that the decisions surrounding the operation were largely determined by normative commitments, both of a general and a more specific nature. Although there were varying degrees of interest in the operation from the individual EU member states, as well as differences in the physical ability to contribute to the operation, a number of pre-existing norms and commitments shaped their assumption of responsibility for the first wholly independent EU operation. The analysis suggests that normative entrapment was indeed present and shaped decision outcomes. It is argued that the extraordinarily permissive environment surrounding the operation (in particular, the backdrop of the bitter disagreements over Iraq) made the presence of normative entrapment all the more noticeable. Operation Artemis presents weaker support for the decisive presence of competitive bargaining, which would normally lead to a lowest common denominator (LCD) outcome at the EU level. The strongly norms-led nature of the lead up to the operation precluded much opportunity for ‘hard bargaining’, although there is some evidence of this at the national level, most notably in France. Although Artemis was limited in terms of scope and participation, it does not represent a LCD outcome, since the constraints on the operation resulted largely from the remit of the UN mandate and, to a lesser extent, from any decisive intra-EU competitive bargaining. The article is organized as follows. The first section puts Operation Artemis into context, with particular attention to the normative framing of the decision-making environment underpinning it. The second part looks at the origins of Artemis, both in terms of the DRC and its neighbours’ history but also the existing EU and national interests in the region. The third section considers the challenge from Kofi Annan, the UN Secretary-General, for an interim EU presence and reflects on the factors shaping the French response. This is then followed by an overview of the issues that led to the agreement of the other principal actors. The actual operation itself is covered en passant since it was brief and sandwiched between two UN operations and, by nature, inconclusive. The final section ponders the theoretical implications of the case study. It concludes that, whilst generally supporting the thesis that the normative and policy commitments already made by

94 EU Decision-making in CSDP

the member states had a decisive influence on the decision to prosecute Operation Artemis, the particularities of CSDP give rise to some additional dynamics. This suggests that particular considerations apply to the general political support for an CSDP operation and the actual physical contributions in the field.

Normative and policy context Normative framing on the French coast The normative framing of the decisions that led to Operation Artemis can be found at both the broad and detailed levels of commitment. At the wide-ranging level, the Treaty on European Union (TEU) advances as one of the objectives of the Common Foreign and Security Policy ‘to safeguard the common values, fundamental interests, independence and integrity of the Union in conformity with the principles of the United Nations Charter’. EU members will also seek ‘to preserve peace and strengthen international security, in accordance with the principles of the United Nations Charter, as well as the principles of the Helsinki Final Act and the objectives of the Paris Charter’ (TEU, 2002, Article 11). Although there are debates surrounding the specific role and responsibilities of the permanent members of the UN Security Council, the TEU nevertheless establishes a clear normative commitment to the UN Charter. The normative foundations of the EU’s external action was further reinforced by the European Security Strategy, which although formally adopted after Operation Artemis, nevertheless codifies what was already accepted amongst the EU member states when the draft version was discussed by the European Council in June 2003. The strategy states that, ‘The fundamental framework for international relations is the United Nations Charter. The United Nations Security Council has the primary responsibility for the maintenance of international peace and security. Strengthening the United Nations, equipping it to fulfil its responsibilities and to act effectively, is a European priority’ (European Security Strategy, 2003, emphasis added). At the more specific level, France and the United Kingdom had been instrumental in advancing the notion that the EU should play ‘its full role in the world’ at the St. Malo summit in December 1998. Both agreed, in the form of a declaration, to the development of CSDP in order to respond to the challenges of effective crisis management; the EU’s ‘military impotence’ having been made painfully apparent in the Western Balkans (Giegrich and Wallace, 2004: 166).

Simon Duke 95

It soon became clear that CSDP was to be framed in more international terms. Nowhere was this expressed more clearly than at the Anglo-French summit at Le Touquet in February 2003, where ‘the potential scope of CSDP should match the world-wide ambition of the European Union’s Common Foreign and Security Policy and should be able to support effectively the EU’s wider external policy objectives to promote democracy, human rights, good governance and reform’. The resultant declaration also recommended that the parties ought to ‘propose to our Partners that the EU should examine how it can contribute to conflict prevention and peacekeeping in Africa, including through EU autonomous operations, in close co-operation with the United Nations’ (Anglo-French Summit, 2003a: Para. 1). The Le Touquet statement also addresses a wide variety of tools for fostering stability in sub-Saharan Africa, including the need to prevent conflict and/or re-establishing peace, which are ‘of constant concern’ (Anglo-French Summit, 2003b). This evidence suggests an abundant normative ‘framing’ of the decision-making environment at both the general (the TEU, European Security Strategy) as well as the more specific levels regarding Africa (St. Malo and Le Touquet). The normative framing would subsequently shape the policy options considered by the individual EU members, but, just as importantly, it created the expectation from Kofi Annan that the EU would fulfil its commitments. The policy context Decisions can be shaped by normative elements and also by policy commitments made in the CSDP context. Four policy commitments, when combined with the aspects above, decisively shaped the decision-making environment. First, Operation Artemis was one of four actions undertaken in 2003. The time overlaps with one, Operation Concordia in the former Yugoslav Republic of Macedonia (fYRoM), would have consequences for Artemis. These can be summarized as twofold. First, the willingness to support politically and in material terms an operation in the DRC was partially conditioned by the relative importance attached to Concordia and by the individual resources and personnel tied up in the Balkans. The Copenhagen European Council in December 2002 had already indicated the EU’s ‘willingness to lead a military operation in Bosnia following SFOR’, drawing on NATO assets and capabilities (European Council, 2002). Concordia, which was strongly backed by the United Kingdom and the United States, made it more likely that Artemis would be framed as a French-backed demonstration of the EU’s autonomy in crisis management.

96 EU Decision-making in CSDP

The second factor to be taken into account was that Concordia was a ‘Berlin Plus’ operation, meaning that it utilized NATO assets. As the first CSDP military operation, it helped to prove that a partnership between the EU and NATO was possible. Yet Artemis was also the result of primarily French pressure to mount a ‘Europe only’ operation to prove a point about the EU’s capabilities. The issue of the EU’s autonomy vis-à-vis NATO came to the surface at a meeting on 29 April 2003 between the leaders of Belgium, France, Germany and Luxembourg (dubbed, unkindly, the ‘Chocolate Summit’ by Donald Rumsfeld). At this meeting, the idea of a ‘core group’, who might move ahead more rapidly on defence integration, including the possibility of establishing planning and command facilities for the EU was advanced. The summit was seen as not only provocative and anti-US, since all had opposed the US position on Iraq, but as damaging for NATO. The United Kingdom’s ‘Food for Thought’ paper was a strong rebuttal of these proposals. In terms of this case study, the spat illustrated the profound difficulties that CSDP could face if the two main underwriters of the policy failed to see alike. The third aspect that had a strong influence on the shaping of policy towards the DRC was the military intervention in Iraq, which caused bitter divisions in Europe. These divisions supported Operation Artemis by demonstrating the EU’s autonomy but also by promoting a positive image of the Union’s new CSDP. The divisions over Iraq, as well as the experience of operations in the Western Balkans, prompted the drafting of the European Security Strategy in June 2003 and its subsequent adoption by the European Council in December. This had the effect of further reinforcing a normative basis for action. Fourth, the discussions surrounding Operation Artemis have been portrayed as an attempt to avoid the marginalization of French foreign policy views, as compared to those of the United Kingdom, in the lead up to enlargement of the EU from 15 to 25 (Club des Vigilants, 2006). Disagreements over Iraq had emphasized differences between the EU members, particularly those who were more pro-US (and NATO) and those who were more pro-Europe. In the lead-up to Operation Artemis, France was far more careful in involving the candidates and seeking, and receiving, their support. The respective national decisions to contribute to Operation Artemis were therefore shaped by the confluence of the four inter-related considerations: the normative commitments undertaken in the EU context as well as the UN setting; the need to demonstrate a stand-alone capability at the European-level to contrast with the heavy dependence on

Simon Duke 97

NATO (and the United States) in the Western Balkans; the bitter divisions caused by Iraq made the need to demonstrate that the EU could live up to its responsibilities in the global arena a pressing concern and; finally, the prospect of an enlarged EU necessitated a more comprehensive and strategic response to the EU’s role in the world.

Origins of Operation Artemis Conflict had been endemic to the Ituri since 1999 with sporadic clashes between local, national and regional players over questions of land ownership and hence rights to natural resources. The DRC was of more than passing interest to a number of European companies especially since it became an oil producer in the mid-1970s (Petrofina, TotalFinaElf, Shell and others were all present onshore and, in some cases, offshore). The greatest wealth, however, lay in the extensive mineral riches. The conflict also had a regional dimension since the shifting alliances in Ituri involved Rwanda and Uganda. In 1996–97, both countries supported the Alliance des Forces Démocratiques pour la Libération du Congo (ADFL), which opposed Joseph Mobutu Sese Seko. In May 1997, Mobutu was overthrown by Laurent Kabila who, in turn, was confronted by a Rwandan-Ugandan-backed rebellion the following year. The first UN involvement was in August 1999, when 90 peacekeepers were deployed to promote the Lusaka Ceasefire Agreement reached earlier in that year. The agreement, signed between the DRC, Angola, Namibia, Uganda and Zimbabwe, rapidly fell apart necessitating an increase in the UN presence to almost 2,000 in February 2000 and, eventually to 8,700 in 2003, just prior to the EU presence. The Lusaka peace agreement was thrown into further disarray following Kabila’s assassination in January 2001. Joseph Kabila then succeeded his father. The Pretoria Agreement, between the DRC and Rwanda, was signed on 30 July 2002 and, on 6 September 2002, the Luanda Agreement between the DRC and Uganda. Under the latter, the withdrawal of the Ugandan forces was scheduled to take place following the establishment of a peace-building strategy by the Ituri Pacification Committee (IPC). The occupying Rwandan and Ugandan forces subsequently withdrew from the eastern parts of the country in October 2002. Shortly thereafter, in December, all parties agreed to form a government of national unity, which was eventually installed in July 2003.

98 EU Decision-making in CSDP

The IPC met in April under the supervision of the Mission des Nations Unies en République Démocratique du Congo (MONUC), which provided a road map for reconciliation and an interim administration. However, the plans soon went awry with the eruption of further violence in Ituri in May 2003, following the withdrawal of Ugandan troops who had provided much of the order and a semblance of stability in the District. Lendu militias (backed by Kinshasa) and the Hema-dominated (and Rwandan backed) Union of Congolese Patriots systematically slaughtered and raped civilians, while Hema-dominated militias seized Bunia in mid-May while Lendu militias terrorized the surrounding countryside. The Hema-Lendu conflict included an economic dimension as both fought for control over land and natural resources. The situation confronting the international community in 2003 was already grave, with an estimated 500,000 displaced people and up to 60,000 killed in Ituri alone. As a result of the deteriorating situation, thousands fled Bunia with many heading to the airport where a Uruguayan battalion had established its base as the MONUC sector 2 headquarters. The 712 Uruguayan troops had arrived in Bunia in April 2003 and were few compared to the Ugandan and local forces. The inability of MONUC to protect the civilian population as well as the rapid withdrawal of the Ugandan troops, contributed to spiralling violence the following month, including the murder of two UN military observers in Mongbwalu on 13 May. Annan’s challenge Two days after the observers were murdered, Kofi Annan addressed a letter to the President of the UN Security Council calling for the ‘rapid deployment to Bunia of a highly trained and well-equipped multinational force, under the lead of a member state, to provide security at the airport as well as to other vital installations in the town and to protect the civilian population’. He further stipulated that, ‘The force would be deployed for a limited period until a considerably reinforced United Nations presence could be deployed’ (Annan, 2003a). The letter also made it apparent that any such force would be authorized under Chapter VII of the UN Charter. The authorization came in the form of UN Security Council Resolution 1484, calling for the ‘deployment of an Interim Emergency Multinational Force [IEMF] in Bunia in close cooperation with MONUC’ (United Nations Security Council, 2003). The resolution further made it clear that the IEMF would be deployed on a ‘strictly temporary basis’ to allow the Secretary-General to reinforce MONUC’s presence in Bunia by mid-August 2003. The deployment of the IEMF would last until 1 September.

Simon Duke 99

In response to the worsening situation and Annan’s request, the Council requested the Secretary-General/High Representative on 19 May 2003 to study the feasibility of an EU operation in the DRC. The initial request for assistance led to the dispatch of nine French army officers to Bunia and Kinshasa to assess the types of force requirements needed to address the ongoing violence. The assessment concluded that any operation would be ‘high risk’ in political and military terms as well as ‘sensitive and complex’ (Astill, 2003). Perhaps unsurprisingly, Solana’s main focus at the time was the Western Balkans and the situation in Iraq; in public Solana made ‘no attempt to highlight African security challenges, and reaffirmed that the DRC was not on the EU’s “CFSP public radar” at the time’ (Hendrickson et al., 2007: 41). This strongly suggests that the early French commitment to lead the operation resulted in Annan’s appeal to Solana to build consensus amongst the EU members. The relatively short, yet significant, gap between the feasibility visit and the passage of UN Security Council Resolution 1484 allowed France to contribute to the shaping of the mission after ‘difficult debates among [the] French army’s Heads of Staff’ (Amnesty International, 2003: 8). For the French military, the acceptability of the mission lay in part upon the mandate being limited in terms of scope (primarily to make the airport safe and to protect Bunia) and time (until 1 September 2003). It was also agreed that, as a pre-condition, any operation would have to be granted a UN Chapter VII mandate and that the intervention force would have the official support of the DRC, Rwanda and Uganda. By the time the UN Secretary-General formally requested the UN member states on 30 May 2003 to provide a temporary stabilization force in the Ituri district, the feasibility of an EU operation had already been established. The Joint Action to launch Operation Artemis was agreed to on 5 June 2003. One day later, the first detachment of some 100 troops from a French marine infantry regiment arrived in Bunia (Council of the European Union, 2003).1 The timing would suggest that the political decision, made in the form of a CFSP Joint Action, was something of a formality since it was based upon an existing UN mandate and a direct request for assistance from the UN SecretaryGeneral. The Joint Action was nevertheless of twofold importance since it assured ‘une indéniable légitimité’ on the operation and convinced a number of member states, who otherwise had no strong interest in western Central Africa, to lend their support (Bagayoko, 2004: 103). The plan of operations was adopted by the Council shortly thereafter on 12 June. By 18 June, 400 soldiers were on the ground in Bunia and 500 more in Entebbe and, by 6 July, all military forces had been deployed. The

100 EU Decision-making in CSDP

intervention force was officially called the Force Multinationale Interimaire d’Urgence (or the Interim Emergence Multinational Force, IMEF) with the code name Artemis, deploying from 6 June to 7 September 2003. The headquarters (OHQ) were located at the Centre de planification et de contrôle des opérations (CPCO) near Paris. It was composed of CPCO officers, along with officers from other participating states and liaison officers from the EU Military Staff. A total force of 1,800 was divided into roughly equal portions between Entebbe/Uganda, the force headquarters and the remainder in Bunia; over half of the forces being French. In addition to France, the United Kingdom and Sweden contributed troops while Belgium and Germany provided non-combat soldiers. Seven countries contributed headquarters staff (Austria, Hungary, Ireland, Italy, Netherlands, Portugal and Spain). Non-EU contributors included Brazil, Canada and South Africa.2 The ultimate verdict of IMEF’s success depended largely upon the prospects of any longer-term stability under MONUC. It is also possible that IMEF’s limited and well-publicized duration, restricted its real effectiveness. Nevertheless Bunia, whose normal population was around 200,000, was around 45,000 at the launch of Artemis and it had climbed to around 100,000 as of September.

Negotiations within and among member states As evidenced, Kofi Annan’s request was to the international community but specifically to France. The task of shaping consensus in the EU context was heavily influenced by both the general and specific normative framing that enabled Annan to approach France with a reasonable expectation of a positive response. Consensus on what became Operation Artemis was shaped both at the member state level as well as within the individual member states. In the latter category none was more important than France in determining the nature and scope of the response. In order to understand the negotiating process one must first unravel the factors shaping the French response to Annan’s request. At a relatively early stage of planning, the Cellule Africaine de l’Elysée proposed that the operation should take place under the EU’s aegis.3 Prior to this, the Cellule had been approached by the International Crisis Group (ICG), which had dispatched representatives to New York and to Paris to voice its concerns about the potential for genocide – a particularly evocative word in the normative lexicon. The ICG suggested to Chirac, via the Cellule, the possibility of some form of multilateral

Simon Duke 101

intervention. Chirac, who hoped for an opportunity to launch a ‘Europe only’ mission, saw this as a good occasion (Lavallée, 2006: 12). French leadership in this particular case was also consistent with past policies whereby the government actively supported the deployment of a UN-backed peacekeeping force in the DRC. France had also contributed to MONUC in the eastern part of the country. On a more geo-strategic level, Niagalé Bagayoko has also suggested that Operation Artemis offered France a chance to re-establish a foothold in the Great Lakes region, and more generally central Africa. French influence had been waning since the 1994 Operation Turquoise in Rwanda. Turquoise was widely viewed as a biased and humiliating retreat of the French presence from the Central African Republic (Bagayoko, 2004: 104). The normative basis for the EU’ developing crisis management role had been laid out at St Malo, whereby the Union should have the ‘capacity for autonomous action’, notwithstanding any support for the strengthening of the European pillar of NATO. Arguably, the latter was demonstrated on several occasions in the Western Balkans, while the former had not. Hence, according to François Grignon the decision to intervene under the EU aegis lay in the ‘political weight [Artemis] could have to prove the value of an EU military capability for peacekeeping’ (Grignon, 2003: 4). The loi de programmation militaire for 2003–08, presented by French Defence Minister Michèle Alliot-Marie in September 2002, also shows that CSDP is ‘une ambition pour la France, pour l’Europe’ (Programmation Militaire, 2003). French ambitions in this regard are ‘equivalent to those of Europe’ (Rieker, 2006: 522). The third component in French policy was the role of the UN. Since Annan viewed the military intervention against Iraq as no less than ‘illegal’, France was the logical choice to turn to for assistance (Annan, 2003b). France was able to strengthen the normative basis for the operation while at the same time, France’s strong support of the UN strengthened the UN’s image, which had been damaged by the Iraq imbroglio. Dominique de Villepin, was effusive in his support of the organization when he observed that it is the UN ‘that will be tomorrow at the centre of the peace’ and that ‘in this temple of the United Nations’ the members are ‘the guardians of the ideal, the guardians of a conscience’ (De Villepin, 2003). Naturally, such rhetorical appeals had a ready constituency in the EU, reflecting not only the treatybased importance attached to the UN, but also the need to rebuild and restore confidence in the organization following the differences over Iraq.

102 EU Decision-making in CSDP

From the French perspective, Operation Artemis therefore offered the possibility of combining elements of national interest (and prestige), with the strengthening of CSDP, and the resuscitation of the UN. A grasp of the French policy background on this issue is critical to the understanding of the behaviour of other EU member states, especially bearing in mind that any use of military force under the TEU requires unanimity. If one considers the three levels outlined above – national, European, international – the positions of other key EU member states are fairly straightforward. First, the question of supporting French national interest was secondary to the offer of French military support for Operation Artemis as the designated ‘framework nation’. The concept had been endorsed by the Council on 24 July 2002 as the conceptual basis for conducting autonomous operations with recourse to a Framework Nation; the assumption being that the designated member should be in a position to pay for and prepare for the deployment of an CSDP operation. The limited number of member states with the capacity to underpin such a mission meant that the French offer was generally welcome as a means of reaching commonly held normative goals. Other larger member states made less than ideal ‘framework nations’ on political grounds since Germany, the Netherlands and the United Kingdom all supported the governments in Uganda and Rwanda. In 2003, the United Kingdom still had close relations with Rwanda, which again suggests that it was not the obvious candidate to assume leadership of a possible EU military operation. France, by way of contrast, had historically close relations with much of Africa but, with reference to the DRC, this was not based on any colonial past, nor any extensive economic interests. It was, according to an official French Foreign Ministry website, a special relationship founded in particular on the French language connection that ‘caused France to play a leading role in supporting the peace process, and, specifically, supporting the national transition processes’ (Ministère des Affaires étrangères, 2008). While competition with Belgium in the economic field cannot be ruled out as a factor shaping French policy, French interests in the DRC were more likely to have been spurred by considerations of the ‘francophonie’ in Africa, with the DRC as the largest francophone country on the continent. The care France took to frame its interests in a multilateral context also alleviated any concerns that this was solely a French operation in EU guise. The traditional differences between France and a number of its NATO allies did not come to the fore since NATO has few strategic

Simon Duke 103

interests in Africa, whereas the EU does. The African context was therefore important since it allowed the symbolic gesture to be made, but in an environment that was not overtly politicized for the US or the Alliance. It is also worth noting that the EU’s framework nation concept was modelled very closely on NATO and may have thus had a reassuring effect on any who suspected neo-Gaullist sub-plots. The desire to promote Europe’s global role, especially in juxtaposition to the international role of the US in Iraq, explains much of the motivation behind Belgian, German and Swedish support for France’s ‘framework nation’ role. The French desire to demonstrate a greater European presence on the international stage, also drew support from the EU Presidency. Papantoniou, the Greek defence minister, shared the view that Artemis could be a politically useful demonstration of the Union’s autonomy (Hendrickson et al., 2007: 40). The French arguments about the need to display some form of ‘Europe only’ capability resonated with the General Affairs and External Relations Council, which concluded that the operation would provide ‘further tangible evidence of the development of the European security and defence policy and of the European Union’s contribution to the international community’s efforts to promote stability and security in Africa’ (General Affairs and External Relations Council, 2003: 16). Germany was initially cautious about Artemis, which was in part a reflection of the traditional political nervousness about contributing to combat operations, as well as doubts about whether this should really be an EU operation. According to Ståle Ukriksen et al. (2004: 513–14) the German Foreign Minister, Joshka Fischer, came under direct British and French pressure to back Operation Artemis. Alister Miskimmon suggests that Chancellor Schröder’s highly public rejection of the Bush administration’s plans to invade Iraq led to ‘actively distancing Germany from America’s foreign policy position and actively allying with France and Russia against Bush’s plan’ (Miskimmon, 2006: 3). The Iraq crisis not only meant that Germany had given up its traditional middle ground in transatlantic and European affairs, but that it had surrendered much of its influence within the Union as a ‘balancer’. One of the options to regain influence was therefore to emphasize Franco-German relations in external relations. Germany’s political backing was facilitated by the fact that both France and the United Kingdom explicitly sought EU endorsement for the operation. However, Germany’s actual contribution remained limited in terms of exposure. The United Kingdom’s position was particularly difficult, given its heavy military commitments elsewhere. The origins of CSDP, with France and

104 EU Decision-making in CSDP

the United Kingdom at its core, meant that any French mission without the support and preferably involvement of the United Kingdom would have lacked credibility. Initially, the United Kingdom did not wish to be involved, but Blair had previously outlined a strong position on the use of force in humanitarian intervention in 1999 at the time of the United Kingdom’s intervention in Kosovo (Gegout, 2005: 438–9). Although the United Kingdom was engaged, its involvement was highly symbolic and value-laden but had the positive effects of, on the one hand, upholding its stance on intervention for humanitarian reasons (even if only in a support role) and, on the other hand, bolstering the commitments made at St. Malo to the development of CSDP. Belgium’s role, as the former colonial overseer, was discreet yet active behind the scenes.4 Any consideration of a more vigorous role in military terms was largely circumscribed by Belgian political nervousness about unilateral military involvement in former colonies after the death of ten Belgian troops in Kigali in 1994. As the result of a parliamentary enquiry subsequent to this event, it was agreed in a legislative document that Belgium ought not to send combat forces (troupes de combat) for active military service to former colonies (Sénat et Chambre des représentants de Belgique, 1998: 6–7). Belgium nevertheless provided transport and medical support and also made intelligence resources available to the mission.

Theoretical implications Normative entrapment and cooperative bargaining The complex normative basis is a key explanatory variable in explaining the decisions underpinning Operation Artemis. The extent of the normative commitments, ranging from the TEU and its invocation of the UN Charter, to the more specific commitments made at St Malo and Le Touquet, all created the expectation of support, if not actual physical involvement. It should also be borne in mind that the political environment in the spring of 2003 was extraordinarily normatively permissive, largely due to the ructions over Iraq. It would therefore have been difficult for any EU member to adopt a ‘no action’ position (with the specific exception of Denmark), especially since the interim EU mission was described by Annan himself as an act of humanitarian intervention. Even if individual member states, such as Germany or the United Kingdom, did not find the humanitarian intervention argument wholly convincing, they were prepared to accept the normative dialogue for other compelling political reasons,

Simon Duke 105

such as the need to demonstrate a measure of autonomy from NATO and the United States in light of the bitter disagreements over Iraq, or to demonstrate the capacity for autonomous action of ESDP (as CSDP was then known). EU member states, prompted by France and the United Kingdom, had also made bold policy commitments at St. Malo, which meant that Artemis was as much about shaping the EU’s global role as it was about strengthening the relatively new ESDP. The framing of primarily Anglo-French interests, as well as those of former colonial powers in Africa, as ‘European’ interests did not meet substantive objections from other EU members or even the candidates, most of whom expressed support. Even those countries that had no apparent direct interest in Africa, such as Ireland or Luxembourg, did so partially out of normative entrapment but also on the expectation that such entrapment would, at a later date, be repaid. The deliberate framing of the proposed military action in the DRC, both in terms of the EU’s formal norms and policy commitments, but also through the invocation of wider international regimes (such as those of the UN Charter), suggests that a strong element of normative entrapment was present. Cooperative bargaining offers a second, but less compelling, hypothesis to explain the emergence of the June 2003 Joint Action. With reference to European Political Cooperation (EPC), CFSP’s predecessor, Simon Nuttal noted the presence of an ‘automatic reflex of consultation’ that gradually modified the way in which member states acted in the foreign and security policy domains from EPC onwards (Nuttall, 1992: 312). As has been pointed out elsewhere, the disagreements over Iraq demonstrated a spectacular breakdown of any such reflex (even though, technically, it was not portrayed as a matter of ‘general interest’ under Article 16 of the Treaty on European Union and thus not subject to CFSP deliberation). Thus the cooperative bargaining thesis is weakened to an extent by the peculiarities of this case and the specifics of CSDP. The foreknowledge that France had volunteered to be the framework nation for what became Operation Artemis had two possible effects. First, it diminished the desire of the member states to reach dramatically diverse positions since political support for EU action in the DRC did not necessarily imply a physical contribution to any ensuing operation as this is always determined on a national case-by-case basis. In this sense, the early emergence of the French as the main provider for the operation not only facilitated agreement, but obviated the need for hard bargaining positions.

106 EU Decision-making in CSDP

Alternative approaches The above account offers an explanation of Operation Artemis drawing primarily on a normative institutionalist perspective. An alternative explanation is the competitive bargaining approach, which is perhaps the most realist-oriented of the explanations. This would reduce Artemis to a lowest common denominator decision, based on competitive bargaining and threatened or actual veto usage. The logic of this approach suggests that since France was willing to provide the bulk of the personnel and logistics, the choice for other EU members was stark: either block France or sign on the dotted line to an essentially French operation under EU guise. Catherine Gegout has been the strongest advocate of this position, arguing explicitly that Operation Artemis represented a lowest common denominator policy, or a French initiative masquerading as a ‘European initiative’ (Gegout, 2005). The ability to underwrite a CSDP operation might appear to give a handful of member states undue influence and scope for coercive bargaining or entrapment. It may also appear to suggest that the defence of norms and values depends upon the whims of a handful of members. In the case of Operation Artemis it could be argued that the strong French interest and operational capacity shaped EU norms and muscled less able (or interested) EU members into backing the operation. Less able member states may also have been dissuaded from hard bargaining, or even exercising a veto, by the knowledge that France could have mounted a unilateral operation with a UN mandate. Such a move would have weakened the EU at a time when it was anxious to demonstrate unity and cohesion rather than the individualism and dissonance that had been displayed over Iraq. One conclusion might therefore be that, aside from the normative veneer, French actions in particular were of a rational choice nature. Such realist-inspired arguments fail to account for the particularly strong normative environment in which the decisions about Artemis were made. All were aware of their UN-based obligations; all were aware of the high stakes for the EU if the fledgling CSDP was seen as circumscribed; and all were aware of the necessity for the EU to be a more visible presence on the international stage in the aftermath of the bitter disagreements over Iraq. There is simply little evidence of competitive bargaining and, even if not all were behaving entirely altruistically, the discussions and eventual Joint Action were framed in normative terms. A realist approach would tend to deal with CSDP operations as distinct events, each fuelled by its own competitive dynamics. But this fails to explain why Artemis resulted in far more cooperation at the

Simon Duke 107

European level (an ‘enhanced reflex’ to put it in Nuttall’s terms). Collaboration in the development of the European Security Strategy (ESS), the battlegroups concept and other various forms of enhanced cooperation in the CSDP context, are all attributable to the lessons gleaned from CSDP operations in 2003, including Artemis. In other words, Artemis cannot be seen as an isolated event but as part of a more complex policy learning process that contributed to the rapid development of CSDP with the support of the EU member states. The argument above may seem to support the policy-learning hypothesis and indeed there is some credence to this. This approach however assumes that the learning process does not affect or transform the fundamental properties of the actors involved. The chain of events set into motion by the Iraq case, Artemis and other CSDP operations suggest that a transformational ‘process’ was at work. In a similar vein, the explanatory power of the normative suasion approach is limited. The distinction between ‘suasion’ and ‘entrapment’ hinges upon whether normative arguments are provoked by an argumentative process between member states during the policy formulation process, as opposed to the conditioning of the same process by previously adopted normative policy positions. The case study suggests that previously adopted norms and policies had a strong role in shaping the decision outcome. There is little to suggest any serious normative competition.

Conclusions: Is CSDP distinct? The evidence from the Artemis case suggests that normative entrapment operated at two levels: first, the general and specific normative statements articulated in EU documents and within the institutions, made it hard to ignore the collective rhetorical commitments and; second, France’s role as one of the principal shapers of the normative environment created a reasonable expectation that it would respond as the ‘framework nation’. The levels therefore illustrate normative convergence as well as rhetorical entrapment at work. In this particular investigation, the ‘logic of appropriateness’ was put into unusually stark relief by the preceding disagreements over Iraq. The highly permissive normative environment surrounding Operation Artemis may well have been unusual but, even if an ‘extreme’ case, it nevertheless supports the pertinence of normative institutionalism to this particular instance. More recent cases, like the 2008 EUFOR Chad/RCA mission, may well suggest a far more nuanced conclusion.

108 EU Decision-making in CSDP

The possible distinctiveness of CSDP lies in the complicated two-level decision-making required to launch a crisis management operation. First, the political decision (a CFSP Joint Action), which is made at the level of the foreign ministers in the Council, and then, second, the national decisions to release the necessary assets to the responsible EU command structure for the duration of the operation. The growth in the number of CSDP operations and the increasing external demand for EU assistance could be perceived as a measure of the deep normative foundations of CFSP. In most operations the number of personnel or resources involved is modest. In these cases, the potential dissonance between the normative convergence and the actual availability of personnel and resources is not of great concern. It remains an open question whether the normative entrapment dimension was strengthened by the limited time and scope of Operation Artemis. Had Artemis concluded the EU’s active engagement in the DRC, this would have significantly undermined normative coherence that underpinned the operation and could even have made it appear a rather selfserving, and even cynical, involvement. To the EU’s credit, Operation Artemis was just the beginning of CSDP involvement in the DRC.

Notes 1 Due to its opt-out on all defence-related provisions on the Treaty on European Union, Denmark did not participate in the decision or the financing and implementation of Operation Artemis. 2 The force consisted of around 1,000 personnel in the Operational Headquarters, and a similar number on the ground in Bunia. The vast majority of forces were French, as were most of the 400 or so combat vehicles, with the United Kingdom and Sweden contributing around 100 and 70 respectively. The UK forces were primarily engineers and medics while Sweden contributed special forces. In addition Belgium sent 48 medical and logistical personnel who were located in Uganda, primarily with transportation responsibilities. Germany provided around 350 troops, again in Uganda, who gave logistical and medical support to the main force. 3 The Cellule Africaine de l’Elysée is a body with no official legal standing and is best described as an informal body, composed of diplomatic and military advisors, to oversee French interests in Africa. 4 The five permanent members of the UN Security Council, plus Belgium, Canada and South Africa, formed the Comité International d’Accompagnement à la Transition (CAT) whose task was to support the transitional government and encourage further progress.

Works cited Amnesty International (2003) ‘Democratic Republic of the Congo: Ituri a Need for Protection, a Thirst for Justice’, 21 October 2003, library/info/AFR62/032/2003.

Simon Duke 109 Anglo-French Summit (2003a) ‘Declaration on Strengthening European Cooperation in Security and Defence’, Le Touquet, 4 February 2003. Anglo-French Summit (2003b) ‘Declaration on Franco-British Cooperation in Africa’, Le Touquet, 4 February 2003. Annan, K. (2003a) ‘DR Congo: Letter from the Secretary-General to the Security Council’, 15 May 2003, Letter reproduced in full at rw/rwb.nsf/AllDocsByUNID/a074c19d1cd53a7ec1256d36004f2a85. Annan, K. (2003b) ‘Iraq War Illegal, says Annan’, BBC News, 14 September 2003, Astill, J. (2003) ‘French Say Their Congo Mission Will Have Little Impact on Fighting’, The Guardian, 10 June 2003. Bagayoko, N. (2004) ‘L’opération Artémis, un tournant pour la politique européenne de sécurité et de défense ?’, Afrique contemporaine, Printemps 2004, 101–16. Club des Vigilants (2006) ‘Opération Mamba: source de rayonnement européen pour la France’, 15 October 2006, 2006/10/operation_mamba.php. Council of the European Union (2003) Council Joint Action on the European Union operation in the Democratic Republic of the Congo, 5 June 2003, 2003/ 423/CFSP. De Villepin, D. (2003) ‘Address on Iraq’, French Minister of Foreign Affairs at the UN Security Council, New York, 14 February 2003. European Council (2002) ‘Presidency Conclusions’, Copenhagen European Council, December 12–13 2002, European Council Press Release 1591/02, 29 January 2003. European Security Strategy (2003) ‘A Secure Europe in a Better World: European Security Strategy’, Brussels, 12 December. Gegout, C. (2005) ‘Causes and Consequences of the EU’s Military Intervention in the Democratic Republic of Congo: A Realist Explanation’, European Foreign Affairs Review, 10(3): 427–44. General Affairs and External Relations Council of the European Union (2003) Press Release 11439/1/03 of 2522nd Council meeting, Brussels, 21 July 2003. Giegerich, B. and Wallace, W. (2004) ‘Not Such a Soft Power: The External Deployment of European Forces’, Survival, 46(2): 163–82. Grignon, F. (2003) ‘The Artemis Operation in the Democratic Republic of the Congo: Lessons for the Future of EU Peace-keeping in Africa’, Paper presented at ‘The Challenges of Europe-Africa Relations: An Agenda of Priorities’, Lisbon: 23–24 October 2003. Hendrickson, R.C., Strand, J.R. and Raney, K.L. (2007) ‘Operation Artemis and Javier Solana: EU Prospects for a Stronger Common Foreign and Security Policy’, Canadian Military Journal, 8(1): 35–42. Lavellée, Ch. (2006) ‘Repenser la Gestion de la Sécurité: Le cas de la politique européenne de sécurité et de défense (PESD) de l’UE’, 9e Symposium annuel des étudiants diplômés, Kingston/Canada, 27–28 October, 2006. Ministère des Affaires étrangères (Access 20 Feb. 2008) ‘France and the Great Lakes Region’, Miskimmon, A. (2006) ‘Germany and the Common Foreign and Security Policy Under the Grand Coalition in Germany’, Paper presented at the German Studies Association Annual Convention, Pittsburgh, PA, 28 September– 1 October.

110 EU Decision-making in CSDP Nuttall, S. (1992) European Political Cooperation. Oxford: Oxford University Press. Peters, G. (2001) Institutional Theory in Political Science: The New Institutionalism. London: Biddles Ltd. Programmation Militaire (2003) ‘Discussion d’un projet de loi’, Sénat, Séance du 14 janvier, #int640. Rieker, P. (2006) ‘From Common Defence to Comprehensive Security: Towards the Europeanization of French Foreign and Security Policy?’, Security Dialogue, 37(4): 509–28. Sénat et Chambre des représentants de Belgique (1998) Opérations en Afrique, 28 January, at Treaty on European Union (TEU) (2002) Official Journal of the European Communities, C 325/5, 24 December. Ukriksen, S., Gourlay, C. and Mace, C. (2004) ‘Operation Artemis: The Shape of Things to Come’, International Peacekeeping, 11(3): 508–25. United Nations Security Council (2003) Resolution 1484, 30 May 2003, http://

7 EU Membership Negotiations with Turkey: Entrapped Again Frank Schimmelfennig

The opening of accession negotiations with Turkey was one of the most controversial external relations decisions the European Union (EU) has ever taken. Whereas there is general agreement among the member states that Turkey is an important strategic partner and should be closely associated with the EU, the prospect of full membership was bound to arouse scepticism. First, the country is relatively poor and agricultural. According to figures for 2005, Turkey’s GDP per capita reached only 28% of the EU average – less than any of the member states.1 The Turkish agricultural sector is still the largest economic sector and employs around one-third of the workforce (as compared to an average of 5% in the EU). Turkish membership is thus likely to increase the divergence of living standards in the EU, create a high potential for labour migration, and instigate demand for high net payments from the structural and agricultural funds. Second, the impact of socio-economic divergence is magnified by the size of Turkey’s population. With more than 70 million inhabitants, Turkey would currently be the second largest member state and likely to surpass Germany by the time it joins the Union. Although the EU’s institutional rules reduce the effect of population size on political power, Turkey would rank among the big member states with regard to seats in the European Parliament and votes in the Council and could gain at least considerable blocking power. Third, Turkey’s European identity is controversial. Obviously, the largest part of Turkey’s territory is geographically located in Asia. The core issue, however, is culture. Even though the Turkish state is staunchly secular, Muslim religious beliefs are deeply entrenched in a population that has been relatively untouched by secularization and religious tolerance.2 In 111

112 EU Membership Negotiations with Turkey

addition, democracy – a core marker of EU identity – has been of a comparatively authoritarian and illiberal variety in Turkey. The strong involvement of the military in politics, the lack of rights for national minorities, limits on the freedoms of belief and expression, and the practice of torture have been among the most frequently named deficiencies. In sum, Turkey would increase heterogeneity in the EU considerably and add to the decision-making and distributional problems of the Union. Correspondingly, public scepticism and opposition toward Turkey’s accession has been strong in the EU. Support for Turkey’s membership has remained at only around 30% for the last decade – the lowest approval rating for all candidate countries. Finally, the decisions to give Turkey a membership perspective and to open accession negotiations have been highly controversial among member state governments and have tended to produce long and conflictive negotiations as well as uneasy compromises. After Turkey’s application for membership in 1987, the Commission formulated a negative Opinion in 1989. During the 1990s, a substantive group of member state governments blocked giving Turkey a general prospect of membership similar to that of the Central and Eastern European countries. When the EU finally accorded Turkey candidate status at the Helsinki European Council in 1999, deep divisions regarding the desirability of Turkish accession persisted. In 2005, before the EU opened accession negotiations with Turkey, Austria and Cyprus threatened to veto the decision. Reflecting the divisions among the member states, the Negotiating Framework declared accession negotiations as ‘open-ended’ and contained provisions for the suspension of talks. They envisaged no less than ten years of negotiations, long transition periods, and permanent safeguard clauses. Moreover, Austria and France announced that they would hold referendums on the accession of Turkey – the first referendums on enlargement in an old member state ever. Given the fundamental problems of Turkish membership, the popular scepticism and the strong and persistent divisions among the member states are hardly surprising. Rather, it is puzzling that the EU has been able to agree on offering Turkey the concrete prospect of EU membership in 1999 and opening accession negotiations in 2005. This is what the chapter seeks to explain. More precisely, it traces the process from divergent member state preferences to common policies toward Turkey. Rather than discussing once again the costs and benefits or the merits and liabilities of Turkey’s membership, and rather than explaining why individual governments are for or against the accession of Turkey, the chapter takes the divergence of member state preferences in the late 1990s as its starting point.

Frank Schimmelfennig 113

The chapter puts forward a normative institutionalist explanation (Thomas, ch.2, this volume) of how agreement on accession negotiations was reached and maintained in spite of persistent and, indeed, increasing differences among major member states. More precisely, it follows the entrapment hypothesis: Its fundamental community norms oblige the EU to consider all applications from European countries according to the same standards of liberal democracy. Turkey’s application could thus not be dismissed by reference to socio-economic or cultural incompatibility. To the extent that Turkey complied with liberal-democratic norms, member states opposed to Turkish membership for economic or cultural reasons could not legitimately block the path to accession but were rhetorically entrapped. This entrapment process was supported by the fact that the Commission was in charge of preparing the progress reports on Turkey. By basing its assessments and conclusions on the liberal-democratic accession criteria, it set the agenda for the ensuing intergovernmental negotiations and framed them in favour of norm-consistent decisions. With regard to the alternative hypotheses, the chapter will show, on the one hand, that governmental preferences have not converged over time. Even though some governments modified their positions, such changes resulted from changes in government – and not from policy learning or normative suasion. On the other hand, whereas competitive bargaining on the basis of exogenous preferences was indeed the dominant pattern of interaction, normative constraints prevented the process from resulting in deadlock or Lowest Common Denominator policy. The opening of accession negotiations with Turkey provides a ‘hard case’ for the normative-institutionalist explanation of enlargement developed in the context of the EU’s Eastern enlargement (Schimmelfennig, 2001; Sedelmeier, 2005). While the initial conditions are similar in both cases – divergent member state preferences and net costs for the EU in comparison with the status quo of association – the opposition to Turkish accession runs deeper and the potential costs of Turkish membership are higher. If it can still be shown that the decision-making process was effectively constrained, and that sceptical member state governments were rhetorically entrapped, the normative institutionalist explanation will be further strengthened. The study will consist of four sets of observations over time. The distribution of preferences ahead of the Luxembourg European Council in 1997 and the decision not to advance Turkey’s prospect of membership serve as the starting point. The process leading to official candidate status in 1999 constitutes the second set of observations. This is followed by the analysis of the (linked) decisions in 2004 and 2005

114 EU Membership Negotiations with Turkey

to open accession negotiations. Finally, I will include the decision of December 2006 to partially suspend the accession negotiations until Turkey fully extends the Customs Union to Cyprus. In all of these decisions, member state preferences have been divergent. Comparing two decisions advancing (1999 and 2004/2005) and two decisions inhibiting Turkey’s progress toward membership (1997 and 2006) gives us the opportunity to learn more about the conditions under which divergent preferences are transformed into common policies. On the one hand, the comparison shows that entrapment has increased in strength over time. While it has contributed to, but was not ultimately relevant for, the favourable decision of 1999, the decision to give Turkey an explicit membership perspective and link the opening of accession negotiations to the fulfilment of liberal-democratic political conditions set the stage for the rhetorical entrapment that was to follow in 2004/2005, when the Commission certified that Turkey had made significant progress in complying with the EU’s political norms. On the other hand, the 2006 decision demonstrates that rhetorical entrapment only works to constrain the sceptical member-state governments if candidates continue to comply with fundamental community norms and fulfil their promises. By refusing to implement the Customs Union with Cyprus, Turkey provided the principled opponents of membership with legitimate grounds to demand the suspension of accession negotiations. In the remainder of the article, I will analyse the preferences of the member states over time, develop theoretical expectations about the path from divergent preferences to common policies, and finally trace the intergovernmental negotiation processes of 1999, 2004/2005 and 2006.

Member state preferences The analysis of preferences is based on a selection of those member state governments that have actively participated in the debate during the period of observation by taking a public position for or against advancing or blocking progress toward Turkish accession. The selection privileges those governments with extreme positive or negative positions and is likely to neglect governments that were undecided or held moderate views. It is not the aim of this section, however, to give an accurate description or explanation of the distribution of preferences. Rather, it is intended to show that member states have held intense and highly divergent preferences on Turkish membership; that divergence has persisted over time; that changes in government preferences

Frank Schimmelfennig 115

have been the result of domestic developments; and that reluctance toward the accession of Turkey has increased rather than decreased during Turkey’s candidacy. Thus, this section provides evidence against the hypothesis of preference convergence as a result of policy learning or normative suasion. The sample contains the ‘big four’ member states Britain, France, Germany, and Italy, Turkey’s two neighbours Cyprus and Greece, and a few smaller countries from different groups of member states (Austria, Denmark, the Netherlands, and Poland). Table 7.1 shows the positions of the governments. The plus sign stands for a position in favour of advancing Turkey’s membership prospects; the minus sign indicates a preference for blocking these prospects. In 1997 and 1999, this meant being in favour of or opposed to giving Turkey a concrete membership perspective by according the country candidate status. In 2004/05, this referred to opening accession negotiations with Turkey. In 2006, the minus sign stands for suspending many negotiating chapters and creating institutional hurdles for resuming negotiations; the plus sign indicates a preference for not freezing negotiations at all or suspending only a few chapters. Table 7.1

Member state preferences on Turkish accession

Member state





Austria Cyprus Denmark France Germany Greece Netherlands Poland Italy UK

– n.a. – + – – – n.a. + +

– n.a. – + + + – n.a. + +

– – + +/– +/– – +/– + + +

– – +/– – – – – + + +

Source: Author’s compilation.

In 1997, the opponents of granting Turkey candidate status were in a clear majority. Principled opposition based on cultural grounds could be found among the Christian Democrat and conservative parties. In March 1997, the group of the European People’s Party in the European Parliament framed the EU as a Christian community and categorically excluded the membership of a Muslim country. The conservative heads of government of Belgium, Germany, Ireland and Spain supported this

116 EU Membership Negotiations with Turkey

declaration (Reuter, 2000: 51). Because of its territorial conflicts with Turkey and the Turkish occupation of Northern Cyprus, Greece was another principled opponent of Turkish membership. Other member states, in particular in Northern Europe, stressed the deficient human rights situation in Turkey. This position, however, resulted in a conditional rather than principled rejection of Turkish membership. Whereas Turkey could hardly change its Muslim population, it could do something to improve its human rights record. Moreover, this position would not have precluded linking progress toward membership to liberal democratic reforms. France and Italy appeared to be most inclined to grant Turkey a more concrete membership perspective. After the Luxembourg summit, both French President Jacques Chirac and Italian Foreign Minister Lamberto Dini expressed their regret that a more open policy toward Turkey – an important strategic partner for stability in the Mediterranean – had not been possible because of mainly German and Greek opposition.3 Italian Prime Minister Romano Prodi claimed to have been the only Christian Democrat leader to have dissociated himself from the principled, cultural opposition of his party family. Finally, while the UK has been the most consistent advocate of integrating Turkey, Austria has been an equally consistent adversary throughout the period of investigation. In light of this constellation, the most consequential change between 1997 and 1999 was the softening of the German and Greek positions. The causes of this change were predominantly domestic. In Germany, the centre-right government was replaced by a coalition of the Social Democrats and the Greens in 1998. Whereas the Christian Democrats continued their principled rejection of Turkish membership in the opposition, the new government did not share the religious-cultural concerns of its predecessors and advocated a proactive strategy to bring Turkey in line with European norms and closer to membership.4 In Greece, the hardliner foreign minister Theodore Pangalos was replaced with George Papandreou in 1999. He stood for a new foreign policy outlook advocating the inclusion of Turkey as a way of solving the security problems in the Aegean Sea cooperatively. In addition, the earthquakes in Turkey and Greece in the summer of 1999 brought about a wave of mutual sympathy and help in the Greek and Turkish publics, which supported a more accommodating foreign policy. The positions of the other member states remained basically stable (Önis, 2000: 473). Due to domestic changes and pressures in the member states, the overall constellation of preferences became less favourable to Turkish accession in 2004 and 2005.5 In France, President Chirac continued to

Frank Schimmelfennig 117

support the opening of accession negotiations but his own centre-right political camp voiced clear opposition. This weakened the French support for Turkey, which became even more lukewarm after the government’s defeat in the referendum on the Constitutional Treaty in 2005. In Germany, the situation changed in a similar direction when the Grand Coalition took over in 2005. Whereas in 2004, the ‘red-green’ government had been strongly in favour of opening accession negotiations, the new chancellor Angela Merkel and her party proposed a ‘privileged partnership’ rather than membership during the election campaign. In the Netherlands, Prime Minister Jan-Peter Balkenende and his foreign minister had advocated a quick start of negotiations but also came under increasing domestic pressure from their own centre-right camp, not least after the failed referendum on the Constitutional Treaty. The new member state Poland followed the UK (and US) lead in welcoming the eventual accession of Turkey to the EU. So did the Danish, Italian and Spanish governments. By contrast, the government of Cyprus (representing the Greek part of the Island) sought to block any further accommodation of Turkish membership interests unless Turkey fully recognized the Republic of Cyprus.6 Greece aligned itself with the Cypriot position and put its own territorial conflicts with Turkey on the agenda again. Austria threatened to block the opening of accession negotiations unless the Negotiating Framework explicitly named alternatives to full membership as the goal of negotiations.7 Finally, at the end of 2006, the member state governments remained split in similar ways over the question of sanctioning Turkey for not extending its Customs Union with the EU to Cyprus. Whereas the proTurkey faction advocated milder sanctions, the opponents of Turkish membership used Turkish non-compliance to demand the suspension of more than eight chapters or even a new consensual decision of the Council before accession negotiations could be resumed. Both Merkel and Chirac asked for an 18-month deadline for Turkey to comply with the Additional Protocol to be followed by a review of the accession negotiations in the Council. Cyprus, Greece, and Austria were also reported to favour tougher measures against Turkey. In contrast, the UK, Sweden, and Spain opposed further obstacles and would have preferred suspending less or none of the negotiating chapters. Denmark’s Prime Minister Fogh Rasmussen supported the partial suspension proposed by the Commission.8 In sum, three patterns can be detected. First, member government preferences have remained divergent throughout the period of investigation. Second, whereas the positions of some governments changed over time,

118 EU Membership Negotiations with Turkey

these changes are best explained by domestically induced changes in government and not as the product of international processes of learning or persuasion. Third, from 1999 to 2006, the distribution of governmental preferences has become more negative for Turkey. At the same time, the EU has paved the way for entering into accession negotiations. This opposite movement of preferences and policies constitutes the main puzzle to be explained.

Democratic norms and European promises: The normative context of EU-Turkey negotiations The institutionalized normative context of the European Union – its ‘community environment’ – constrained the EU negotiations on the question of Turkish accession in significant ways. It assigned different degrees of legitimacy to the divergent preferences of the member states and to their arguments for or against Turkish membership. It thereby strengthened the negotiating power of those actors whose preferences and arguments were in line with the community norms of enlargement and worked in favour of a norm-consistent negotiation outcome (Schimmelfennig, 2001). The core identity of the European Union as established in the Treaties is that of a pan-European liberal democratic community. States sharing this identity are legitimate members of the community. Correspondingly, the core enlargement criteria of the European Union state that candidates need to belong to ‘Europe’ and conform to the basic norms of liberal democracy. At the time of Turkey’s accession bid, Article 6(1) of the Treaty on European Union (TEU) declared that the ‘Union is founded on the principles of liberty, democracy, respect for human rights and fundamental freedoms, and the rule of law, principles which are common to the Member States’. Accordingly, the Treaty’s enlargement article (Article 49) stipulated: ‘Any European state which respects the principles set out in Article 6(1) may apply to become a member of the Union.’ These principles were reaffirmed in the so-called Copenhagen criteria of enlargement in June 1993, which required of prospective members ‘the stability of institutions guaranteeing democracy, the rule of law, human rights and respect for and protection of minorities’ and ‘the existence of a functioning market economy’. The Association Agreement between the European Economic Community and Turkey was signed in September 1963 and, in Article 28, envisaged that ‘the Contracting Parties shall examine the possibility of the accession of Turkey to the Community’, as soon as the economic

Frank Schimmelfennig 119

association and the customs union had ‘advanced far enough’. Although the Agreement did not prescribe an automatic transition from association to membership, it established an unequivocal membership perspective for Turkey. At the time, Commission President Walter Hallstein also clearly declared that ‘Turkey is a part of Europe […] geographically […] historically […] militarily, politically, and economically’.9 In 1996, the Customs Union between Turkey and the EU entered into force. Thus, the objectives of the Association Agreement had been put into practice. The only remaining legitimate criteria for blocking the path towards Turkey’s membership were the democratic and human rights conditions in the country. The entrapment hypothesis thus generates the following testable expectations for different stages of the accession process: 1. The EU officially judges and decides the eligibility of Turkey to accession negotiations on the basis of the democratic and human rights situation in the country. Conversely, all other criteria that shape the preferences of the member states and the debate on Turkey’s membership – be they religious-cultural, economic, geographic, or military-strategic – are of lower legitimacy and therefore not part of the discourse of the EU, or only marginally so, as represented in the official documents of the EU organizations. Moreover, the general eligibility of Turkey to EU membership is not questioned officially. 2. Turkey’s status of candidate for membership is a consequence of its recognized status as a ‘European’ country and the attainment of the preceding and preparatory institutional steps. Blocking candidate status becomes the more difficult to sustain, the more countries in a similar situation are granted a membership perspective. Advocates of candidacy status point successfully to the need of the EU to keep its membership promise (made from the Association Agreement onwards) and to treat all applicants equally according to the enlargement criteria of the EU. 3. The opening of accession negotiations depends on Turkey’s compliance with the constitutive political norms of the EU. Blocking accession negotiations becomes the more difficult to sustain, the more Turkey complies and the more the European Commission, the authoritative EU organ for reviewing the fulfilment of accession criteria, confirms compliance. Advocates of accession negotiations with Turkey point towards Turkish progress in meeting the EU’s criteria and call on the EU to keep its conditional promise of membership.

120 EU Membership Negotiations with Turkey

4. Progress in accession negotiations equally depends on compliance with the constitutive political norms of the EU. Only a breach of these norms (or the promises made to respect them) constitutes legitimate grounds for suspending or cancelling these negotiations. Opponents of Turkish membership are only entrapped as long as Turkey conforms to EU norms.

Granting Turkey candidate status: Strategic interests and preference change At the meeting of the EC-Turkey Association Council in April 1997, the EU reaffirmed that Turkey was eligible for membership and that it would be judged on the same criteria as the other applicant countries. In December 1997, however, the European Council at Luxembourg followed the Commission’s ‘Agenda 2000’ recommendations to exclude the country from the list of candidates for membership. The Commission justified its recommendation on the grounds that Turkey did not fulfil the Copenhagen criteria. The Freedom House ratings for liberal democracy listed in Table 7.2 reveal that this recommendation was entirely consistent with the fundamental community norms of the EU and did not constitute a discrimination of Turkey. On the scale of the Freedom House Index (FI), which runs from 1 (full liberal democracy) to 7 (full autocracy), Turkey scored a poor 4.5. A southeast European regional comparison shows that Bulgaria and Romania, which were given candidate status in 1997, scored significantly better (2.5 and 2), whereas Croatia, which was rated only slightly better than Turkey, was also excluded. Although only a few, mainly Northern, member states had focused on Turkey’s domestic political deficiencies as their main reason for refusal, the 1997 decision not to grant Turkey candidate status was thus in line with the community norms and could not be challenged Table 7.2

Turkey Bulgaria Croatia Romania

Freedom House Index for selected EU membership applicants 1997









4.5 2.5 4 2

4.5 2.5 4 2

4.5 2.5 4 2

4.5 2.5 2.5 2

4.5 2 2 2

3.5 1.5 2 2

3.5 1.5 2 2

3 1.5 2 2.5

3 1.5 2 2

Source: Freedom House, available at

Frank Schimmelfennig 121

legitimately by Turkey or the proponents of a more accommodating strategy. In its conclusions, the European Council reaffirmed Turkey’s eligibility to membership in principle and reminded Turkey that progress in its relations with the EU depended on the continuation of political and economic reforms. Thus, neither the culture-based and categorical opposition of the Christian Democrat governments nor the strategic interests of the UK and some of the Mediterranean countries were officially recognized as legitimate grounds. Two years later, at the Helsinki summit of 1999, the European Council concluded that ‘Turkey is a candidate country destined to join the Union on the basis of the same criteria as applied to the other candidate states’ and ‘will benefit from a pre-accession strategy to stimulate and support its reforms’. However, this change in EU policy cannot be attributed to political reforms in Turkey. Although the Helsinki Council welcomed ‘recent positive developments in Turkey’, the political situation had not substantially improved. In its Regular Report of 1999, the Commission found that ‘the situation concerning civil and political rights in Turkey has not evolved significantly’. Yet Turkey’s non-candidate status became more and more awkward as an increasing number of countries in Turkey’s neighbourhood obtained a membership perspective. First, the EU decided to open accession negotiations with another group of five candidate countries in 1999, including Bulgaria and Romania. The other countries of the Balkans were promised eventual membership at the Cologne Council in June of the same year. In this situation, the advocates of a candidate status for Turkey invoked the credibility of the EU and referred to its obligation to heed past promises. For instance, the new Commissioner for enlargement Günter Verheugen stated that ‘if there were no history to the Union’s relations with Turkey, we could consider a completely different strategy’, such as ‘a highly developed association. … History, however, precludes this option.’10 Whereas the advocates conceded that Turkey did not currently fulfil the political criteria to open accession negotiations, they claimed that candidacy status would provide a strong incentive to reform.11 It is questionable, however, whether these arguments were either necessary or sufficient to bring about the decision to grant Turkey candidate status. A more plausible explanation combines the perceived need to upgrade the Turkish status for strategic reasons with a change in pivotal member state preferences. For one, the member states were surprised by the harsh reaction of the Turkish government to their 1997 decision. Turkey refused to participate in the European Conference set up in

122 EU Membership Negotiations with Turkey

Helsinki for the ‘European states aspiring to accede to’ the EU, blocked meetings of the EU-Turkey Association Council, suspended talks on the solution of the Cyprus conflict, and threatened to veto the use of NATO facilities for EU military missions. There was a widespread perception that the EU had to make an accommodative gesture to safeguard the strategic partnership and to ensure Turkey’s cooperation on these important security issues (Önis, 2000: 470). Crucially, the change in German and Greek preferences made this concession possible (Müftüler-Bac and McLaren, 2003). Whereas the new German government under Chancellor Schröder had already declared itself in favour of ‘candidacy status’ at the Cologne Council of June 1999, Greece still prevented the Council from adopting any new conclusions on Turkey. It required the change at the top of the Greek foreign ministry and the improved climate after the earthquakes in the summer of 1999 to overcome Greek resistance. This sequence of events prior to the Helsinki Council demonstrates the relevance of exogenous changes in preferences. As a result, ‘for the first time in EU history, no country was determined to prevent Turkey from becoming an eventual member’ (Rumford, 2002: 55). In sum, the 1999 decision to accord Turkey the status of a candidate for membership was not the result of learning, suasion, entrapment, or cooperative bargaining. Rather, the domestically induced softening of the preferences of the two main opponents of Turkish membership moved the lowest common denominator from ‘no candidate status’ to ‘candidate status’. Thus, the 1999 decision is best explained in an intergovernmentalist perspective. At the same time, however, it created the conditions for entrapment to become effective in the future.

Opening accession negotiations: Normative coherence and rhetorical entrapment The common policy decided at the Helsinki Council reflected member state agreement on enhancing Turkey’s status but not a consensus on the desirability of Turkish membership. In addition, it was certainly facilitated by the perception that actual accession negotiations would be a matter of the distant future. Together with granting Turkey the status of a candidate, the EU emphasized that the opening of accession negotiations would require compliance with the Copenhagen political criteria on which Turkey had not made substantive progress in the past. At the same time, candidate status changed the institutional context of future decisions on Turkey’s membership. First, it considerably strength-

Frank Schimmelfennig 123

ened the rather vague membership commitment of the 1963 Association Agreement. Second, it strengthened the role of the Commission in the process. It was now up to the Commission to assess Turkey’s progress with regard to the Copenhagen criteria and to recommend the opening of accession negotiations. Third, it constrained the EU to use the same criteria for Turkey that it had used for the Central and Eastern European countries. Consequently, Turkey’s application would be judged primarily on the merits of democracy, human and minority rights, and the rule of law. Cultural, religious arguments were excluded from the assessment, and criteria of economic performance were of secondary importance only. This meant that Turkey could be certain to enhance its prospects for accession negotiations by improving its dismal human rights record and that it would become difficult for the principled opponents of Turkish membership to block the opening of accession negotiations if Turkey fulfilled the political criteria. As a result, the credibility of Turkey’s membership perspective was significantly enhanced: it became worthwhile for the Turkish government to engage in domestic reform. In October 2001 and August 2002, the Turkish government introduced two reform packages. By abolishing the death penalty in peacetime and allowing the teaching of Kurdish and its use in radio and TV broadcasts, Turkey fulfilled two core EU demands in the fields of human and minority rights. In its Regular Report of 2002, the Commission therefore concluded that ‘Turkey has made noticeable progress… [but] does not fully meet the political criteria’ (Commission of the European Communities, 2002: 139). Moreover, the Commission demanded to see the implementation in practice as well as further progress especially in the fields of freedom of expression, the fight against torture and civilian control of the military (Commission of the European Communities, 2002: 47). The European Council meeting in Copenhagen in December 2002 also welcomed ‘the important steps taken by Turkey towards meeting the Copenhagen criteria’ and concluded: ‘If the European Council in December 2004, on the basis of a report and a recommendation from the Commission, decides that Turkey fulfils the Copenhagen political criteria, the European Union will open accession negotiations with Turkey without delay.’ This promise demonstrates the institutional and normative constraints set up by the candidate status. The decision was to be based on the political Copenhagen criteria only, and the assessment on whether these criteria were met was delegated to the Commission. This tangible goal prompted the Turkish government to accelerate the pace of reform. In addition, the prospects for success were strongly improved after the

124 EU Membership Negotiations with Turkey

change in government to the moderately religious AKP, which was not part of the Kemalist establishment. Among other things, Turkey continued judicial reform (including the abolition of the State Security Courts), amended the Penal Code, and ratified Protocol 13 to the European Convention on Human Rights abolishing the death penalty under all circumstances. In 2003 and 2004, Turkey also tackled the two main remaining and most politically sensitive demands of the EU: the civilian control of the military and Cyprus. The National Security Council, formerly a military-dominated ubergovernment, was transformed into a civilian-led and consultative body. It was further decided that the defence budget would be monitored by parliament and that the military would cease to be represented in civilian fields such as education and broadcasting. Finally, in April 2004, the Turkish northern part of Cyprus accepted the United Nations’ Annan plan for a peaceful reunification of Cyprus. Turkey’s undeniable progress on the way to liberal democracy (see Table 7.2) and its compliance with the Copenhagen political criteria could, of course, not convince those that opposed the membership of Turkey in principle. Not only were the member state governments divided as ever (see Table 7.1) but two of the major leaders in favour of Turkish membership – French president Chirac and German chancellor Schröder – came under increasing domestic pressure. Whereas in the German case, resistance was mobilized only by the opposition, Chirac lost the support of his own party. On the one hand, Chirac (usually not an ardent supporter of Commission power) began to hide behind the reports of the Commission and the promises made to Turkey and claimed that governments had to follow the Commission’s recommendations and open accession negotiations if Turkey fulfilled the Copenhagen criteria. On the other hand, he began to cater to the domestic opposition by stating that France pursued a neutral stance toward Turkish membership12 and by conceding, in October 2004, that the ratification of Turkish accession would become the subject of a referendum. During the intergovernmental meetings at European level, however, he remained in line with the common policy. In its Regular Report of 2004, the European Commission presented a very detailed and balanced assessment of the political Copenhagen criteria. ‘In conclusion, Turkey has achieved significant legislative progress in many areas, through further reform packages, constitutional changes and the adoption of a new Penal Code, and in particular in those identified as priorities in last year’s report and in the Accession Partnership. Important progress was made in the implementation of

Frank Schimmelfennig 125

political reforms, but these need to be further consolidated and broadened’ (Commission of the European Communities, 2004a: 55). At the same time, the Commission’s recommendation was unambiguous: ‘In view of the overall progress of reforms, and provided that Turkey brings into force the outstanding legislation mentioned above, the Commission considers that Turkey sufficiently fulfils the political criteria and recommends that accession negotiations be opened’ (Commission of the European Communities, 2004b). In December of the same year, the European Council accepted the Commission’s assessment and followed its recommendation under two conditions. First, Turkey needed to adopt six additional pieces of legislation. Second, and more controversially, the Turkish government agreed to sign an Additional Protocol to the Ankara Agreement of association that would extend the Customs Union to all new member states including Cyprus. The Turkish Prime Minister Erdogan refused to sign the Additional Protocol at the summit but promised to do so before the actual opening of negotiations on 3 October 2005. At the same time, he insisted that this signature would not be tantamount to recognizing the Republic of Cyprus. The year 2005 brought more difficulties for the supporters of Turkey’s accession bid. First, the failed referendums in France and the Netherlands were widely interpreted not only as a negative vote on the Constitutional Treaty but also on the rapid enlargement of the EU. The opponents of Turkish accession used the referendum results as evidence for the need to change course. More directly, the referendum further weakened President Chirac. Second, the early parliamentary elections in Germany resulted in the defeat of the Social DemocratGreen government. The Social Democrats stayed in government with the Christian Democrats but had to accept Angela Merkel as chancellor, who had consistently campaigned for ‘privileged partnership’ rather than full membership. Thus, two key advocates of accession negotiations had become neutralized. Third, the Republic of Cyprus, which had the strongest stake in Turkey’s decision on the Additional Protocol, had become a member state. Thus, whereas the pro-camp was weakened, the anti-camp was reinforced by a country with particularly intense preferences. The opponents of Turkish membership could not deny Turkish progress on the way toward liberal democracy and could not legitimately call into question the Commission’s report and recommendation to open accession negotiations. However, they brought up alternative routes to block or prevent the talks and eventual membership. For one, they tried

126 EU Membership Negotiations with Turkey

to include alternatives to full membership such as ‘privileged partnership’ into the Negotiating Framework. In addition, they sought to exploit the Turkish reluctance to recognize Cyprus. Both attempts failed in 2005. Those conservative Commissioners that tried to include alternative objectives into the Commission’s proposal for the Negotiations Framework had no success. The proposal listed accession as the only and ‘shared objective’ of the negotiations – even though these would be open-ended. Moreover, the Commission reiterated that, in contrast to the extension of the Customs Union, the recognition of Cyprus under international law was not a precondition of accession talks.13 Turkey signed the Protocol extending the Customs Union on 29 July 2005. Yet French Prime Minister de Villepin as well as the Greek and Cypriot governments demanded the recognition of Cyprus as a precondition for opening accession negotiations.14 Moreover, the centre-right majority of the European Parliament refused to ratify the Additional Protocol because Cypriot ships were still barred from Turkish ports. After weeks of debate, the EU on 21 September agreed to require Turkey to normalize relations with the Republic of Cyprus ‘as soon as possible’ in the course of accession negotiations (rather than before their end, as the EU had demanded before). In addition, the Additional Protocol was to be fully implemented by the end of 2006. But the Cypriot government did not achieve its goal of making recognition (and the full implementation of the Additional Protocol) a pre-condition of opening accession talks. Neither did the Austrian government succeed in including alternatives to full membership in the Negotiating Framework. After ‘a pretty grueling thirty hours of negotiations’,15 accession negotiations were opened in the night from 3–4 October without significant changes to the Framework proposed by the Commission. In sum, the 2004/2005 decision to open accession negotiations with Turkey bears witness to the causal relevance of entrapment. The candidate status of 1999 anchored Turkey’s application firmly in the normative and procedural rules of enlargement. After 1999, it was clear that Turkey’s accession prospects would wax and wane with its achievements in liberal democratic reform, that the Commission would regularly assess the progress made and recommend opening negotiations when EU demands were met. When Turkey – rather unexpectedly and rapidly – demonstrated its willingness to reform and made substantial progress, the member states, including those that were principally opposed to Turkish accession, found themselves entrapped. After Turkey had fulfilled its part of the political accession conditionality deal, they were compelled to keep their part of

Frank Schimmelfennig 127

the promise as well and could not legitimately deny negotiations. They could also not legitimately change the deal by adding new preconditions or altering the promise from membership to something less. Last-minute attempts to this effect failed. In order to placate their dissatisfied domestic constituents, they could only put up a good fight and, as an instrument of last resort, promise them a referendum on Turkey’s accession. As a further indication of effective constraints, the new German chancellor, Angela Merkel, played according to the rules in spite of her domestic campaign against full membership for Turkey. However, the opponents of Turkish membership could only be made to acquiesce in accession negotiations as long as Turkey continued to comply with EU norms and keep its own promises. Otherwise, the opponents could seize the opportunity to block the path to Turkish accession. This became obvious in the course of 2006.

Suspending accession negotiations: Reverse coherence and entrapment With accession negotiations secured and parliamentary elections approaching, the reformist zeal of the Turkish government weakened in 2006. The Commission’s regular report revealed a mixed picture with small progress in many fields and stagnation in others. The main bone of contention, however, was Turkey’s refusal to fully extend the Customs Union to Cyprus. It continued to deny access to Cypriot vessels and aircraft (or those coming from Cyprus). The opponents of Turkish membership did not fail to seize this breach of international treaties and Turkish promises to demand sanctions. Now the supporters of Turkey were entrapped. The most they could try to achieve was to soften the tough sanctions demanded by the opponents. On 29 November, 2006, the Commission presented its recommendations on the continuation of Turkey’s accession negotiations: eight chapters relevant to Turkey’s restrictions on Cyprus should not be opened, and no chapter should be declared provisionally closed, until Turkey lifted the restrictions against Cyprus. However, the Commission left sufficient room for the negotiations to continue, rejected calls for a deadline or the suspension of the entire negotiations, and made sure that it would be the Commission rather than the Council that reviewed Turkish compliance.16 At their meeting on 11 December, 2006, the foreign ministers of the EU accepted the recommendations of the Commission. After the policy was adopted, the foreign minister of Finland (holding the EU

128 EU Membership Negotiations with Turkey

presidency at the time), stressed that ‘nobody questioned the possibility of Turkey joining the EU after it fulfils all the criteria’. On the part of the brakemen, Austrian foreign minister Ursula Plassnik made clear that ‘a number of areas of negotiation will have to remain on hold, but nobody will slam doors’.17 Within the camp of Turkey’s supporters, Spanish foreign minister Miguel Angel Moratinos conceded that ‘Turkey’s refusal to extend an EU customs pact to all 10 nations that joined the bloc in 2004 left the EU with no option but to take action’.18 In sum, the decision of 2006 demonstrates that rhetorical entrapment cuts both ways. As long as Turkey complied with EU norms, it backed the supporters of Turkish accession and constrained the sceptics. However, when Turkey failed to comply, it gave legitimacy to the claims of the sceptics to slow down the accession process and forced the supporters of Turkey’s membership bid to join in. At the same time, however, those that had asked for a complete halt or a principled review of the accession process did not get their way. Turkey’s accession prospects, the normative conditions under which they would be realized, and the role of the Commission in the process remained intact.

Conclusions The case of accession negotiations with Turkey is a hard case for Normative Institutionalism. Because of its geography, economy, and culture, Turkey is the most contentious candidate for EU membership ever. Because enlargement requires unanimity, member governments with strong preferences against Turkish membership could have stopped the process at every stage, as they had done in 1997. Yet entrapment has worked. Once Turkey was accepted as a candidate in 1999, when the constellation of member state preferences was particularly favourable, its further progress to membership has only depended on meeting the institutionalized enlargement criteria of the EU. When Turkey had gone a long way in fulfilling the EU’s demands, negotiations were opened, even though member state opposition had increased rather than softened in the meantime. Entrapment is a conditional process, however. First, its effects depend on the strength and legitimacy of the norm. In the case of Turkey, both the high constitutional status of the enlargement rules and the recent precedent set by the Central and Eastern European countries favoured entrapment. Second, it must be clearly established that the norm applies

Frank Schimmelfennig 129

to the case at hand. Relevance increased strongly when Turkey was recognized as a candidate for membership in 1999. Third, entrapment is facilitated when there is strong role of independent actors committed to the norm. The Commission’s role in enlargement demonstrates this. Finally, entrapment only works if the beneficiaries of the norm act appropriately as well. Once Turkey failed to comply with its promise to fully apply the Customs Union in 2006, the opponents of Turkish membership were free to block the negotiations.

Notes 1 GDP per capita in purchasing power standards for 2005 according to EU Rapid Press Release, STAT/06/166, 18 December 2006. 2 See Alber (2006: 376–7) based on data from the European Value Survey and the European Quality of Life Survey. 3 Süddeutsche Zeitung, 24 December 1997 (on Dini); Le Monde, 7 December 1997 (on Chirac); and press conference of Jacques Chirac, 13 December 1997 (available at 4 Süddeutsche Zeitung, 9 June 1999. 5 For an overview of government positions in this period, see the compilation by Sven Prange, available at standpunkte.shtml. Unless otherwise indicated, I draw on this compilation. 6 Süddeutsche Zeitung, 16 July 2005 and 3 September 2005. 7 Süddeutsche Zeitung, 1 October 2005. 8 ‘EU Leaders Polarized Over Turkey Talks’, Turkish Daily News, 30 November 2006; ‘France and Germany Step Up Pressure on Turkey’, EurActiv, 4 December 2006. 9 Cited according to Engert (2010: 32). 10 Günter Verheugen, Den Haag, 4 November 1999, SPEECH/99/151, Rapid Database. See also Der Spiegel 42/1999. 11 ‘I’m Sorry’, Turkish Daily News, 9 June 1999; Romano Prodi in a speech before the European Parliament, Süddeutsche Zeitung, 2 December 1999. 12 See e.g. Süddeutsche Zeitung, 10 April 2004. 13 Süddeutsche Zeitung, 30 August 2005. 14 Süddeutsche Zeitung, 3 August 2005; 3 September 2005. 15 UK Foreign Minister Jack Straw in a Press Conference on 3 October 2005, available at 16 EU Rapid Press Release, 29 November 2006, IP/06/1652. 17 ‘Turkey EU membership talks “partially on hold”’, EurActiv, 12 December 2006. 18 International Herald Tribune, 11 December 2006.

Works cited Alber, J. (2006) ‘Zehn Fragen zum EU-Beitritt der Türkei’, in J. Alber and W. Merkel (eds) Europas Osterweiterung: Das Ende der Vertiefung?. Berlin: edition sigma.

130 EU Membership Negotiations with Turkey Commission of the European Communities (2002) Regular Report on Turkey’s Progress Towards Accession, SEC(2002) 1412, Brussels, 9 October. Commission of the European Communities (2004a) Regular Report on Turkey’s Progress Towards Accession, SEC(2004) 1201, Brussels, 6 October. Commission of the European Communities (2004b) Recommendation of the European Commission on Turkey’s Progress Towards Accession, COM(2004) 656 final, Brussels, 6 October. Engert, S. (2010) EU Enlargement and Socialization. Turkey and Cyprus. London: Routledge. Müftüler-Bac, M. and McLaren, L.M. (2003) ‘Enlargement Preferences and PolicyMaking in the European Union: Impacts on Turkey’, Journal of European Integration, 25(1): 17–31. Önis, Z. (2000) ‘Luxembourg, Helsinki and Beyond: Towards an Interpretation of Recent Turkey-EU Relations’, Government and Opposition, 35(4): 463–83. Reuter, J. (2000) ‘Athens Türkeipolitik im Wandel. Griechisch-türkische Beziehungen vor und nach dem EU-Gipfel von Helsinki’, Südosteuropa-Mitteilungen, 40(1): 47–64. Rumford, C. (2002) ‘Failing the EU Test? Turkey’s National Program, EU Candidature and the Complexities of Democratic Reform’, Mediterranean Politics, 7(1): 51–68. Schimmelfennig, F. (2001) ‘The Community Trap: Liberal Norms, Rhetorical Action, and the Eastern Enlargement of the European Union’, International Organization, 55(1): 47–80. Sedelmeier, U. (2005) Constructing the Path to Eastern Enlargement: The Uneven Policy Impact of EU Identity. Manchester: Manchester University Press.

8 EU Policy on Economic Partnership Agreements: Trade… and Aid? Ole Elgström1

The empirical focus of this chapter is the internal negotiation processes within the European Union (EU) in its efforts to conclude Economic Partnership Agreements (EPAs) with 79 developing countries. I seek to explain two different policy outcomes: the agreements on an EU negotiation mandate in 2002 and on the Union’s final negotiation offer in 2007. EPAs are negotiated with six regional groupings of Third World countries (Central Africa, West Africa, Southern Africa Development Community, East and Southern Africa, the Caribbean and the Pacific) and are considered a novel form of cooperation with developing countries (Holland, 2002). The EPA talks are formally defined as trade negotiations and actual negotiations are handled by Directorate General (DG) Trade, guided by a Council negotiation directive. Because the Common Commercial Policy accords important policymaking powers to the Community, it is commonly expected that the EU should speak with a single voice in trade negotiations (Dür, 2006; Meunier, 2005). EPAs are, however, ‘not free trade agreements in the usual sense’ (Mandelson, 2007a). They are supposed to be ‘tools for development’ and reflect traditional aid-and-trade ties to former European colonies. The result is a situation with competing normative frameworks, where free trade principles co-exist with development concerns, but where trade representatives enjoy a significant formal, institutional advantage. How have the member states and the Commission in this particular context overcome divergent preferences to reach agreement on common negotiation positions? Member states have internally and, more importantly, sometimes also openly questioned the Commission’s free trade-oriented approach. The idea that free trade is the best pathway to growth and development, advocated by DG Trade and some member states, was challenged by more 131

132 EU Policy on Economic Partnership Agreements

‘development-friendly’ members who argued that the special needs of the LDCs and the goal of poverty eradication must be taken into account also in this kind of trade negotiation. They have with some success tried to modify the bargaining approach of DG Trade to include more of development concerns. The evolving EU policy in the area is therefore very much a negotiated construction. To explain the outcomes and the characteristics of the negotiation processes, I apply the normative institutionalist approach developed in this volume (Thomas, ch.2). The legal framework conferred significant institutional advantages to the Commission, legitimating its normative frame and thereby constraining especially the most consensus-sensitive member states. The constraining force of free trade principles was, however, weakened by the existence of a legitimate competing normative development framework. The result was a bargaining process, guided by the prevailing consensus norm and by a normatively-based concern to reach a negotiated agreement, with a mutual compromise as the outcome. To explain the public member state protests against the Commission’s bargaining approach, we need, however, to resort to a twolevel game approach (Putnam, 1988; Evans et al., 1993): this behaviour can only be understood, I argue, by examining the impact of domestic opinion. In brief, the EU’s policy can also in a case of commercial policy easily be interpreted in terms of normative institutionalism. An understanding of internal policy-making in terms of an institutional approach that situates EU decision-making in a normative setting serves us better than approaches that focus on competitive bargaining, policy learning or normative suasion. The description and analysis of the processes are based on interviews and documents. Semi-structured interviews, lasting approximately an hour and relying on a number of fixed questions and themes for discussion, have been carried out with senior officials in ministries and permanent representations to the EU of several member states (Denmark, France, the Netherlands, Poland, Sweden and the UK; in total 11 interviews) as well as with four DG Trade officials and with one official at DG Development. Background interviews were also carried out with four representatives of development NGOs in Brussels. All interviews were given under a promise of anonymity. I start by examining the history and background of the EPA negotiations, highlighting how the EU’s experience with previous aid-andtrade regimes resulted in policy commitments that led to its present trade-oriented approach. Next, I delineate the major lines of disagree-

Ole Elgström 133

ment within the EU camp and their associated competing normative frameworks. In the following section the process leading to the EU negotiating mandate is analysed, followed by an account of an increasing level of politicization and internal negotiations, which eventually resulted in final compromise proposals to the African, Caribbean and Pacific (ACP) countries. In these analytical sections, the negotiation process is interpreted with the help of the institutionalist approach and by references to two-level game theory. I then compare the explanatory power of this approach with that of the competing hypotheses set out in this volume’s theoretical chapter. I end with some concluding remarks.

The historical legacy For almost 30 years, the so-called Lomé conventions constituted the most significant part of the EU’s relations with the Third World (Babarinde, 1994; Grilli, 1993; Holland, 2002; Lister, 1997; Ravenhill, 1992). These aid and trade agreements were considered to be of major importance for the development efforts of a large number of countries – 46 at the time of the first agreement in 1975, 79 today – in ACP. Trade was in these agreements used as an instrument of growth and development, notably by abandoning the otherwise sacred General Agreement on Tariffs and Trade (GATT) principle of reciprocity. The ACP group thus enjoyed nonreciprocal preferential access to the EU market. The aid component aimed at improving living conditions in the recipient states. The size of the aid budget was determined by intergovernmental negotiations between the EU member states before each new agreement. The Lomé regime mirrored a European partnership identity that emphasized the special ties between the EC and its former colonies, a special responsibility for the EC and interdependence between rich and poor. The result was a rhetorical emphasis on Lomé being a contractual relationship between equal partners (Elgström, 2005b: 188; Grilli, 1993: 93; Ravenhill, 1992). In the early 1990s, a growing scepticism over the effectiveness of the existing Lomé regime developed. Decades of assistance and nonreciprocal trade advantages did not seem to produce the desired developmental effects. At the same time, the EU adopted the ‘Washington consensus’, reflecting the idea that a sound macroeconomic and political environment was a precondition for development, and also introduced structural adjustment lending and human rights conditionality into the fourth Lomé agreement (1990–2000) (Crawford, 1998: 132–7; Lister, 1997). These steps constituted an adaptation to prevailing

134 EU Policy on Economic Partnership Agreements

international norms and reflected a ‘virtual international consensus in favour of aid conditionality’ (Lister, 1998: 31). Though elements of the old partnership rhetoric remained, the basic power asymmetry between the ‘partners’ now became even more apparent. In 1996, the Commission published a Green Paper, which set forth wide-ranging changes. It introduced the revolutionary idea that future relations might be based on reciprocity, it stressed enhanced political dialogue on human rights and good governance, and ventured that the ACP group as a whole was perhaps not an ideal partner to cooperate with in corpore (Holland, 2002: 167–86). The EU’s positions in the negotiations that led to the Cotonou agreement in February 2000 reflected this new normative framework (Babarinde and Faber, 2004; Holland, 2002: 186–9). The Union now wholeheartedly adopted liberal principles of free trade and democratic governance. The concrete consequences were the EU’s insistence on the abandonment of the non-reciprocity, on WTO compatibility, and, in general, on the introduction of a trade regime based on liberal principles and on good governance. On all these points, the Union saw its general approach win through. On the other hand, internal negotiations within the EU (Elgström, 2005b) softened the text considerably, taking away the most radical proposals from the Green Paper. A coalition of member states, strongly supported by a network of development NGOs, that advocated an alternative normative framework with a focus on poverty alleviation and that questioned the benevolent effects of unfettered free trade, succeeded in introducing a social development perspective into EU policies and to retain elements of the existing ‘partnership spirit’ (Babarinde, 1994; Babarinde and Faber, 2004; Lister, 1998: 22). The Cotonou Partnership Agreement foresaw the initiation of negotiations on regional Economic Partnership Agreements (EPAs). The Parties to the Agreement had thus agreed ‘to conclude new WTO-compatible trading agreements, progressively removing barriers to trade between them and enhancing cooperation in all areas relevant to trade’ (Commission, 2002a; see also Article 37.5 of the Agreement). The EU had a strong commitment, partly based on its own success story, to regional integration as the best way to achieve development. In the words of Article 35.2 of the Partnership Agreement, ‘regional integration is a key instrument for the integration of the ACP countries into the world economy’. After strong pressure from the Commission, the members of the ACP, who had initially fought to protect the integrity of the ACP as a group, divided themselves into six regional groupings: Central Africa, West Africa, Southern Africa Development Community, East

Ole Elgström 135

and Southern Africa, the Caribbean and the Pacific. The aim was both to foster regional integration within these groupings and to set up these groupings as negotiating partners to the EU. Negotiations between the EU, represented by DG Trade, and these six regions started in 2002. The idea was to have EPAs in place no later than 1 January 2008, when the WTO waiver that covered the preferential aspects of the trade chapters of the Cotonou agreement was to expire. Although the creation of homogeneous EPAs, including all ACP states, was clearly the main goal for the Commission – this being the most efficient and least costly solution for the EU members’ commercial interests – alternative arrangements were to be considered for those ACP countries that preferred to stay outside.

Conflict dimensions, actor constellations and interests The major lines of disagreement within the EU concerned two issues: • The extent to which the EPA process should reflect development concerns, notwithstanding its official categorization as trade negotiations. • To extent to which the EU should give concessions and open its internal market to competition from ACP countries. The two conflict dimensions are partly over-lapping, partly cross-cutting. The first one reflects potentially competing normative framework, while the second is founded in the precarious balancing act between national material self-interests and normatively-based altruistic concerns. The EU has a strong general political and normative commitment to free trade. This holds true also for its Third World policy: A free exchange of goods and services is believed to be the best instrument to promote growth and development for less developed countries. In the words of former Trade Commissioner Peter Mandelson, ‘I believe in progressive trade liberalisation. I believe that opening of markets can deliver growth and the reduction of poverty’ (Mandelson, 2005). Helping the ACP countries to become a part of the global economy is therefore seen as fuelling a major engine of development. The liberal free trade principle is, however, not the only major EU norm that is applicable to the EPA negotiations. As described above, the EU’s relations with the ACP countries have always included strong elements of solidarity and an attachment to goals like poverty alleviation and social development. The objectives of the Cotonou agreement

136 EU Policy on Economic Partnership Agreements

reflect these underlying values: ‘to promote and expedite the economic, cultural and social development of the ACP states … The partnership shall be centred on the objective of reducing and eventually eradication poverty …’. Advocates of this normative framework contended that development aspects and the special needs of the ACP countries must be clearly taken into account, the essential trade character of the EPA negotiations notwithstanding. We may thus pinpoint the existence of two competing normative frameworks in the first conflict dimension. They compete in the sense that different actors put various emphases on the one or the other: all actors within the Union subscribe to both norms, but more or less enthusiastically. In brief, the so-called like-minded countries (Denmark, Sweden, the Netherlands and the UK) are the EU actors that most vociferously have advocated development concerns over the years (and are also well-known as relatively generous bilateral aid donors) while the Mediterranean countries (not to mention the new members in Central and East Europe) have been more passive in this regard. France holds a special position in the sense that it has been a major proponent of the Lomé regime since its initiation and is also the most persistent advocate of a special relationship between the EU and the ACP countries, not least its former colonies. On the other hand, France does not have the reputation of being an avant-garde aid donor. The main advocate of the free trade principle has been the Commission, in the shape of DG Trade, backed by for example Germany. Although in principle ardent supporters of liberal trade, the like-minded countries have voiced concerns that unfettered free trade could have negative consequences for the poorest ACP states, and that it therefore had to be cushioned with special treatment for the Least Developed Countries (LDCs) (cf. Elgström, 2005b: 193). This more sceptical view of free trade as a vehicle of development for LDCs has also been prominently represented by the development NGOs. The second bone of contention mirrors the existence of material selfinterests in trade relations with the ACP countries. Some member states have prominent agricultural interests that might be damaged by a further opening up of borders. France is one of these, but also Mediterranean producers of for example fruit and sugar, like Spain and Portugal. Northern member states are usually less apt to have strong material interests in products covered by EU-ACP trade. The Commission seems to be somewhat divided: while DG Trade would like to see a gradual reduction of EU trade protectionism, its officials have to reckon with the strong conservative interests of DG Agriculture.

Ole Elgström 137

It should furthermore be noticed that some member states have a deeper attachment to the traditional ACP links than others. France is the prime example, with its emphasis on ‘the spirit of Lomé’ and on continued responsibilities and unique ties to its former colonies, but also the other original founders of the Lomé regime have demonstrated similar sentiments. The Scandinavian countries have long-standing experiences as donors of foreign assistance to Africa and see the EU focus on the ACP countries as a natural continuation of this. On the other hand, Germany and the new member states have no such special attachments and would like to see a more global approach, while Spain and Portugal have a traditional focus on Latin America and Northern Africa. The Mediterranean and Eastern countries have been relatively uninterested in the EPA-negotiations and are apt to follow the Commission’s proposals when their material self-interests are not at stake. The potential de facto veto players in the EPA case, in the sense that these actors were key to any negotiated agreement, were France, the UK and the Commission. France was one of the most conservative actors in both conflict dimensions, and also one of the most powerful. The Commission was the most conservative player in the trade vs. aid dimension, where it strongly favoured a ‘pure trade deal’. The UK, on the other hand, is a major EU power and was strongly committed to change in both dimensions, wishing to open EU markets and to up-grade development concerns in EPA agreements. Its position as a ‘revisionist’ actor made, however, its ambitions problematic as the potential lowest common denominator outcome would have been the status quo in both cases: to maintain the predominant definition of EPAs as pure trade agreements, and to deny any further substantial opening of EU markets. When applying the theoretical framework of this volume to the case of EPA negotiations, the specific normative set-up of this case has to be borne in mind. EPA negotiations, situated in the borderland between trade and aid policy, are characterized by the simultaneous co-existence of two competing substantive normative frames. The existence of multiple, competing norms typically reduces the costs of violation and expands the range of acceptable behaviour (cf. Tallberg, 2004). Different actors may quote different sets of institutionalized norms, according to their policy preferences, and try to use these norms to their advantage in the negotiation game. On the other hand, the institutional arrangement – officially framing the EPA negotiations as trade negotiations and thereby giving the Commission the most prominent role – favours one set of institutionalized norms (free trade norms) to the disadvantage of the other (development norms). Thus, there is a dominant ideational frame, with

138 EU Policy on Economic Partnership Agreements

official blessing and higher normative status. The consequence is that the consensus norm and rhetorical entrapment are potentially very much relevant also in this case. Member states with a different value priority have to react to the dominant frame and justify their activities in relation to this framework.

Negotiations on an EPA mandate Negotiations to draft an EU mandate (formally, a ‘negotiation directive’) for the upcoming EPA negotiations took place in the ACP Group of the Council and in the 133-Committee (two working groups under the Council with responsibility for relationships with the ACP countries and for trade negotiations, respectively) in the Spring of 2002. Their deliberations were based on a draft recommendation, written by DG Trade (Commission, 2002a). While previous negotiations with the ACP countries had been conducted by the General Directorate for Development (DG Dev; formerly DG VIII), the task to conduct the EPAnegotiations was given to DG Trade. The reason was a recent major re-organization, in which the trade division of DG VIII was – after a highly acrimonious process – transferred (with most of its existing personnel) to DG Trade, leaving only development policy issues to the new, considerably weakened DG Dev (interviews). This organizational arrangement signified that the EPA negotiations were framed as a trade negotiation, rather than as a development, or even trade and development, negotiation. As a result, there was a high degree of coherence, mirroring EU trade policy norms, behind the EU mandate. The Commission’s draft recommendation on a mandate was in a sense a compromise proposal. Well aware of the various positions in the Council, the Commission produced a deliberately vague text that promoted both trade objectives and developmental concerns, although with a marked emphasis on the former. As regards access to the EU’s own internal market, DG Trade had the ambition to move EU positions forward towards a higher degree of openness, with the goal of promoting the EU’s image as a liberal actor and as a friend of the developing countries (interviews). The Commission’s aspiration in this area resulted in formulations that provoked reactions in member states with commercial interests in areas where competition from the ACPs could be expected. Council negotiations resulted in very limited changes to the Commission proposal (see Commission, 2002b). The prevailing decision-making rule, requiring member state unanimity to change the Commission proposal, is the main explanation for this. Objections did exist to the pro-

Ole Elgström 139

posal, but these came from different directions and focused on different aspects of the text. Thus, there was little prospect for consensus in the Council. The final negotiating directive on market access for ACP exports to the EU was, however, far less forthcoming than the Commission’s preliminary proposal (cf. European Research Office, 2002). It was only stated that EPAs ‘shall build upon and further enhance the market access conditions currently provided under the Cotonou Agreement’ (Council 2002, paragraph 3.1) and no concrete commitments were made. The member states that were less concerned with commercial self-interests (perhaps because they had so few of these) and that instead stressed the development concerns of the ACPs objected to this outcome. In a statement entered into the minutes, the Swedish, Danish and UK governments called for ‘access to the European market on EBA (Everything But Arms; e.g. open access) conditions for all ACP countries’ (European Research Office, 2002). This unusual formal marking must be considered a very strong political signal and an expression of intense normative disapproval of the majority position. The decision by the like-minded countries nevertheless to agree upon the mandate as a whole can be partly explained by their commitment to the consensus norm. Member states are in general very reluctant to openly disagree if they face a clear EU majority. All three countries wanted the EU to present a united front to the outside world in an area, which they believed was important and valuable also for the LDCs. Therefore, there was little incentive for them to deny consensus on the proposed mandate. They also wished to support DG Trade’s ambition to create a more open trade regime and their desire to avoid further demands from the more protectionist camp led them to accept the rather trade-centred approach of the mandate, and to abstain from efforts to insert more ‘development friendly’ formulations (interview). It could also be argued, however, that their passivity in this regard could be explained by reference to normative entrapment, as the three countries may have been disempowered by their pre-existing commitment to the ‘Washington consensus’. The evidence available does not allow us to distinguish between the two possible normative institutionalist explanations. The very broad formulations of the mandate in practice meant that DG Trade was given rather free reins in its conduct of the negotiations. DG Trade officials have seldom experienced that the formulations of the mandate have curtailed their freedom of action (interviews). Rather, they have been able to use the directive in the internal debate as a way of countering criticism of their way of handling the negotiations (‘we

140 EU Policy on Economic Partnership Agreements

just do what we are required to do according to the mandate’). DG Trade could in this way portray its actions as consistent with agreedupon common norms. The mandate has thus functioned as a policy commitment, to which member states – but also the Commission itself – have felt obliged to adhere.

Norm-based negotiations: A clash between competing values Negotiations between the EU and the six regional ACP constellations started in September 2002. On the EU side, DG Trade was to begin with in full control and handled the negotiations without much interference from member states, which after having agreed upon the mandate demonstrated little interest in and commitment to the ensuing negotiations. They had in general little material interest in the EPA negotiations and were at the same time involved in other, arguably more important trade negotiations. DG Development was preoccupied with its own problems and left the EPA talks to the trade directorate. Among DG Trade officials, EPAs were widely seen as ‘just another free trade agreement’. They tended to conduct the negotiations as if they were ‘a traditional trade negotiation’ (interviews). This framing reflected the organizational culture of this Directorate General. It also mirrored a normative conviction. The directorate’s guiding philosophy includes a strong belief that ‘opening of markets can deliver growth and the reduction of poverty’ (Mandelson, 2005), and that an integration of the ACP into the liberal world economy is needed. DG Trade’s approach came, however, as a shock to many ACP representatives, who were used to the more development-friendly and accommodative negotiation mode in DG Development (interviews). DG Trade was perceived as a confrontational negotiator that paid little attention to development concerns while pushing the developing countries to open their markets to EU exports (interviews). Many commentators contrasted the development-friendly language in public speeches by Commission top officials with EU attitudes and praxis in the concrete negotiations. Even the ACP Council of Ministers, in an official declaration on the EPA negotiations, regretted ‘the disconnect between the public statements of the Commissioners of Trade and Development on the development aspects of EPAs and the actual position adopted during EPA negotiating sessions’ (ACP, 2005). The complaints from the ACP countries, and perhaps even more evidently from an increasingly active and vocal development NGO

Ole Elgström 141

community, finally reached officials and decision-makers in the likeminded member states, which in the EPA case has been semi-formalized into what is known as ‘friends of the EPA’ group (interviews). The core of this grouping includes delegates from Denmark, the Netherlands, Sweden and the UK, but sometimes also representatives from France, Belgium and Ireland. The ‘friends of the EPA’ may be characterized as an informal coalition, based on a normative consensus, that serves as a forum for discussions and information exchange, and which ordinarily seeks to coordinate potential activities. In this case, however, joint actions were slow to appear. The negotiation mandate that all member states had agreed upon just a few years ago constituted a policy commitment that limited their propensity to act. The Commission was not, after all, acting contrary to its directives, only interpreting the mandate in a particular way. Furthermore, the free trade objectives stressed by the Commission had as such strong support from most member states; it was the one-sided emphasis on this particular normative framework, to the detriment of the development framework that the like-minded members disliked. Many factors thus spoke against member state action. Nevertheless, the UK trade and development ministries in March 2005 issued a joint statement in which they recommended that ‘the EU should take a non mercantilist approach and not pursue any offensive interests’, proposed that the EU ‘should make an upfront offer of complete duty and quota-free market access to each ACP regional group, with no strings attached’ and that ‘investment, competition and government procurement should be removed from the negotiations’. They also asked for additional resources ‘to enable the ACP countries to benefit from trade reforms and build their export competitiveness’ (DTI and DFID statement 2005). This extraordinary statement was a challenge to the Commission’s policy and could be (and was, by Commission spokespersons) interpreted as criticism of the existing negotiating mandate. To understand the British initiative, we need to direct our attention to the British domestic arena. In 2004, the ‘stop EPA’ campaign, headed by Oxfam but supported by a large number of British and European development NGOs and by some political celebrities, had been initiated. The campaign struck down on the perceived lack of a development focus in the negotiations and criticized the Commission for its ‘self-serving trade and investment liberalisation agenda’ that did not reflect ‘the interests and needs of ACP countries’ (The Guardian, 2005). Intensive lobbying occurred in connection with the Labour Party congress and in the run-up to the British elections. The UK inspired Commission for Africa, which

142 EU Policy on Economic Partnership Agreements

had been set up in advance of the UK G8 and EU Presidencies, had published a report that was very critical of EU trade and agriculture policy. The UK declaration should be understood against this background, as a tactical move in a two-level game. Another possible interpretation is to see the declaration as a normatively-based reaction to the one-sided free trade approach of the Commission from a government with a competing normative agenda. The public nature of the reaction and its timing speak, however, in favour of an interpretation that sees the statement primarily as a response to domestic pressure. The Commission reacted strongly to the British initiative. In a note for the attention of EU delegates in ACP countries that was leaked to the press, the move was condemned as a ‘major and unwelcome shift in the UK position’ and as ‘counterproductive as it could well make progress with EPA negotiations more difficult by reinforcing the views of the more sceptical ACP states’. Using normative framing tactics, DG Trade portrayed the UK statement as inconsistent with previous commitments: ‘Some recommendations move well away from agreed EU positions set out in the Cotonou agreement and negotiating directives. Others are not compatible with WTO agreements’. In an effort to delegitimize the position of the two British ministries, it was noted ‘that their statement is contrary to the agreed EU position and harmful for our common objective’ (Commission, 2005); an accusation that the UK forcefully denied, pointing to the vagueness of the negotiating mandate and claiming that its stance was compatible with the directive. The UK also continued its strategy of using public channels to influence the Commission. In an open letter of October 2006 to EU development and trade Commissioners Michel and Mandelson, its trade and development ministers stressed their concern about the current state of the negotiations and expressed their wish ‘to see these agreements deliver real benefits to ACP countries’ (Financial Times, 2006). The UK ministers echoed the worries that they had previously expressed on market access: ‘The EU must also allow ACP countries as much time as they reasonably need to open their own markets, while providing effective safeguards to prevent unfair competition from subsidised European products.’ This move may be interpreted as an attempt of normative persuasion where the UK referred to ‘unobjectionable’ norms (Elgström, 2005a) like ‘fairness’ and solidarity with poor countries – both elements of formally recognized EU norms – in order to undermine the Commission’s insistence of a ‘pure trade deal’.

Ole Elgström 143

The member states that were critical of the Commission’s normative approach also utilized their formal institutional possibilities to raise their concerns. Heated debates took place in the ACP Group and the 133-committee where the Commission’s strategy was openly questioned. An irritated and pressured Commission representative linked the critique to the on-going (and in the eyes of the Commission unfair) NGO campaign, exclaiming ‘who has written your comments? Is it Oxfam?’ (interview). To underline the importance of social development aspects of EPAs, the like-minded countries distributed non-papers to the ACP Group and to the Commission. Countries with established contacts into DG Trade used these informal networks in an attempt to influence the future approach of the Commission (interviews). In brief, we see in this phase of the negotiation process intense arguing and efforts to persuade others of one’s own normative standpoint. There is, however, no evidence that actors were open-minded and prepared to change their priorities. This speaks against the normative suasion hypothesis.

Concessions and normative convergence: Towards a mutual compromise In a tactical move that in the end proved decisive, Denmark, backed by the other like-minded countries, made an attempt to raise the issue to a higher political level, the Council of Ministers, trying to get the Council to issue a principled statement that EPAs were primarily to be seen as ‘tools for development’. The underlying logic was to produce an authoritative normative declaration, which the Commission would be bound – at least politically – to pay attention to. The move was initially only partly successful; the wish to have the Council issue a clearly formulated statement finally resulted in a watered-down, non-committing formulation in the Council Conclusions (interviews). Nevertheless, repeated high-level political expressions of concern for the development aspects of EPAs are claimed to have gradually affected Commission behaviour: from 2006 DG Trade put much more emphasis on development aspects, primarily in its verbal behaviour but also in practice (interviews). It also started to invite DG Development to participate in both negotiations with the ACP groupings and in the ACP- and 133-committees and the two General Directorates began to present a common outward front. In the final year of negotiations, the Commission also went back from its earlier insistence that development aspects were not to be included in the EPA agreements, as they were dealt

144 EU Policy on Economic Partnership Agreements

with in other parts of the Cotonou agreement, and agreed to include ‘development chapters’ in the final texts. These changes in Commission attitudes and practice mirror an insight that an increased consideration of development aspects was necessary to ensure backing, and final ratification, by member states, nota bene prominent ones like Germany, the UK and France (interviews). It may also be interpreted as a consequence of rhetorical entrapment (Schimmelfennig, 2003). In its efforts to defend its negotiation position, the Commission increasingly began to emphasize its positive attitude towards development-oriented agreements (Mandelson, 2007a; interviews). Picturing the EU policy in rosy colours, not only being instrumental to liberalization and growth but also to social and human development and taking into consideration not only EU interests but also the interests of impoverished people in the ACP countries, opened the EU both to demands for negotiation concessions and to post-negotiation accusations of not living up to its promises. Proclaiming EPAs a tool for development made it more difficult to withstand ACP and Member State demands for an unambiguous development profile in the agreement. In the end, the Commission realized that full-fledged EPAs would be impossible to achieve within the time-limit set. The result was a decision to offer the ACP countries and regions ‘EPAs lights’, or ‘interim EPAs’, restricting the agreements to cover only goods, market access and some development aspects in a first step. The aim, however, was still to conclude comprehensive agreements. The Commission also concluded that interim agreements would not necessarily be signed with all ACP regions, but with ‘all interested ACP countries and regions’, thus opening the possibility that ‘in some regions, not every country member will be able or willing to sign an agreement now’ (Mandelson, 2007b). Interim, two-step agreements, that almost totally reflected the EU offer, were initialled with most of the ACP partners in December 2007. The only full EPA agreement was concluded with the Caribbean ACP states. The interim agreements focused on trade in goods and included a decision to remove all remaining tariffs and quotas to the EU for all exports from the ACP, but with transition periods for sugar and rice and a gradual liberalization in ACP countries while excluding a number of sensitive products from liberalization. They also included chapters on development that endorse a range of development-supporting measures. Commitments on assistance in specified forms, quantities and time-frames are, however, not part of the agreements.

Ole Elgström 145

To summarize, the EPA negotiations departed from a formal institutional setting that legitimated one normative framework (EPAs as free trade agreements) to the disadvantage of another (EPAs as development tools). The dominance of the trade frame and the power of the consensus norm have discouraged member state opposition and made them play along with the Commission most of the time. The Commission’s use of references to the negotiating mandate when attacked, and its repeated references to shared ideals of free trade (also as the best way to reach development) have constrained member state action. But only to a certain limit: the existence of a legitimate competitive normative framework has rendered the Commission’s framing tactics less binding. And when domestic political forces have been sufficiently strong, member states have chosen to break the ranks and openly criticize the Commission’s negotiating approach. These initiatives also made heavy use of normative framing tactics. Over time, repeated expressions of high-level political support for development concerns seem to have influenced Commission attitudes, leading to a negotiation approach that increasingly reflected both normative frameworks. The final EU offer to the ACPs reflected these changes: development provisos were included as was a market access proposal that at least partly satisfied the like-minded countries. This outcome can be interpreted as a mutual compromise, based on entrapment and consensus concerns. Although the internal EU negotiations can hardly be called cooperative there is thus still impressive evidence that supports the hypotheses of the normative institutionalist approach.

Alternative explanations An interpretation of the EPA negotiations based on a normative institutionalist framework has been offered above. How well then do the competing theoretical approaches presented in this volume (Thomas, ch.2) fare in comparison? The competitive bargaining hypothesis, predicting tough negotiations and a lowest common denominator outcome based on the veto-right of member states, faces special problems in issues of trade as member states do not have any veto in EU commercial policy – ratification is in the EPA case decided by qualified majority in the Council of Ministers. Nevertheless, the Commission still has to calculate its chance of having a final agreement approved by the Council (interviews). Based on the discussions in the ACP Group and the 133 Committee, it has to assess whether objections to specific points are so strong that it needs to adapt its proposal. In practice, the

146 EU Policy on Economic Partnership Agreements

major members – France, Germany and Great Britain – must be on board to produce approval, and are therefore de facto veto players. In the internal EPA negotiations, however, both member states, including the great powers, and the Commission have made concessions. In our first case, agreement on a negotiation directive, member states that were sensitive to an opening of European markets managed to water down the Commission’s initial formulations, while the Commission was more successful in keeping its ‘trade, not aid’ approach to the negotiations. In the second case, agreement on a final EU offer, development concerns were taken on board in a compromise satisfying both the Commission and the like-minded countries. Market access for all ACP states (‘no tariffs, no quotas’) was guaranteed, albeit with two, time-limited exceptions. In this way, free trade activists and protectionist forces were both given an acceptable solution. In brief, the outcome is a mutual compromise, rather than a lowest common denominator-like solution. The competitive bargaining perspective – with its foundations in Realist theory – has in general serious problems in dealing with normbased state behaviour. In the Realist tradition, states are not driven by values or normative concerns. Therefore, a Realist would have a hard time explaining why EU member states strive for altruistic objectives, like fairness and poverty eradication, but also to understand the normbased striving for consensus, demonstrated in this study. The approach is better placed to explain behaviour that can be linked to material objectives, and thus has no difficulty in interpreting EU member states’ concerns with market access and no problem in accounting for the final decision to allow exceptions for rice and sugar in the EU’s final offer on market access. The policy learning and normative suasion hypotheses are not well fitted to an analysis that emphasizes norm competition rather than increasing norm convergence. In the issue-area at hand there is little of ‘thick understanding’ although many of the values heralded by EU aid-and-trade negotiators are on the surface shared by all member states. But beneath the surface of general norm adherence we have detected various priorities, reflecting different ways of understanding the basic mechanisms of north-south relations. The participants in the internal EU debate on trade and development have well-grounded convictions, based on long traditions of national development assistance and can therefore not be said to be in a situation of uncertainty. New member states have with a few exceptions been relatively unengaged in the discussions. The changes that have occurred among the participants seem to be based on rational calculations on what is needed to reach a viable compromise, rather

Ole Elgström 147

than preference change based on policy learning or normative suasion. Unanimity was reached on the EU negotiation mandate and on a final EU offer to the ACPs but this consensus does not mirror an underlying convergence of values but rather a mutual compromise between advocates of different normative frames.

Conclusion: How to interpret the EPA negotiations? The question of agreement despite divergent preferences is most satisfactorily answered by relying on a normative institutionalist approach. It is the existence of an institutionalized consensus norm and the expected negative consequences from breaking this that help us understand the continuing desire for a common EU foreign policy also in this field. Member states, and especially the most consensus-sensitive ones, tend to follow the agreed official EU policy line despite conflicting preferences and priorities. Efforts, by the Commission and by leading member states, to normatively frame the debate, and to rhetorically entrap other actors by references to previous commitments are ubiquitous and had demonstrable consequences for the outcomes. Member states do take account of the potential effects of their actions, in terms of potential praise or shaming, and adapt their behaviour to the institutional setting. Still, I have in this text argued that this holds true only up to a certain limit: if sensitivity to domestic constituencies is high enough and the issue gets politicized, then member states may break ranks regardless of institutionalized normative constraints, especially if the issue-area is linked to proclaimed key self-images of the government. Domestic policy concerns thus constitute a scope condition in this case. Another scope condition is external pressure: the Commission’s reluctant decision to include development sections in the final offer was arguably not only the result of internal member state activity, but also of pressure from the EU’s counterpart, the ACP, whose main bargaining objective was to include strong and effective development provisos in the final agreement. Basically, the normative institutionalist framework advocated in the Introduction to this volume fits also in this chapter – despite this being a case of trade negotiations. The definition of EPA negotiations as a trade negotiation did give definite institutional advantages to the free trade policy principles of DG Trade. However, the existence of a normatively legitimate alternative made these norms less compelling, encouraging normative dissent. The result was, as predicted by the framework, a negotiated mutual compromise. As trade policy generally – also in the WTO context – ‘is no longer about just trade’ (Young, 2007), but has

148 EU Policy on Economic Partnership Agreements

increasingly become framed as a complex, multi-dimensional issue-area and not least as an instrument to foster development, it may well be that the patterns discovered in the EPA case may apply also to EU decisionmaking behaviour in trade negotiations more widely. On the other hand, EPA can be seen as atypical trade negotiations, and as an ‘easy case’ for normative explanations, in the sense that EU member states in general have relatively limited economic interests in EU-ACP trade. The wider relevance of our findings from the EPA case is therefore a question for future research.

Note 1 The author would like to express his gratitude to the Bank of Sweden Tercentenary Foundation for financial support.

Works cited Books and articles Babarinde, O. (1994) The Lomé Conventions and Development. Aldershot: Avebury. Babarinde, O. and Faber, G. (2004) ‘From Lomé to Cotonou: Business as Usual?’, European Foreign Affairs Review, 9: 27–47. Crawford, G. (1998) ‘Human Right and Democracy in European Development Policy’, in M. Lister (ed.) European Union Development Policy. London: Macmillan. Dür, A. (2006) ‘Assessing the EU’s Role in International Trade Negotiations’, European Political Science, 5: 362–76. Elgström, O. (2005a) ‘Consolidating “Unobjectionable” Norms: Negotiating Norm Spread in the European Union’, in O. Elgström and C. Jönsson (eds) European Union Negotiations. London and New York: Routledge, pp.29–44. Elgström, O. (2005b) ‘The Cotonou Agreement. Asymmetric Negotiations and the Impact of Norms’, in O. Elgström and C. Jönsson (eds) European Union Negotiations. London & New York: Routledge, pp.183–99. Evans, P.B., Jacobson, H.K. and Putnam, R.D. (eds) (1993) Double-Edged Diplomacy – International Bargaining and Domestic Politics. Berkeley: University of California Press. Grilli, E.Z. (1993) The European Community and the Developing Countries. Cambridge: Cambridge University Press. Holland, M. (2002) The European Union and the Third World. Houndmills: Palgrave. Lister, M. (1997) The European Union and the South. London: Routledge. Lister, M. (1998) ‘Europe’s New Development Policy’, in M. Lister (ed.) European Union Development Policy. London: Macmillan. Meunier, S. (2005) Trading Voices. The European Union in International Commercial Negotiations. Princeton and Oxford: Princeton University Press. Putnam, R. (1988) ‘Diplomacy and Domestic Politics: The Logic of Two-Level Games’, International Organization, 42: 427–60. Ravenhill, J. (1992) ‘When Weakness is Strength’, in I.W. Zartman (ed.) Europe and Africa. The New Phase. Boulder, Co.: Lynne Rienner.

Ole Elgström 149 Schimmelfennig, F. (2003) The EU, NATO and the Integration of Europe. Rules and Rhetoric. Cambridge: Cambridge University Press. Tallberg, J. (2004) ‘The Power of the Presidency: Brokerage, Efficiency and Distribution in EU Negotiations’, Journal of Common Market Studies, 42(5): 999–1022. Young, A.R. (2007) ‘Trade Politics Ain’t What It Used to Be: The European Union in the Doha Round’, Journal of Common Market Studies, 45(4): 789–811.

Official documents and other sources ACP (2005) ACP Council of Ministers Declaration of the 81st Session, held in Brussels 21–22 June. Commission (2002a) Draft Recommendation for a COUNCIL DECISION authorizing the Commission to negotiate Economic Partnership Agreements with the ACP countries and regions, Brussels, 9 April. Commission (2002b) EU Mandate for Trade Negotiations with African, Caribbean and Pacific Countries, Brussels, 17 June. Commission (2005) Note for the Attention of Delegations in ACP Countries. Subject: Recent UK Statements on EPAs, Brussels, 11 April. Council (2002) EU Council of Ministers Directives for the Negotiation of Economic Partnership Agreements with ACP Countries and Regions. DTI and DFID (2005) Economic Partnership Agreements: Making EPAs Deliver for Development, Department of Trade and Department for International Development, 22 March; European Research Office (2002) The ACP Guidelines and the EU Negotiating Mandate: A Comparison, September 2002. Financial Times (2006) ‘UK Urges EU to Ease Trade Laws for Poor Nations’, by Alan Beattie, 15 October. Mandelson, P. (2005) Statement to the Development Committee of the European Parliament by Trade Commissioner Peter Mandelson, Brussels, 17 March. Mandelson, P. (2007a) ‘Economic Partnership Agreements: Tackling the Myths’, in l’Express Dimanche, 10 June, php?news_id=84960. Mandelson, P. (2007b) Comments at the INTA Committee, European Parliament, 22 October, pdf. The Guardian (1995) ‘Blaming Brussels’, by Jeffery, S., 19 May.

9 EU Policy on Global Climate Change: The Negotiation of Burden-Sharing John Vogler

The contrast between the inadequacies of the Common Foreign and Security Policy (CFSP) and the achievements of the Union in successfully promoting the ratification of the Kyoto Protocol, in the face of outright US opposition, are striking. The Union provided the ambition that drove the search for an agreement based on targets and timetables under the 1995 Berlin mandate, implemented the world’s first international emissions trading system in response to its Kyoto obligations and performed the complex diplomacy that led to eventual entry into force of the Protocol on 16 February 2005. Since then, as scientific evidence of the onset of dangerous climate change has become ever more insistent, international discussions have turned to building a regime to supplement or replace Kyoto on the expiry of its commitment period in 2012. ‘Twin track’ negotiations on the future of the UN Framework Convention (UNFCCC) and its Kyoto Protocol were initiated at the 2005 Montreal Conference of the Parties (CoP 11/MoP1) and developed in the 2007 ‘Bali Roadmap’ (CoP 13/MoP3). Just as the 1995 ‘Berlin Mandate’ agreed at UNFCCC CoP 1 set a deadline for agreement at Kyoto in 1997 so the Bali Roadmap set the goal of new agreement at the CoP to be held in Copenhagen in late 2009. This meeting, attended by heads of government, failed to produce the kind of binding and comprehensive new climate agreement that had been sought by the EU. In what was widely regarded as a failure of EU leadership, the outcome was a nonbinding ‘Accord’ which did little more than set the direction for future negotiations. The focus of this article is upon two periods in this long process of regime creation. The first involves the EU’s response to the Berlin Mandate between 1995 and 1997 and the second looks at the search for a 150

John Vogler 151

post-2012 regime from 2005. The UNFCCC architecture is extraordinarily complex and Parties, like the European Community (Union) and its member states, are involved in a bewildering range of technical and political issues, but the analysis below is restricted to one essential aspect – targets and timetables for greenhouse gas (ghg) emission reductions.1 Not only is this the headline issue in climate discussions, but it is one of the matters that most sharply differentiates the EU from the United States. For the EU, leadership in the emerging climate regime has meant public commitment to emissions reduction targets in advance of international negotiations – 15% before Kyoto and latterly the 20/30% position for the post-2012 discussions (against a 1990 baseline in both cases). At Kyoto an actual aggregate reduction of 5.2% for developed Parties was achieved, the EU contribution being 8%. In advance of the Copenhagen CoP of 2009 the EU had promised a 20% reduction by 2020 and a 30% reduction contingent on similar reductions by other developed countries. No such result was forthcoming, in circumstances of economic recession that made the EU 20% target appear less than ambitious.2 The EU’s internal negotiation problem was that these declaratory targets, to be credible, required an allocation of burdens (or ‘effort’) amongst the member states which had potentially serious national implications for growth rates and competitiveness. Thus, the negotiation of climate policy provides a particularly stern test for a theory EU foreign policy-making that claims to demonstrate that institutional factors and normative entrapment can over-ride or modify hard bargaining based upon national economic interests. Counteracting the pursuit of shortterm energy and economic gains is not only a sense of genuine alarm over the growing evidence on the likelihood and impact of climate change, but also increasing awareness of the significance of climate leadership as a critical aspect of the EU’s international identity as an actor. The idea that the EU has a unique leadership role in global climate politics is assiduously fostered by the Commission, the Council and leading member states. It represents a key component of the Union’s emerging identity (Bretherton and Vogler, 2006, ch.2). Climate leadership involves normative and policy commitments that have become very significant ‘for the functionality and credibility of the EU as a global actor’ (Thomas, ch.2, this volume).

Climate policy-making Although observers have referred to ‘Herculean problems of coordination’, especially during CoPs and notably in 2000’ (Grubb and

152 EU Policy on Global Climate Change

Yamin, 2001), the EU’s complex internal arrangements have not prevented it from being the fulcrum around which international climate negotiations have moved. In common with other areas of external environmental policy, competence is shared between the Community (Union) and member states when dealing with international climate issues. However, the greater part of competence for the issues under discussion (such as energy policy or taxation) lies with the member states, which places a heavy burden on the rotating presidency. The Commission is also heavily involved and the EU is said to negotiate ‘at 28’ – both the member states and the Community, now the Union, (as an REIO) being signatories of the UNFCCC and Kyoto Protocol.3 Climate negotiations are conducted according to a mandate drawn up by the Environment formation of the Council of Ministers. Before it meets there will be technical-level discussions led by the presidency. Here an important continuing role has been played by the relevant Council working group, in 1997 the Ad Hoc Group and now the WPIEI/CC. It comprises representatives from DG Environment of the Commission and climate specialists from member state environment ministries who meet once or twice a month in Brussels and also attend UNFCCC CoPs and other meetings. As well as supporting their national positions members of the WPIEI/CC have also developed a strong collective commitment to the EU’s international climate leadership role (Costa, 2008: 536). The group drafts climate change conclusions for the Council which are passed upwards through COREPER. Finally, as the salience of climate issues have risen, the European Council has been increasingly active in the endorsement and on occasion the negotiation of international policy and the details of the internal arrangements that support it. The tight coupling of internal and external climate policy has been increased by the introduction of the Emissions Trading Scheme (ETS) along with the Linking Directive (for associating the ETS with the CDM and other Kyoto mechanisms) The ETS directive (2003/87/EC) plus the elements of the 2008 energy and climate package are all subject to the rigours of co-decision with the European Parliament and Council (Commission, 2008a, b). Thus the complexity and character of EU climate decision-making has changed between 1997 and 2008 and is likely to change further. The 1997 and 1998 decisions were taken in the Council of Ministers and while this remained the case for target setting in 2007, the parallel and highly significant allocative decisions under Phase II of the ETS relied upon centralized and legally enforced review by the Commission.

John Vogler 153

Member state preferences As signatories to the 1992 UNFCCC the European Community and its member states accepted a non-binding objective (under UNFCCC Art.4.2), already agreed by the Council in 1990, to return their ghg emissions to 1990 levels by the year 2000. The Commission, went on to calculate that a ‘technical potential’ existed for a reduction of CO2 emissions of the order of 10% below 1990 levels by 2010 at ‘no or low costs provided positive synergies between climate change and several other policy areas of the Community are exploited’ (Commission, 1996: v–vi). However, the ambitions and capabilities of member states displayed and continue to display some extensive disparities. The figures in Table 9.1, representing the situation in 2005–7, illustrate the wide variations in economic scale, level of development and ghg emissions of the 27 member states. Some, like France are heavily reliant upon nuclear, others upon coal or gas. Equally, the environmental orientation of member states, their willingness and ability to shoulder the burdens of ghg mitigation has varied between North and South and now between West and East. These differences constitute the basic structure of inter-related external climate policy and ‘burden sharing’ negotiations within the Council. The inseparability of external and internal policy in relation to climate cannot be over-emphasized. Establishing common positions in advance of Kyoto and more recently in respect of a post-2012 regime may have far-reaching implications for the performance of member state economies and, indeed, for the political fortunes of their governments. Prior to the 1997 Burden Sharing Agreement (BSA), it would be possible to refer to three distinct groups of member states. First the ‘advanced’ green states which had been in the vanguard of environmental policymaking in Europe, notably Germany, the Netherlands Austria and the Scandinavians (Andersen and Liefferink, 1997), only Germany and to a lesser extent the Netherlands, being responsible for substantial portions of the EU emissions total. Germany under both the CDU government and the Red-Green coalition of 1998–2005 proved to be an early convert to the necessity for extensive action to counteract climate change and it was environment minister Angela Merkel who, at the UNFCCC CoP I in Berlin in 1995, reiterated a national commitment (first made in 1990) to a 25–30% reduction in ghg emissions by 2005 against 1987 levels (Commission, 1996: 44). Its achievement appeared feasible because of the ‘one off’ benefits associated with the re-unification of Germany and the opportunity that it provided to count the emissions savings

154 EU Policy on Global Climate Change Table 9.1

EU member state economies and emissions

Member State


GDP per cap $

2005 Emissions

Germany United Kingdom France Italy Spain Poland Netherlands Belgium Austria Sweden Greece Portugal Finland Czech Republic Denmark Romania Hungary Ireland Slovakia Bulgaria Slovenia Lithuania Luxemburg Latvia Estonia Cyprus Malta

19.5 14.5 14.4 12.9 8.7 4.0 4.0 2.6 2.2 2.1 2.0 1.6 1.5 1.5 1.5 1.5 1.4 1.3 0.7 0.6 0.4 0.3 0.2 0.2 0.2 0.1 0.1

32,684 32,949 31,377 30,383 28,810 14,609 33,079 33,908 36,198 32,548 24,733 20,673 34,162 20,539 37,399 9,100 18,992 45,135 18,705 10.844 24,459 16,756 76,025 15,061 19,243 23,419 21,081

1,001 657 553 582 440 399 212 143 93 67 139 86 69 146 63 53 81 70 49 70 20 22 13 11 21 10 03

BSA Target –21% –12.5% 0% –6.5% 15% (–6%) –6% –7.5% –13% 4% 25% 27% 0% (–8%) –21% (–8%) (–6%) 13% (–8%) (–8%) (–8%) (–8%) –28% (–8%) (–8%) – –

Source for emissions data: 2007 EC annual greenhouse gas inventory submission to the UNFCCC secretariat. The rounded figures are for million tonnes of CO2 equivalent, excluding LULUCF. Note that bracketed numbers in the BSA column are for new members who are not within the current BSA but do have Kyoto targets as Economies in Transition, with the exception of Cyprus and Malta. The GDP data is for 2006 and derived from Eurostat 2008.

associated with the closure and re-structuring of inefficient ex GDR plant as national reductions against a 1990 baseline.4 Germany remained the foremost proponent of extensive climate action by the Union – proposing 40% reductions by 2020 at the Heiligendamm G8 meeting in 2007. ‘No other country has been as important to establishing the EU burden sharing goal as Germany’ (Schreurs and Tiberghien, 2007: 6). However more recent German policy toward the 2008 energy and climate package might prompt some revision of this evaluation.

John Vogler 155

Another important ‘green leader’ state was the Netherlands. The Dutch occupied the EU presidency at a critical juncture in 1997 and had already established themselves as proactive developers of climate policy (Anderson and Mol, 2002). As a state with a substantial part of its territory at or below sea level, an early concern with avoiding the predicted consequences of the enhanced greenhouse effect was understandable. In 1989 Dutch policy-makers established the requirement for a stabilization of ghg emissions at 1990 levels by 2000 and then a 3–5% reduction of CO2 (Commission, 1996: 51). It is, however, worth noting that being largely dependent upon coal for power generation and having abandoned nuclear development, the Netherlands was in a much more difficult position in terms of actually implementing ambitious reduction targets than some other member states. Austria and Denmark planned a 20% reduction of CO2 emissions by 2005 against 1988 levels while Finland hoped to halt the growth of its energy-related emissions by the end of the 1990s and Sweden proposed stabilization at 1990 levels by 2000 (Commission, 1996: 37, 40, 41, 55). A second group of countries would include the UK, France, Belgium, Luxemburg and Italy, relatively rich and developed members of the Union with a substantial legacy of historic emissions to their name. Their enthusiasm for environmental policy initiatives was mixed and there was no immediate desire to go beyond the stabilization of emissions. The United Kingdom was able to make substantial reductions against a 1990 baseline at little or no cost, because of the destruction of its deep coal mining industry and the consequent ‘dash for gas’ in power generation. In 1995 its national commitment was to stabilization of emissions at 1990 levels by 2000 (Commission, 1996: 57). The arrival of the Blair administration meant that the opportunity to play a leading role in climate politics at Kyoto and beyond began to be seized. New Labour had campaigned in 1997 on a manifesto commitment to 20% CO2 reductions by 2010. Despite various internal shortcomings in relation to this target, Blair was to emphasize climate change, particularly in his strategy for the UK G8 presidency in 2005. Thus after 1997 the UK was to move from the position of climate change laggard to something of a leader, particularly in the controversial struggle to re-engage the United States. Blair, for example, personally intervened with President Bush to persuade the US to alter its blocking stance during the 2005 Montreal CoP (Interview, Council Secretariat, 2006). In the mid-1990s Belgium stood out as the only member state not to have ratified the UNFCCC and it, along with France and Italy, only had a national commitment to reduction to 1990 levels by 2000. France, with a

156 EU Policy on Global Climate Change

78% reliance upon nuclear for its power generation played ‘a very limited role’ in climate politics up until 2005 (Schreurs and Tiberghien, 2007: 39).5 Italy, faced with fast increasing emissions and under the leadership of Berlusconi, was to become a major opponent of extending the Kyoto commitments within the EU, playing what one official described as an ‘obstructive role’ in the Council (Interview, Council Secretariat, 2006). A third group comprised the ‘cohesion’ countries (defined as those with a GDP per cap. below 90% of the EU average and eligible thereby for special financial assistance), characterized by low per capita emissions, a limited contribution to the overall EU emissions total but the expectation that this contribution would increase very rapidly as economic growth took hold over the next decade or more. For them it was a question of the levels at which emissions increases could be held at the year 2000. The figure proposed by Portugal, with the lowest GDP per capita of the entire Union, was 40%, for Spain 24%, Ireland 20%, and Greece 15% (Commission, 1996: 46, 47, 52, 54).6 The 2004 accessions exacerbated this situation with a group of largely coal dependent central and eastern European countries lagging far behind the levels of development enjoyed by their western counterparts.

The approach to Kyoto The Union had committed itself to the Berlin Mandate in 1995 and the Commission was involved, alongside the member states in the Ad Hoc Group on the Berlin Mandate (AGBM) talks preparatory to the projected final agreement on a Protocol at Kyoto in December 1997. The mandate itself was relatively flexible in order to include the US. The parties agreed to initiate a process leading to ‘a protocol or other legal instrument’. There was no mention of 1990 as a base year and instead of ‘targets and timetables’ the term ‘quantified emission limitation and reduction objectives within specified time-frames’ (QUELROs) was deployed. During CoP 1 various member states had urged actual reductions from 1990 levels (ENB, 1995: 2, 12, 19) but there was no agreed EU position on an overall target for the Annex I developed countries (an outcome of CoP I, strongly supported by the EU was that nonAnnex I developing countries would not have to undertake mandatory emissions reduction commitments). EU support for QUELROs continued in the series of regular meetings of the AGBM that were designed to pave the way to agreement on a protocol at the Kyoto CoP 3. At the AGBMs the EU tabled a draft protocol and

John Vogler 157

placed heavy, but unsuccessful emphasis upon the need for agreed mandatory ‘policies and measures’ to reduce ghg emissions. At AGBM 3 in March 1996 Germany proposed reduction targets of 10% by 2005 and 15–20% by 2010 (against a 1990 baseline) but this was not an agreed EU position (ENB, 1996: 12, 27). There was clearly both a requirement and a major opportunity for EU leadership, something that was being stressed by the Commission’s DG XI (Environment) in internal discussions (Ringius, 1997). It was at AGBM 6 held between 3–7 March 1997 that the Dutch Presidency was able to make a decisive EU commitment to a 15% emissions reduction target by 2010 for the Annex I countries (ENB, 1997: 10, 12, 45). This served to shape the negotiations on QUELROs in the few months remaining until the Kyoto CoP. Nailing the EU’s colours to this ambitious negotiating target required internal agreement on the collective means of delivering such reductions. This was finally achieved at a Council meeting on the same day that the AGBM meeting convened, 3 March 1997. The EU entered into the Kyoto negotiations on the basis of Council Conclusions of 3 March and 19 June 1997 to the effect that industrialized parties to the Protocol should ‘individually or jointly, reduce their emission levels of greenhouse gases by at least 7.5% in 2005 an by 15% in 2010 as compared with 1990 values. The objective would be achieved by the Community as a whole (the ‘EU bubble’) by means of burden sharing which would allow certain member states not to change and even to increase their emissions while others would undertake to reduce them beyond the overall target’ (Council, 1997).

Internal negotiation of the Kyoto mandate and burden sharing While it had been clear throughout the early 1990s that some form of burden-sharing would be necessary, if the EU was to move beyond its agreement of October 1990 to stabilize CO2 and exercise leadership in the global negotiations, ‘no substantial progress had been achieved until 1997’ (Ringius, 1997: 17).7 Developments or rather the lack of developments in the AGBM and the imminence of the Kyoto CoP made the matter urgent. The Commission argued strongly that in order to attain a global leadership role the EU had to propose a politically credible target and that this must be no less than a 10% reduction by 2005 (from a 1990 baseline) (ibid, 1997: 19). The coincidence of the next AGBM meeting and EU Spring Environment Council in March 1997 set a clear deadline.

158 EU Policy on Global Climate Change

The Netherlands assumed the presidency on 1 January 1997 in the aftermath of a December 1996 Environment Council that had failed to agree on a EU position on specific ‘targets and timetables’. Dutch officials proceeded to implement consultations at a technical level to resolve the issue of finding a common position on emissions reductions that would equitably accommodate the divergent national energy interests of member states. By all accounts the strategy had been long prepared and involved the ingenious solution of disaggregating national emissions sources in what was known as the ‘Triptique’. The idea was to overcome entrenched national resistance centred on single emission targets by dividing all emissions up into three sectors 1) the light domestic 2) energy intensive export-oriented and 3) power generation; and treating them separately for the purposes of re-calculating national emissions allowances. The ‘Triptique’ was unveiled to national representatives on the Ad Hoc Group on Climate Change (see fn.2) at a workshop held at Zeist on 16–17 January 1997. There, four potential solutions arising from the Triptique analysis were discussed in advance of a political decision by the upcoming Environment Council meeting (Ringius, 1997: 24–6). The Dutch presidency then proposed an overall EU target of 10% by 2005 and 15% by 2010 to serve as a negotiating position at the AGBM and Kyoto plus a set of national emissions reduction targets, by which it might be achieved, derived from the Zeist workshop and the Triptique. The latter went significantly further than informal pledges made by some member states (which would have resulted in an overall EU reduction of 11%. For Germany, Belgium, Greece, Ireland, and Finland 5% increases in targets were proposed. For Luxembourg and the UK, 10% increases. For all the others, and most significantly the Netherlands, itself, no increases were proposed. These differences formed the substance of the Ad Hoc Group on Climate on 17–18 February at which various member states held the increases to be unacceptable. The Environment Council met under the pressure of the immediacy of the AGBM meeting and the knowledge that anything less than a stated position of a 10% reduction by the industrialized countries would be a disaster for EU aspirations to climate leadership. However, irritated by the apparently self-serving proposal of the Dutch to restrict their own contribution to a 5% reduction; Germany, Belgium and the UK rejected the presidency’s proposals as did Greece and Portugal. This left the meeting with a set of minimum contributions by a range of member states amounting to no more than a 10% reduction for the EU as a whole.

John Vogler 159

The day was saved by the Danish minister who came up with a pragmatic solution. This was simply to accept the contributions as an interim agreement on a 10% reduction but to go forward to the global negotiations with a target of 15%, thus allowing the EU to take the lead in the AGBM discussions. If a 15% reduction was to be negotiated at Kyoto then the EU would have to revisit the matter and find ways of achieving the further 5% reduction. If not there would not be a problem.8 In the event this proved to be an accurate estimate. The EU entered into the Kyoto negotiations with a widely admired 15% target in its negotiating mandate. What emerged from Kyoto was a 5.2% average reduction for developed countries by the end of the first commitment period 2008–12. The EU’s share in what was described as a global bubble was an 8% reduction in ghg emissions against a 1990 baseline. In 1998 the Council moved to translate this into a revised version of the BSA with significantly reduced national emissions commitments. Even so they have been hard to achieve. There is some fit between the 1997 BSA negotiations and the competitive and the cooperative bargaining hypotheses (Thomas, ch.2, this volume). The Environment Council of 3 March had many of the hallmarks of competitive bargaining as member states backed away from Table 9.2

EU burden sharing 1997–98 Presidency Proposal

BSA 1997

Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxemburg Netherlands Portugal Spain Sweden UK

–25% –15% –25% –10% –05% –30% +05% +05% –10% –40% –10% +25% +15% +05% –20%

–25% –10% –25% 0% 0% –25% +30% +15% –07% –30% –10% +40% +27% +05% –10%

–13% –7.5% –21% 0% 0% –21% +25% +13% –6.5% –28% –06% +27% +15% 0% –12.5%





Sources: Ringius, 1997, pp.7 & 32; Council of the European Union, 2002.

BSA 1998

160 EU Policy on Global Climate Change

the presidency proposals and insisted upon protecting their national interests in the face of perceived ‘non-cooperation’ by the Netherlands and other partners. Because of the potential impact upon industrial competitiveness of national mitigation measures it might also be possible to characterize this phase of the negotiations in terms of a zerosum competition. However the Triptique-based discussions served to ‘reframe’ them, establishing a formula within which distributive bargaining could occur over detail (Zartmann, 1976).9 It certainly helped to establish a ‘contract zone’ from a disparate set of national positions founded upon differing assumptions and baselines. Yet how far was this pre-Council exercise truly ‘integrative’ in character? An argument that it was, could be constructed on the basis of the way in which the export competitive sector, where differing energy costs within the Single Market might have been regarded as setting up a zero-sum competition, was abstracted. Similarly, there was the formula devised to justify the increases allowed to the cohesion countries where a slow convergence to equivalent per capita emissions across the Community by 2030 was envisaged. In respect of the cohesion countries the existence of side payments, that might serve to increase the sum of the game by persuading cohesion countries that if they were to reduce the scale of their projected emissions they would be compensated through the Community’s structural and other funding mechanisms, is disputed. Ringius (1997: 39) is certain that this was not the case and that leader countries simply relied upon exhortation rather than any attempt to make an explicit compensatory link between the Cohesion and Structural funds and abatement costs while Desai and Michaelowa (2001: 332) claim that the expectation of side payments was a factor in Portugal’s acceptance of the final BSA where a 27% increase represented a very ambitious climate change policy (ibid: 339). Taken as a whole, and in line with other environmental policy areas (Vogler, 1999), there does not appear to have been a retreat to a lowest common denominator solution or a situation in which the EU convoy was condemned to move at the speed of the slowest ship. Member states had already moved beyond a simple ‘business as usual’ or ‘stabilization at 1990 levels’ approach. A real overall reduction was accepted as the necessary basis for climate leadership. Nonetheless, a moment of ‘lowest common denominator’ behaviour was observable amongst those states that reacted against the presidency proposal. The BSA agreement was enabled by the very favourable circumstances associated with the 1990 baseline. Germany, a convinced green leader, was able to cover 85% of the required emissions reductions without excessive sacrifice. Similarly the United

John Vogler 161

Kingdom contribution was well within its own national target, set by the incoming Blair government at 20%. Indeed while other countries revised their contribution downwards in 1998, the Blair government was able to offer more with its agreement to an additional 2.5% reduction. The institutionalist might expect ‘normative entrapment’ to play some part in setting the pre-Kyoto negotiating target – member states being prepared to make greater contributions than would be predicted on the basis of strict adherence to their national energy interests. Also involved would be socialization into acceptance of the emerging self-image of the EU as global climate leader. Difficult as it is to determine such things, some supporting evidence exists. Austria, Denmark, Germany and the Netherlands had already committed themselves to climate targets from which it was difficult to retreat, ‘They shared an interest in providing prestigious environmental leadership at the EC and global level, and would undermine their political credibility as lead countries if they lowered their targets during the negotiations’ (Ringius, 1997: 38). At least part of this commitment was, however, to domestic political groups. The clearest evidence of Union membership as an incentive to moderate negotiating position comes from the cohesion countries. Spain was in 1997–8 a ‘policy-taker’ its political elite fully aware of the ground that had to be made up in implementing EC environmental policies. Costa (2006: 225) notes that ‘even the cruder defence of national interests… was expressed in a way that did not question EU international and internal policies’ and ‘the strong adhesion of Spain to a discourse that was perceived as legitimate simply because it had been endorsed by the EU’ (ibid: 228). Some evidence of Spanish desire to be part of an EU climate consensus is to be found in its willingness to accept a reduction from a 27% to a 15% increase in the final 1998 BSA agreement when other countries lightened their burdens and it was becoming clear that the target might be difficult to meet in the face of sharply rising Spanish emissions. Something similar appears to have occurred with Portugal – it too accepted a reduction in 1998 from a 40% to a 27% increase. Dessai and Michelowa (2001: 332) claim that Portuguese negotiators ‘wanted to prevent international criticism and thus accepted the proposal despite lack of feedback with interest groups at home’.

Implementing Kyoto and post-2012 planning As well as accepting a collective target of 5.2% reductions for the developed countries at Kyoto the EU, at first reluctantly, was forced to embrace

162 EU Policy on Global Climate Change

the ‘flexibility mechanisms’; Joint Implementation, the Clean Development Mechanism and above all emissions trading as means whereby Annex I Parties could achieve the emissions reductions (now relating to a six gas basket of ghgs) to which they were committed. It is one of the great ironies of recent environmental politics that, having been unable to resist emissions trading in the negotiations, the EU was soon to become its greatest international advocate.10 In 2005 it launched the world’s first international emissions trading scheme, the ETS, which bore the burden of having to deliver over 40% of the EU’s pledged Kyoto reduction commitment.11 The Kyoto Protocol was in no sense an operational agreement and it was to take years of difficult negotiations, in which the EU provided the key motivating force, to draft all the detail required to run the mechanisms and to ensure compliance. Thus, it was only at CoP 13 held at Montreal in late 2005 that final agreement was reached upon a Kyoto Protocol that could be fully implemented by its Parties. In the meantime the EU had struggled through difficult meetings at The Hague in 2000 and then CoP 6 bis at Berlin and at Marrakesh in 2001. The major, but predictable, setback to the prospects of Kyoto occurred in March 2001 when the incoming US administration of George W. Bush actually denounced the US signature (being joined by Australia which refused to ratify) and then proceeded to go over to active opposition. Undaunted, the EU took the landmark decision at the Gothenburg European Council of June 2001 to persevere with the Protocol regardless. This meant not only leading the attempt to flesh out its terms but, just as significant, marshalling EU diplomatic and economic resources to ensure entry into force. This was far from easy in the face of outright US hostility because the terms of the Protocol required ratification by 55% of signatories which must include Annex I Parties responsible for in excess of 55% of emissions. In these circumstances, persuading Japan, Canada and finally Russia to ratify ‘required a real demonstration of EU resolve as an actor’ (Vogler and Bretherton, 2006: 3). It also heightened the sense that the EU was now a global player capable of responding to demands that US hegemony be challenged. As Environment Commissioner Wallstrom observed in 2001 ‘I think something has changed today in the balance of power between the US and the EU’ (ENB, 2001: 1). The EU’s leadership role in climate politics began to take on significance well beyond the specifics of the UNFCCC and Kyoto, extending to the very identity of the Union in the international system. This chimed in with the rising public salience of climate change issues, associated not only with mounting scientific evidence but also with events such as the European summer heat-wave of 2003. Thus one might expect that ‘normative entrapment’

John Vogler 163

would be evident in the internal deliberations which were bound to follow the successful entry into force of the Protocol in February 2005. No sooner had the Protocol entered into force than it was time to consider its successor or its development. Art 3.9 stipulated that discussions on the future post-2012 commitments of developed countries be commenced in 2005 while Article 9 refers to the broader future of the Protocol and its relation to the UNFCCC. This set a timetable for deliberation leading up to CoP 13/MoP I scheduled to begin at the end of November 2005 in Montreal. All that had been achieved in a difficult CoP 12 at Buenos Aires in the previous year was an agreement to hold a seminar on post-2012 in Bonn in May 2005. The EU’s position agreed at the March 2005 European Council reiterated the need to stabilize CO2 emissions at below 550 ppmv if dangerous climate change was to be avoided. ‘“Global mean surface temperature increase should not exceed 2°C above pre-industrial levels” requiring “significantly enhanced aggregate (greenhouse gas) reduction efforts by all economically advanced countries” with reduction pathways of the order of 15–30% by 2020’ (European Council, 2005: 46). The preceding Environment Council had also stipulated a 60–80% reduction by 2050 but this was too much for some member states and by the European Council meeting this target had disappeared entirely. The main problem with setting targets appeared to lie with Italy, which was obstructive during the Luxembourg presidency, wanting to propose changes on economic rather than environmental grounds and evidently concerned with its own rising emissions (Interview, Council Secretariat, 2005). These problems with setting a target for post-2012 continued through to the March 2006 Environment Council (ENDS, 2006). Instead of hitting its Kyoto target, Italy increased emissions by 12%. In contrast, Sweden proposed to reduce its emissions by 25% from its 1990 baseline by 2020 (ENDS, 2006a), while Spanish ghg emissions hit a new high of 53% above the 1990 level in 2005 (ENDS, 2006b). The credibility of the Union’s position was still critically dependent upon the fortunes of its ETS, of which phase I entered into force at the beginning of 2005. This proved to be unsuccessful in terms of establishing a respectable carbon price – in May 2006 the price collapsed and at the end of 2007 it was as low as €0.02 cents (it had been as high as €30 per tonne) (Stern, 2008: 22). Most member states had protected their national interests by over-estimating their emissions in their National Allocation Plans. All of the work by the Union to ensure that Kyoto was developed into a form that would provide the basis for a working climate regime could be put at risk if the Union were to fail to

164 EU Policy on Global Climate Change

deliver its 8% reduction through the operation of the BSA which now, after the 2004 enlargement, ran alongside the separate Kyoto commitments of the East European accession states. The 2004–5 trends indicated in the EU’s greenhouse gas inventory report to the UNFCCC Secretariat were not auspicious (EEA, 2007).12 Between 2004 and 2005 total EU 15 emissions decreased by 0.8% or 35.2 million tonnes. The situation for the entire EU 27 appears more favourable because of the large reductions obtained by ex-Comecon economies since 1990. Thus the equivalent EU 27 decrease is 7.9% (ibid: 7). The accession states are treated differently under Kyoto because they are classed as ‘economies in transition’, but as with the older cohesion countries, they remain wary of disabling their economic growth through agreeing to new emissions restrictions. To repeat the ETS I experience in its next phase 2008–12 would have done serious damage to the Union’s aspirations to future climate leadership and more specifically to the prospect of fulfilling the 8% Kyoto target. Thus the Commission embarked on a much more rigorous review of the National Allocation Plans, lowering many of the proposed totals and taking infringement procedures against errant member states at the European Court of Justice (ECJ). In contrast to ETS Phase I the Commission rejected and demanded cuts in the proposed allocations of all but three of the 23 NAPs submitted by mid-2007 (see Table 9.3). This was far from the negotiation model employed to determine the content of the EU’s bubble in 1997. Instead the Commission made a determination of member state obligations in terms of 12 criteria which included, inter alia, the national ‘gap to close’ in relation to individual BSA targets, consistency with the terms of Kyoto and fairness in relation to the overall EU effort. Assessment used ‘independently verified’ and ‘high quality’ emissions data (Commission, 2006). In moves reminiscent of the 1997 burden sharing debates most of the states that had acceded in 2004 demanded increases in their allowed emissions on the grounds that they failed to take into account the economic growth that would be required to bring their GDP per capita up to average EU levels. Six of them initiated legal action against the Commission on these grounds in 2007 and Vaclav Klaus, the Czech President, was quoted as saying that Communism had been replaced by environmentalism as ‘the biggest threat to freedom, democracy, the market economy and prosperity at the beginning of the 21st century’ (Cantor and Castle, 2007). Alongside these developments the EU continued its international activities in search of a post-2012 arrangement. In 2005, the year of the

John Vogler 165 Table 9.3

ETS emissions and caps

Member State


2005 Emissions

Austria Belgium Cyprus Czech Rep. Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Poland Slovakia Slovenia Spain Sweden UK

33.0 62.10 5.7 97.6 19.0 45.5 156.5 499.0 74.4 31.3 22.3 223.1 4.6 12.3 3.4 2.9 95.3 239.1 30.5 8.8 174.4 22.9 245.3

33.4 55.58 5.1 82.5 12.62 33.1 131.3 474.0 71.3 26.0 22.4 225.5 2.9 6.6 2.6 1.98 80.35 203.1 25.2 8.7 182.9 19.3 242.4

32.8 63.30 7.12 101.9 24.38 39.6 132.8 482.0 75.5 30.7 22.6 209.0 7.7 16.6 3.95 2.96 90.4 284.6 41.3 8.3 152.7 25.2 246.2





ETS II Caps Proposed Allowed 30.7 (93.6%) 58.5 (92.4%) 5.48 (77.0%) 86.8 (85.2%) 12.72 (52.2%) 37.6 (94.9%) 132.8 (100%) 453.1 (94.0%) 69.1 (91.5%) 26.9 (87.6%) 22.3 (98.6%) 195.8 (93.7%) 3.43 (44.5%) 8.8 (53.0%) 2.5 (63.0%.) 2.1 (71.0%) 85.8 (94.9%) 208.5 (73.3%) 30.9 (74.8%) 8.3 (100.%) 152.3 (99.7%) 22.8 (90.5%) 246.2 (100%) 1903.43


All figures are in million metric tonnes of CO2. Source: European Commission (2007). As of August 2007 four NAPs remained unapproved, Bulgaria, Denmark, Portugal and Romania.

UK joint presidency of the EU and G8, the Blair government pursued a well-orchestrated strategy to engage the US and major developing countries in the discussion of post-2012 without sacrificing the EU’s growing commitment to Kyoto and emissions trading. The Union’s approach to the CoP13/MoP 1 in Montreal was very different from its proactive position before Kyoto. It was described as an ‘open and reliant (upon the Canadian Presidency of the CoP) stance in which the October 2005 Environment Council drafted a mandate which was highly flexible in anticipation of the initiation of a process involving not only the Kyoto Parties but the US, Australians and developing

166 EU Policy on Global Climate Change

countries as well (Interview, Council Secretariat, 2006). In the event, the initiation of such a process was achieved at Montreal but not without some internal disagreement with Belgian, Hungarian and Dutch objections over the lack of direct linkage between protocol and convention tracks and some resentment amongst smaller member states at being kept ill informed as to the British presidency’s strategy (ibid). Thus some progress had been achieved but the EU had failed to arrive at the kind of headline reduction target that had underpinned its leadership in advance of Kyoto. During 2006 the situation in the Council was greatly eased by the removal of the Berlusconi government in Italy that had openly urged the discontinuation of Kyoto after 2012 because of US opposition. Now the new Italian government was to dismiss this position as an ‘anomaly’ (ENDS, 2006c). Nonetheless the next CoP (14/MoP 2), held at Nairobi at the end of 2006 saw no movement on targets and timetables. Understandably, in view of its location, its main agenda items related to developing country issues such as the regional distribution of CDM projects. Subsequently, at the December 2006 Environment Council, the Finnish presidency raised for the first time the issue of a post-2012 target for the EU. UK, Germany, Italy and Sweden, supported by Environment Commissioner Dimas, favoured a 30% reduction by 2020. Hungary, Slovakia Poland Spain and others opposed, arguing that the EU should wait for other Kyoto parties before making ‘a hasty declaration of commitment’. Commissioners Verheugen (Enterprise) and McCreevy (Internal Market) were also reported to be against EU unilateral adoption of any target greater than 15% (ENDS, 2006d). On the other side Energy Commissioner Piebalgs reportedly backed Commissioner Dimas in arguing that a pro-active climate policy would have economic advantages for new European green industries (Cicerone, 2006). The latter position seems to have been accepted within the College of Commissioners because on 10 January 2007, the Commission adopted an energy and climate package that was subsequently accepted by the Environment and European Councils of March (Commission, 2008). The compromise endorsed at the March 2007 Council stated that a 30% reduction by the EU against a 1990 baseline would be the EU’s contribution ‘… provided that other developed countries commit themselves to comparable emissions reductions and economically more advanced developing countries to contributing adequately according to their responsibilities and respective capabilities (European Council, 2007: 31). Otherwise until a ‘global and comprehensive agreement for the period

John Vogler 167

beyond 2012 was achieved and without prejudice to its position in international negotiations’ the EU’s firm independent commitment was to achieve at least a 20% reduction in ghg emissions. The long-term collective target of 60–80% reductions by 2050 by developed countries was also restored and the Council prefaced its conclusions by underlining ‘the leading role of the EU in international climate protection’.

The 2008 energy and climate package The European Council tasked the Commission with finding equitable and cost efficient ways of sharing the burdens and effort required to attain the 20-20-20 target that it had set (European Council, 2007: 33). The resulting ‘package’ of draft decisions and directives was published in January 2008. Under phase three of the ETS the entire tradable emissions sector was to be removed from the control of member states – avoiding previous haggling over National Allocation Plans. Instead the Commission would allocate permits and set an EU wide cap involving a 21% reduction of total ETS emissions by 2020 against a 2005 baseline (this entailed an increase in the ETS share of the overall contribution to the EU target of 60% as against the current figure of 41%). From 2012, member states would mainly auction permits to users within their jurisdiction and, in a division reminiscent of the Triptique approach there will be special treatment for sectors exposed to international competition and carbon leakage (Commission, 2008a). Remaining non-tradable emissions (agriculture, transport, building etc) were to be subject to ‘effort sharing’ by member states with an aggregate 10% reduction achieved against a 2005 baseline. Once again there were echoes of the 1997 debates in that the Commission proposed an equitable method for apportioning effort which involved division according to GDP per capita (Commission, 2008b).13 There were also proposals on carbon capture and sequestration and on vehicle emission standards. Already subject to intensive member state and industrial lobbying during its formulation, the package wound its way through the codecision process until final bargains were struck at a tense European Council on 11–12 December 2008. In the context of economic recession member states, including Germany and France expressed serious concern that the current policy, in the absence of an international agreement on carbon leakage, would fatally damage the competitiveness of steel, cement and aluminium producers and force the flight of these industries to China and elsewhere (Economist, 2008, 22 March: 9). The new Berlusconi government in Italy threatened a veto and East

168 EU Policy on Global Climate Change

Europeans, led by Poland (also threatening a veto) denounced the unfairness of the auction of ETS permits, demanding a free allocation to their overwhelmingly coal-fired power plants (European Voice, 2008, 12 December). The French presidency proved capable of orchestrating a series of bargains and compromises before and during the December European Council to the effect that agreement was achieved on a substantially modified package, which left all parties other than the environmental lobby expressing satisfaction that their aims and interests had been satisfied. Substantial derogations from the auctioning of ETS permits were agreed to meet the demands of power generators and Central and East European objections were addressed through a solidarity fund re-distributing 10% of auction allowances from high to low GDP per capita countries with an additional 2% mainly re-distributed to Poland, Romania and Bulgaria.14 To address the carbon leakage problem a complex methodology was agreed to determine the exemption of ‘at risk’ industries subject to international competition in the event that global rules were not forthcoming. Controversially in both the ETS and ‘effortsharing’ policies, the permitted offsets from buying in JI and CDM credits were also raised up to the level of 50% of EU wide reductions 2008–20 (Commission, 2008c: 8). When the full details of these important deals, taken in the shadow of the Copenhagen CoP, are subject to future analysis it is more than probable that they will reveal a hardbargaining process to accommodate national economic interests. However, it remains plausible that member states felt compelled to reach an agreement that the Commission could claim would support the climate policy targets and sustain EU leadership. For as a commentary in early 2008 predicted, ‘no one member state or constituency would risk standing in the way of addressing such an important issue as climate change’ and the Commission’s proposal are equitable. ‘We should therefore expect this package in large part to be implemented’ (Institute of International and European Affairs, 2008: 9–10).

Conclusions EU climate change policy certainly fulfils the conditions appropriate to the operation of normative entrapment – an uncontested normative frame (involving the necessity for action by the EU) significant public attention, plus external conditions that have strengthened the relevance of the EU’s original bid for international climate leadership (Thomas, ch.2, this volume). There has been a growing commitment to EU leader-

John Vogler 169

ship and willingness in both pre-Kyoto and post-2012 diplomacy to run ahead of other developed countries by committing to emissions reductions targets and timetables. Also, there can be few areas of the EU’s external relations that have such a compelling normative content and engage such widespread parliamentary and public support. Yet climate policy also has the potential to impose very significant costs on member state economies and particular sectors which are heavy energy users – raising the likelihood of hard interest-based bargaining. In 1997 the pre-existing EU climate commitment certainly played a part and the strongest evidence for normative entrapment appears in the willingness of the cohesion countries to accept cuts in their projected emissions, which did in fact prove difficult to achieve. The negotiations involved some adroit integrative re-framing followed by some hard bargaining in the Council. However, the context for reaching and sticking to an agreement was highly favourable. Two member states were able to deliver most of the planned EU commitment with little trouble against a highly favourable baseline and the final Danish ploy of simply setting aside 5% of the target reduction pending the international negotiation was never tested. Instead an EU target of 8% rather than the 10% allowed for in the original BSA allowed for some relaxation. The post-2012 case has some similarities with its predecessor in that the EU again attempted, with some difficulty, to devise a target and timetable commitment. On this occasion it was blocked in the Council until the change of Italian government in 2006. The emergence of the ETS and the new climate and energy package mean that the bargaining that underpins external climate diplomacy now differs from that conducted by the member states in 1997. Unlike then, the Commission has moved into a central allocative role under phase II of the ETS and its proposed successor and there is the additional complication of lengthy co-decision procedures. The stakes for member states were much higher than during the approach to Kyoto for the costs of ETS auctioning were very real, especially for aggrieved central and east European governments already in serious recession and reliant upon coal-fired power generation. But the normative pull of the EU’s commitment to climate leadership has also strengthened since 2001 alongside the salience of climate policy. This has been underlined by an increasing perception of the ways in which measures to reduce emissions can also, potentially, enhance energy security and reduce the kind of vulnerability evident in the Russia-Ukraine gas supply crises of 2006 and 2009 (Commission, 2006a). Hard interest-based bargaining and threats of Council vetoes were evident but so, it may be surmised, was a degree of normative entrapment as the EU struggles with the complex

170 EU Policy on Global Climate Change

inter-locking of internal and external climate policy. In the end the failure to achieve negotiating targets at Copenhagen was clearly not a consequence of internal EU dissension, although the array of European leaders gathered at the final stage of the Conference created some confusion. Rather, the outcome appears to have reflected some deeper changes in the structure of international climate politics – the reengagement of the US and the new significance of the BASIC countries, Brazil, South Africa, India, China – over which the Union could have little control.

Notes 1 Note that from 1st December 2009 under the Lisbon Treaty the Union has replaced the Community. 2 The Copenhagen Accord simply lists in an appendix (to which Parties made submissions in January 2010) an uncoordinated set of ‘quantified economywide emissions targets for 2020’ representing the various pledges made in advance of Copenhagen. ‘As part of a global and comprehensive agreement for the period beyond 2012, the EU reiterates its conditional offer to move to a 30% reduction by 2020 compared to 1990 levels, provided that other developed countries commit themselves to comparable emission reductions and that developing countries contribute adequately according to their responsibilities and respective capabilities.’ Appendix I to the Copenhagen Accord, UNFCCC Decision -/CP.15. 3 In the period under discussion in this chapter the relevant part of the Commission was DG Environment, previously DG XI. From the beginning of 2010 a new DG Climate Action has been given the main responsibility in this area. The Community and now the Union has the legal right to participate in the UNFCCC under a clause that recognizes Regional Economic Integration Organizations (REIOs) of which the Union is the only extant example. 4 The 25–30% reduction target was subsequently reduced to 21% at the outset of negotiations on the BSA. Actual German emissions fell very sharply in the first half of the 1990s but subsequently the rate of decrease levelled off. 5 Green politics was also relatively under-developed in France. Some change occurred in the period 2005–07 when Chirac prioritized environmental policy and led the call, which became EU policy, for the creation of a UN environmental organization. 6 These different national commitments and baselines were further complicated by subsidiary targets, put forward by some, but not all, member states for other greenhouse gases, nitrous oxide and methane. These three gases were those mentioned in the original UNFCCC (the CFCs – also potent ghgs being controlled under the Montreal Protocol). Part of the Kyoto deal was to further elaborate and complicate the regime and the operation of the mechanisms by adding three additional ‘industrial’ greenhouse gases to form a ‘six gas basket’.

John Vogler 171 7 Under the Irish presidency a workshop including the EC Ad Hoc Group on Climate, Commission officials, Dutch energy specialists and a professional facilitator met in Dublin in September 2006 and various negotiating targets were discussed without agreement (Ringius, 1997: 18–20). 8 Analysts of the EU’s climate change are indebted to the extensive research and interviewing of participants carried out by Lasse Ringius. He provides a detailed re-construction of the ad hoc group and Council meetings upon which the brief description above is based (Ringius, 1997: 15–33). 9 It is worth mentioning that the originators of the concept of integrative bargaining, Walton and McKersie (1965) viewed bargaining as four interrelated social processes distributive and integrated bargaining, attitudinal structuring and the management of boundary role conflict. Integrative bargaining both contradicts and complements distributive bargaining. Walton and Mckersie refer to mixed bargaining where integrative bargaining increases the size of the pie to be distributed. There are some similarities to Zartmann’s (1976) distinction between commitment-convergence and formula-detail bargaining. 10 For an interesting analysis of this change in terms of the way in which the EU was able to escape norm-entrapment – in terms of its stated opposition to emissions trading – through US abandonment of the Protocol, is to be found in Cass (2005). Wettestad (2005) provides a very detailed account of development of the original ETS directive. 11 The Emissions Trading Scheme (ETS) introduced in January 2005 covered power generation in excess of 20 Megawatts. This permit-based ‘cap and trade’ system covered some 12,000 enterprises in the member states and relied upon their estimate of emissions under National Allocation Plans (NAPs). Allowances – or rights to emit – were mainly provided free to plants on the basis of their historic emissions (grandfathering). Phase I probably had little effect upon actual emissions but provided a learning experience for the 2008 second phase and the planned 2013 third phase. These provide an altogether more rigorous framework in terms of allocation, the auctioning of allowances and the imposition of a cap on CO2 emissions. 12 Relative to the 1990 base year EU 15 (BSA) ghg emissions were in 2005 only reduced by 2%, or 86 million tonnes of CO2 equivalent. On a linear path plotted to 2010 they were thus 4 index points above where they should have been in relation to achieving the 8% target (EEA, 2007). 13 ETS and Effort sharing are only part of the Commission’s package which also includes proposals on carbon capture and storage, energy efficiency and targets for renewables and most controversially the use of biofuels. 14 The original plan was that all allowances should be auctioned under Phase 3 of the ETS. Now an ‘optional and temporary derogation’ was introduced for certain member states such that 70% of their allowances might be free of charge in 2013 but with a view to achieving full auctioning by 2027. The rules about eligible plant are complex and will be further determined by a ‘comitology’ process (Commission, 2008c: 3–8). Additional allowances of 2% over and above the 10% in the solidarity fund are apportioned 29% to Romania, 27% to Poland and 15% to Bulgaria (Euractiv, 2008, 12 December).

172 EU Policy on Global Climate Change

Works cited Andersen, M. and Liefferink, D. (eds) (1997) European Environmental Policy: The Pioneers. Manchester: Manchester University Press. Andersen, M. and Mol, A.P.J (2002) ‘The Netherlands in the UNFCCC Process – Leadership between Ambition and Reality’, International Environmental Agreements, 2: 49–68. Bretherton, C. and Vogler, J. (2006) The European Union as a Global Actor. London: Routledge. Cantor, J. and Castle, F. (2007) ‘EU Wrangling on Carbon Emissions Moves into the Courts’, International Herald Tribune, 31 July. Cass, L. (2005) ‘Norm Entrapment and Preference Change: The Evolution of the European Union Position on International Emissions Trading’, Global Environmental Politics, 5(2), February: 1–23. Cicerone (2006) ‘The Battle Over Climate Policy in the EU, 2006–1’, http:// Commission of the European Communities (1996) Communication from the Commission under the UN Framework Convention on Climate Change, COM (96) 217 final, 11.06.1996. Commission of the European Communities (2006) Communication from the Commission to the Council and to the European Parliament on the assessment of national allocation plans for rhe allocation of emission allowances in the second period of the EU Emissions Trading Scheme, Brussels, 29.11.2006, COM (2006) 725 final, p.2. Commission of the European Communities (2006a) A European Strategy for Sustainable, Competitive and Secure Energy, Brussels, 8.3.2006, COM (2006), 105 final. Commission of the European Communities (2008) Questions and Answers on the Commission’s proposal for effort sharing, Press Release MEMO/08/34, 23 January. Commission of the European Communities (2008a) Proposal for a Directive of the European Parliament and of the Council amending Directive 2003/87/EC so as to improve and extend the greenhouse gas emission allowance trading system of the Community, COM (2008) 16 final, 23.1.2008. Commission of the European Communities (2008b) Proposal for a Decision of the European Parliament and of the Council on the effort of Member States to reduce their greenhouse gas emissions to meet the Community’s greenhouse gas emission reduction commitments up to 2020, COM (2008) 17 final, 23.1.2008. Commission of the European Communities (2008c) Questions and Answers on the revised EU Emissions Trading System, MEMO/08/796, Brussels, 17 December. Costa, O. (2007) ‘Spain as an Actor in European and International Climate Policy: From a Passive to an Active Laggard?’, South European Society and Politics, 11(2): 223–40. Costa, O. (2008) ‘Is Climate Change Changing the EU? The Second Image Reversed in Climate Politics’, Cambridge Review of International Affairs, 21(4), December. Council of Ministers (1997) 2033 Environment, 16/10/- Press: 296 Nr:11332/97. Council of the European Union (2002) Decision 2002/358/EC, 25 April. Dessai, S. and Michaelowa, A. (2001) ‘Burden Sharing and Cohesion Countries in European Climate Policy: The Portuguese Example’, Climate Policy, 1: 327–41. Economist (2008), 22 March.

John Vogler 173 Earth Negotiations Bulletin (ENB) (1995) ‘1st CoP of the UNFCCC: March 28–April 07, Berlin Germany, Summary’, 12(21). ENB (1996) ‘3rd Session of the AGBM: March 05–08, Geneva Switzerland, Summary’, 12(27). ENB (1997) ‘6th Session of the AGBM: March 03–07, Geneva Switzerland, Summary’, 12(45). ENB (2001) ‘CoP 6 bis of the UNFCCC: July 16–30, Berlin Germany, Summary’, 12(176). ENDS Daily (2006) ‘Italy Weakens Ministerial Climate Resolution’, 10 March. ENDS Daily (2006a) ‘Sweden Adopts 2020 Climate Change Target’, 31 March. ENDS Daily (2006b) ‘Spain’s Greenhouse Emissions Hit New High’, 24 April. ENDS Daily (2006c) ‘Italy Returns to the Fold on Post-Kyoto Rules’, 8 November. ENDS Daily (2006d) ‘Council Debates Post-2012 Climate Options’, 18 December. Euroactiv (2008) 11 December, ‘EU Heads Towards Scaled-Down Climate Ambitions’, European Commission (2007) Press Release ‘Emissions Trading: Commission Adopts Decision on Cyprus’s National Allocation Plan for 2008–2012’, IP/07/1131, 18/07/2007. European Council (2005) Presidency Conclusions – Brussels 22/23 March. European Council (2007) Presidency Conclusions – Brussels 8/9 March. European Environment Agency (2007) Annual European Community greenhouse gas inventory 1990–2005 and inventory report 2007, Submission to the UNFCCC Secretariat, EEA technical Report No.7/2006. European Voice (2008) 22 December, Eurostat (2008) Europe in Figures: Eurostat Yearbook 2008. Luxembourg: Office for Official Publications of the European Communities. Grubb, M. and Yamin, F. (2001) ‘Climate Collapse at the Hague. What Happened and Where do We Go from Here?’, International Affairs, 77(2): 261–76. Institute of International and European Affairs (2008) Energy and Climate Change Policy Brief, February. Ringius, L. (1997) Differentiation, Leaders and Fairness: Negotiating Climate Commitments in the European Community. Oslo: Cicero, Report 1997: 8. Schreurs, M.A. and Tiberghien, Y. (2007) ‘Multi-Level Reinforcement: Explaining European Union Leadership in Climate Change Mitigation’, Global Environmental Politics, 7(4), November: 19–46. Stern, N. (2008) Key Elements of a Global Deal on Climate Change. London School of Economics and Political Science. Vogler, J. (1999) ‘The European Union as an Actor in International Environmental Politics’, Environmental Politics, 8(3): 24–48. Vogler, J. and Bretherton, C. (2006) ‘The European Union as a Protagonist to the United States on Climate Change’, International Studies Perspectives, 7: 1–22. Walton, R.E. and McKersie, R.B. (1965) A Behavioral Theory of Labor Negotiations: An Analysis of a Social Interaction System. New York: McGraw Hill. Wettestad, J. (2005) ‘The Making of the 2003 EU Emissions Trading Directive: An Ultra-Quick Process due to Entrepreneurial Proficiency?’, Global Environmental Politics, 5(1), February: 1–23. Zartmann, I.W. (1976) ‘Reality Image and Detail’, in I.W. Zartmann (ed.) The 50% Solution. Garden City: Doubleday.

This page intentionally left blank


This page intentionally left blank

10 Normative Institutionalism and EU Foreign Policy in Comparative Perspective Frank Schimmelfennig and Daniel C. Thomas

This research project has challenged prevailing views on the European Union’s development of foreign policy and external relations – both the Intergovernmentalists’ claim that policy-making in this area is dominated (and often fatally undermined) by the member states’ hard pursuit of national interests and counter claims that the path to agreement is smoothed by discursive exchanges that promote durable convergence in national preferences. In particular, the project is organized around a Normative Institutionalist theory of EU policy-making that highlights two ways in which the Union’s substantive and procedural norms, as well as its preexisting policy commitments, enable negotiated agreements among member states with divergent policy preferences. Entrapment is the process by which member states find it difficult to escape the dictates of substantive EU norms and thus accept (however reluctantly) norm-consistent policies that diverge from their actual preferences. Cooperative Bargaining is the process by which veto threats are sidelined by the EU’s procedural norms in favour of consultation and consensus, leading member states to adopt common policies based on mutual compromise. These conclusions serve three purposes. First, we give an overview of the case study findings with regard to the explanatory power of Normative Institutionalism. Second, we explore the conditions under which the Normative Institutionalist mechanisms of policy agreement are more or less likely to be effective. Third, we discuss the implications of these findings for existing scholarship and future research on EU foreign policy. And finally, we consider the implications of our findings for the EU’s ability to promote its values and interests in world affairs. 177

178 Normative Institutionalism and EU Foreign Policy in Comparative Perspective

The power of normative institutionalism Normative Institutionalism has proven to be a robust theory of EU decision-making on foreign policy and external relations whose explanation for the policy-making process and outcome in actual cases is generally (though not universally) superior to that of competing theories. This conclusion is based on the results of 14 case studies examined within the project’s seven empirical chapters (summarized in Table 10.1).

Table 10.1

Negotiation characteristics and normative institutionalism


Negotiation Characteristic

Consistency with Normative Institutionalism

1) Ukraine: support for democratization



2) Ukraine: accession prospect



3) ICC: immunity for UN peacekeepers



4) ICC: bilateral immunity agreements



5) Iraq: support for war


6) Iraq: reconstruction aid



7) Congo: Operation Artemis



8) Turkey: negotiations 1997


9) Turkey: negotiations 2004



10) Turkey: negotiations 2006



11) EPA: negotiation mandate



12) EPA: negotiated offer



13) Climate change: Berlin mandate



14) Climate change: post-2012 regime



Notes: (1) Negotiations are characterized as follows: ENT = entrapment, WENT = weak entrapment, DIS = disentrapment, CBG = cooperative bargaining, LCD = lowest common denominator, DED = deadlock. (2) With regard to the implications of various outcomes for Normative Institutionalism’s core expectations and scope conditions, ENT and CBG are considered strongly supportive (++); DIS, WENT as supportive (+); LCD and DED as non-supportive or contrary (–).

Frank Schimmelfennig and Daniel C. Thomas 179

Our confidence in this conclusion is bolstered by the fact that the case studies covered a broad range of policy areas and institutional settings, all of the cases exhibited clear differences in the initial policy preferences of member states, and most of the case studies provide detailed evidence of the policy-making process. And to be clear, our conclusion that Normative Institutionalism offers the strongest explanation is not to deny that the factors and mechanisms highlighted in other theories and hypotheses may sometimes, if not often, be evident in actual EU foreign policy-making. Nine of the 14 case studies (cases 1, 4, 6, 7, 9, 11–14) demonstrate policy-making processes and outcomes that fit one or both of the Normative Institutionalist hypotheses more closely than they do any of the alternative hypotheses. Of these, all nine exhibit significant evidence of entrapment, while four cases (cases 4, 6, 13, 14) exhibit significant evidence of cooperative bargaining. Furthermore, in three cases (cases 2, 3, 10), we see clear evidence of normative dis-entrapment, where EU actors (one or all) are not expected to behave in accordance with an existing EU norm or policy commitment because external conditions are contrary to the assumptions of the EU norm/commitment or because they have assumed a non-EU role with its own distinctive norms. If many instances of EU policy-making were to exhibit disentrapment, then the percentage of real-world cases to which Normative Institutionalism applies would be significantly reduced, undermining the theory’s explanatory power. But as long as such situations remain relatively rare, as these case studies suggest, we must remember that they are actually consistent with the theory’s assertions. In fact, the disentrapment of the anti-Turkey group of member states in the 2006 negotiations simultaneously entrapped the pro-Turkey camp into supporting a partial freeze in the accession process. In contrast, only two cases (the Iraq war and the 1997 negotiations on Turkey’s EU candidacy) offer clear support for Intergovernmentalism’s competitive bargaining hypothesis, which asserts that negotiation among member states will be dominated by the threat or possibility of veto and thus result either in deadlock or agreement on the lowest common denominator. In both cases, however, the prerequisites for normative entrapment were absent. Two others (Ukraine’s EU accession and ICC immunity for peacekeepers) also exhibit lowest common denominator (LCD) dynamics, but in the context of normative disentrapment. This does not mean that member states did not threaten to veto EU agreement in any of the other cases under examination. But as the second ICC case clearly demonstrates, the consequences of a veto threat are

180 Normative Institutionalism and EU Foreign Policy in Comparative Perspective

minimized when the member states making the threat find itself entrapped by EU norms. None of the case studies conforms to either the normative suasion or policy learning hypotheses, which posit distinctive mechanisms of policy agreement based on preference convergence. Although not encouraging for hypotheses that are so well established in the theoretical literature on foreign policy and European integration, this null finding is actually less definitive than it first appears. Both hypotheses refer to social processes that unfold over time and whose effects may not be immediately evident. In contrast, each of the case studies in this project is designed to examine policy negotiation and choice within a rather limited time period, generally measured in weeks or months rather than years from a starting point where member state preferences were demonstrably divergent. It is thus entirely possible that while normative suasion and policy learning cannot explain short-term movements from divergent preferences to negotiated agreements, they might explain what some EU observers claim is a secular longer-term trend toward preference convergence among member states. Assessing that possibility, however, is beyond the scope of this project’s research design.

Normative institutionalism and the conditions of policy agreement This volume’s framework chapter stipulates several conditions under which rhetorical entrapment and cooperative bargaining are most likely to occur. First, entrapment is expected to overcome policy divergence if there is an uncontroversial norm with clear behavioural implications (determinacy) and the situation in question is consistent with the normative conditions for agreement (relevance). When the norm is contested, indeterminate or irrelevant, the basic conditions of entrapment are not met. In addition, entrapment is facilitated by public attention to the issue (publicity), prior policy commitments (precedent), and by a site of negotiation and deliberation in which EU norms are salient (forum). These conditions make it even harder for actors to escape the normative trap. If the conditions of entrapment are weak or absent, cooperative bargaining is still possible – and again most likely to occur if negotiations take place in a favourable context or forum, in which the EU’s norms of cooperative bargaining are most salient. By contrast, publicity is expected to have ambivalent effects, supporting normative entrapment but undermining cooperative bargaining.

Table 10.2

Conditions of entrapment and cooperative bargaining








1) Ukraine: support for democratization







2) Ukraine: accession prospect




3) ICC: immunity for UN peacekeepers



4) ICC: bilateral immunity agreements






5) Iraq: support for war




6) Iraq: reconstruction aid





7) Congo: Operation Artemis





8) Turkey: negotiations 1997





9) Turkey: negotiations 2004







10) Turkey: negotiations 2006





11) EPA: negotiation mandate




12) EPA: negotiated offer





13) Climate change: Berlin mandate






14) Climate change: post-2012 regime







Notes: (1) When both ‘determinacy’ and ‘precedent’ are coded negative, ‘relevance’ is automatically not applicable (‘N/A’). (2) Publicity is expected to promote entrapment but to undermine cooperative bargaining. (3) Negotiations are characterized as follows: ENT = entrapment, WENT = weak entrapment, DIS = disentrapment, CBG = cooperative bargaining, LCD = lowest common denominator, DED = deadlock.

182 Normative Institutionalism and EU Foreign Policy in Comparative Perspective

Table 10.2 shows the presence and absence of these conditions in the cases studied in the contributions to this volume. All cases involve situations where intra-EU negotiations sought to overcome divergent preferences among major member states. The ‘case’ column displays the demands or policy proposals in question. The values for the conditions are positive if they are expected to work in favour of a positive outcome, negative if they are likely to hinder a positive outcome. The one exception is publicity, which is supposed to favour entrapment but hinder cooperative bargaining. A positive value for determinacy indicates that there is a clear and uncontested norm in favour of the policy proposal. A precedent is present if previous commitments and actions along the lines of the policy proposal exist. A positive value for relevance means that the situation meets the conditions under which the norm prescribes agreement on the proposal, and a positive value for forum indicates that decisions are taken within the institutionalized channels of negotiation and deliberation of the EU. Finally, publicity is present if public attention to the issue is high and/or decisions are taken in a public setting. Richard Youngs’ study of EU policy toward Ukraine (ch.3, cases 1–2) finds that entrapment produced agreement on support for democratization during the Orange Revolution. When, however, Ukraine demanded a membership perspective, the earlier rift among the member state governments reemerged and could not be overcome. In both cases, an uncontested, institutionalized norm existed: support for democracy and a membership perspective for democratic European countries. In the first case (support for democratic change), the reluctance of one group of member states could be overcome because their focus on geopolitical interests was delegitimated by the Union’s strong democracy promotion norm, the situation in Ukraine during the Orange Revolution matched the conditions for the norm to be enacted, public attention was high, there was prior rhetorical commitment to democracy in Ukraine, and policy deliberations took place within the EU. In sum, the constellation of conditions was highly favourable for effective entrapment. By contrast, conditions were less favourable when some member states sought to offer Ukraine a membership perspective: the EU’s prior rhetorical commitment for membership had always been less strong than it was for democratization, public attention had decreased, and the country’s questionable democratic development after the Orange Revolution had the effect of disentrapping un-sympathetic member states. Daniel C. Thomas (ch.4, cases 3–4) compares two cases of EU policy triggered by US efforts to gain immunity from prosecution by the Inter-

Frank Schimmelfennig and Daniel C. Thomas 183

national Criminal Court (ICC) for its citizens and soldiers. Although the recent establishment of the ICC meant that policy precedents were weak, giving in to US demands would have meant compromising the ICC statute, which the EU had strongly supported and which all member states had signed. The two cases are nonetheless distinguished by the different forums in which they occurred: negotiations on immunity for UN peacekeepers took place in the UN Security Council while the issue of bilateral immunity agreements fell within the policymaking scope of the Common Foreign and Security Policy (CFSP). Negotiations at the UN were characterized by low determinacy (one of three EU member states on the Council held the presidency and was thus subject to different norms) and low publicity (they were largely shielded from public view), while negotiations within EU channels were characterized by high normative determinacy (the EU was formally and collectively committed to support the Rome Statute) and high publicity (NGOs and media were fully engaged). The difference in forum, with direct consequences for normative determinacy and publicity, thus explains the emergence of LCD dynamics in the first case and a combination of entrapment and cooperative bargaining in the second. The Iraq crisis of 2002–03 is widely regarded as the most glaring foreign policy failure of the EU. In his study, Jeffrey Lewis (ch.5, cases 5–6) compares the divisions and the breakdown of the cooperation norm among the EU member states on the question of support for the war on Iraq with the agreement on reconstruction aid for Iraq later in the year. The initial deadlock over Iraq policy resulted from a constellation of conditions, in which there was neither a common norm nor a policy precedent that could have generated agreement on supporting the war. In addition, cooperative bargaining was hampered by public posturing and negotiations and deliberations outside the institutionalized channels of EU foreign policy-making. By contrast, the EU eventually reached agreement on multilateral, UN-backed reconstruction aid for Iraq through a mix of entrapment and cooperative bargaining – first, because this kind of aid is in line with the commonly accepted EU norms for global involvement and, second, because the EU actors refrained from public posturing and moved to in camera deliberations. In the case of Congo and Operation Artemis, says Simon Duke (ch.6, case 7), the context conditions for cooperative bargaining and normative entrapment were favourable as well. First, as a humanitarian intervention requested by the UN’s Secretary-General, it clearly conformed to the EU’s normative requirements for a legitimate military operation under European Security and Defence Policy (ESDP). Second, the main

184 Normative Institutionalism and EU Foreign Policy in Comparative Perspective

proponent of the operation, France, sought to conduct it in the framework of ESDP. Third, the failure of the cooperative bargaining norm during the Iraq crisis created strong pressures to conform to the norm in this case. By contrast, publicity and precedent do not seem to have played a major role. Rather, Operation Artemis was used to establish a precedent for joint military operations outside of NATO. Frank Schimmelfennig’s analysis of EU-Turkey accession negotiations (ch.7, cases 8–10) consists of three cases of divergent preferences among the major member states. In 1997, the EU decided not to grant Turkey candidate status as a first step toward accession negotiations. In 2004, the EU agreed on opening negotiations in 2005. Finally, in 2006, the negotiations were partially suspended. In all cases, the treaty-based norm that democratic European countries ought to have a membership perspective was present. In addition, the issue received high public attention and negotiations were conducted within EU fora with the participation of the Commission and the Parliament. By contrast, consistency and precedence vary in line with the outcomes. In 1997, both democracy in Turkey and the prior commitment of the EU to start accession negotiations with Turkey were too weak to generate normative entrapment, resulting in an LCD outcome. In 2004, however, the political situation in Turkey had vastly improved and the EU had committed itself by granting Turkey official candidate status in 1999 and by vowing to follow the same procedures and criteria as in Eastern enlargement. Thus, all conditions for entrapment were present. Yet in 2006 Turkey undermined entrapment by failing to fully implement the Customs Union to Cyprus, and there was no precedent or prior commitment to continue negotiations unaltered. To the contrary, the EU had committed itself to sanctioning Turkey in this case. In his study of the European Partnership Agreements (EPA) with developing countries, Ole Elgström (ch.8, cases 11–12) compares the decisions on the negotiating mandate with that on the final offer the EU made to its negotiating partners. In contrast with the other cases analysed in this volume, the EPA negotiations are characterized by competing normative frameworks and policy commitments: free trade and development. The negotiating mandate put little emphasis on development concerns and clearly favoured the free trade agenda. This had very much to do with the strong position of the Commission in the policy-making process, a new paradigm for relations with developing countries developed in the 1990s, and weak public attention. As Table 10.2 shows, the mobilization of development-friendly public opinion made the difference in adding more development concerns to

Frank Schimmelfennig and Daniel C. Thomas 185

the EU’s negotiating position. According to Elgström’s analysis, public opinion and the NGO sector are less constrained by the EU’s institutional and normative entrapment and cooperative bargaining mechanisms than the member states and can thus have an independent impact on external relations. The issue of targets and timetables in climate policy is one of a strong EU policy commitment, high public attention, and negotiations within the institutionalized framework of the EU. At the same time, it suffers from an indeterminate normative framework. Although the EU has increasingly committed itself to the protection of the global climate and to becoming the global environmental leader, and agreement on targets and timetables is the core reality test for such a commitment, the normative implications of this commitment for the distribution of emissions reductions among the EU member states remained vague. Under these conditions, the negotiations on targets and timetables do not show clear evidence of entrapment. John Vogler’s analysis (ch.9, cases 13–14) suggests, however, that the strong commitment of the EU (and some of its member states in particular) has prevented the negotiations from succumbing completely to competitive bargaining and thus from resulting in a pure LCD outcome. This review of the individual case studies raises a number of analytical questions. Can we generalize the findings across the cases? Are the entrapment and cooperative bargaining hypotheses supported? Which conditions of entrapment and cooperative bargaining are relevant either alone or in combination with others? Are there necessary and/or sufficient conditions of entrapment and cooperative bargaining emerging from this set of case studies? To answer these questions, we offer here a cross-case analysis based on the summary findings in Table 10.2. First of all, the extreme cases reveal broad support for the normative institutionalist framework. As the cases of democracy promotion in Ukraine and the decision to open accession negotiations with Turkey (in 2004) reveal, the presence of all five conditions hypothesized to promote entrapment has indeed resulted in policy agreement through entrapment in spite of initial preference divergence among the member states. Conversely, the Iraq war case with negative scores for all conditions (except for the theoretically ambiguous publicity condition), is the case of the most obvious failure of entrapment and cooperative bargaining. Moving on to the individual conditions, the results are more complex. In logical terms, determinacy and relevance or precedent and relevance is the theorized cause of agreement through entrapment in cases of divergent

186 Normative Institutionalism and EU Foreign Policy in Comparative Perspective

member state preferences. That is, we expect entrapment if there is either a clear and uncontested norm or a clear policy commitment that applies to the situation at hand and requires support of the policy proposal at hand. Indeed, a relevant and determinate norm emerges as a necessary and sufficient condition of the five cases of unambiguous entrapment (cases 1, 4, 6, 7, 9). Where there was a lack of determinacy, or if a single clear and uncontested norm was not relevant to the situation at hand (cases 2, 3, 5, 8, 10–14), the result was no entrapment, disentrapment, or mutual and thus weak entrapment. A relevant precedent proved a sufficient condition of entrapment when determinacy was present as well (cases 1 and 9) but only produced weak entrapment effects when determinacy was absent (cases 13 and 14). It must be said, however, that negotiations on targets and timetables are likely to have resulted in even more competitive bargaining without the EU’s prior commitment to the reduction of greenhouse gas emissions in the latter pair of cases. On the other hand, a relevant and determinate norm did not require precedents in addition to bring about entrapment (cases 6 and 7). Thus, determinacy (in combination with relevance) proved more important than precedent (in combination with relevance). We theorized forum and publicity as non-essential but facilitating conditions for entrapment and as essential conditions of cooperative bargaining. At first glance, the evidence here is mixed. Of the 12 cases that transpire within EU forums, nine (cases 1, 4, 6, 7, 9, 11–14) exhibit significant evidence of entrapment, which suggests a strong relationship. Neither of the two cases of negotiations outside the EU (cases 3, 5) produced entrapment or cooperative bargaining – but we must be careful about drawing strong conclusions from such a small sample. Of the 11 cases that feature high publicity, only six (cases 1, 4, 9, 12–14) exhibit significant evidence of entrapment, which hardly suggests a strong independent effect for this variable. However, the results are more striking with regard to the respective pairs of conditions that we expect would support the entrapment and cooperative bargaining hypotheses. There is strong evidence for the expectation that entrapment would be most likely when negotiations transpire within EU forums and when publicity is high. In fact, of the eight cases that fit this description, six (cases 1, 4, 9, 12, 13, 14) exhibit significant evidence of entrapment. Of the remaining six cases, only three (cases 6, 7, 11) exhibit significant evidence of entrapment. In short, the likelihood of entrapment rises from 50% to 75% when the hypothesized conditions are present.

Frank Schimmelfennig and Daniel C. Thomas 187

On the other hand, the evidence contradicts the expectation that cooperative bargaining would be most likely when negotiations transpire within EU forums and when publicity is low. Of the four cases that exhibit cooperative bargaining, only one (case 6) fits the expected conditions. The other three cases of cooperative bargaining (cases 4, 13, 14) occurred in EU forums but amidst considerable publicity. The effects of forum and publicity are most clearly evident in the paired comparison of observations on the same issue – especially when determinacy, precedent, and relevance remain constant. First, they help to explain the difference in outcome in the two ICC cases. In contrast with the case of immunity for UN peacekeepers in Bosnia-Herzegovina, the EU negotiating channels and the higher public attention in the case of the bilateral immunity agreements clearly contributed to normative entrapment and cooperative bargaining among the member states. Second, the absence and presence of publicity best explains the different outcomes in the EPA cases: whereas weak public attention in the early stages of formulating the negotiation mandate favoured the ‘free trade’ norm supported by the agenda-setting institutional actor (the Commission), the subsequent involvement of a pro-development NGO public accounts for the rebalancing of the final offer to the developing countries. In sum, the ideal conditions to generate foreign policy agreement in situations of divergent member state preferences would be a relevant and determinate norm, a relevant policy commitment or precedent, and negotiations within the institutionalized channels of the EU, and high public attention. However, a determinate and relevant policy norm stands out as the crucial variable: a necessary and sufficient condition. This finding mirrors the results of other studies using a normative institutionalist framework (e.g. Schimmelfennig, 2003; Rittberger and Schimmelfennig, 2006).

Implications for future research These findings have multiple implications for future research on EU negotiations over foreign and security policy and other areas of external relations. Of these, issues related to EU actorness, the conditions of entrapment and cooperative bargaining, normative underdetermination, and issue-linkage deserve particular attention. EU ‘actorness’ This project offers unprecedented insight into a critical determinant of the European Union’s stature as an actor in world affairs: its ability to

188 Normative Institutionalism and EU Foreign Policy in Comparative Perspective

overcome the diverse preferences of its member states and reach agreement on common policies. This framing of the actorness issue builds upon Bretherton and Vogler’s (2006) emphasis on the consistency between the bilateral external policies of the member states and those of the EU. The project thus provides a theoretically-informed explanation for the many foreign policy outputs that are often taken as evidence of EU actorness (e.g., Ginsberg, 2001) and a valuable counterpoint to those who argue that the EU cannot act unless it becomes a unitary state (e.g., Allen, 1998). It suggests that EU actorness in world affairs requires political cohesion, including the sharing of political goals and policy preferences, the adoption of common policies, and the pursuit of common behaviours by EU member states and supranational institutions. Improving our understanding of how and why the political cohesion of the EU varies across issue-areas, relationships, and policy choices poses a challenging and progressive agenda for future research. The aforementioned conclusion that Normative Institutionalism offers a powerful explanation for collective EU decisions on foreign policy should not be confused, however, with the portrayal of the EU as a ‘normative power’ (Manners, 2002). Normative Institutionalism asserts that negotiations among EU member states are shaped significantly by the EU’s substantive and procedural norms and policy commitments, but it does not assert that the EU will necessarily promote normatively ‘good’ outcomes, however those may be defined. As such, the EU’s increasing involvement in military missions overseas may be a problem for a certain conception of the Union as a ‘normative’ (often defined as ‘civilian’) power, but it is not a problem for Normative Institutionalism as an explanation for the sources of EU foreign policy (c.f. Manners, 2006). Conditions of entrapment and cooperative bargaining The principal purpose of the project was to assess the relative power of competing explanations of EU foreign policy-making derived from Normative Institutionalism and Intergovernmentalism as well as theories of social learning and normative suasion. So to conclude that Normative Institutionalism offers the strongest explanation is not to deny that the factors and mechanisms highlighted in other theories and hypotheses may sometimes, if not often, be evident in EU foreign policymaking. As reviewed above, the various case studies in this volume teach us much about the conditions conducive to Normative Institutionalism’s two explanations for how member states reach agreement in the face of diverse policy preferences – entrapment and cooperative bargaining.

Frank Schimmelfennig and Daniel C. Thomas 189

Our conclusions in this regard are nonetheless tentative, as most of the cases were not selected explicitly in order to compare variations in a particular scope condition while holding other conditions constant. Future researchers in this area can therefore build upon our foundation by being more systematic about case selection. Normative under-determination One of the challenges faced by Normative Institutionalism is accommodating the fact that many policy-making situations involve more than one norm or policy precedent with implications for the issue under negotiation. If two or more EU norms are equally relevant to the issue at hand, they may dictate contradictory policies and thus create a situation that the framework paper defines as mutual entrapment. For example, as evident in the Economic Partnership Agreement cases, member states that prefer policy X may find themselves entrapped by the norm dictating policy Y while member states that prefer policy Y are simultaneously entrapped by norm X. Or member states may agree on which EU norm is relevant to an issue, but if the norm is indeterminate, they may differ over its implications for the choice at hand. In either case, the dynamics of entrapment will be under-determining with regard to policy choice. When faced with such situations, member states may engage in cooperative bargaining, or they may make threats in pursuit of their policy preferences, exchange information that could influence each other’s factual understanding of the issue at hand, or seek to persuade each other that one policy option is normatively superior to others. The analytical challenge is not to ignore norms but to achieve greater clarity on how member states act in normatively structured but under-determined situations. Issue-linkage and logrolling This volume has adopted the version of Intergovernmentalism that expects EU foreign policy negotiations to be dominated by competitive bargaining and LCD dynamics. This is consistent with the tendency of most Intergovernmentalists working on EU foreign policy, who treat these dynamics as a fact of life rather than as elements of a hypothesis. As Thomas (ch.2) concedes, though, the logic of Intergovernmentalism does not necessarily indicate that EU negotiations will exhibit these dynamics. Certain constellations of issues and preferences could create incentives for member states to link concessions on off-setting issues and thus achieve mutual gains that avoid the dynamics and consequence of competitive bargaining (McKibben, 2010). For example, König and Junge’s (2009)

190 Normative Institutionalism and EU Foreign Policy in Comparative Perspective

study of Council voting on the Commission’s legislative proposals found that member states often achieve agreement by ‘logrolling across proposals that either belong to the same policy domain or are negotiated during the same period’. The EU policies that result from such Paretoimproving deals would closely resemble Normative Institutionalism’s Cooperative Bargaining model. Although König and Junge’s study did not cover EU foreign policy and external relations, their argument is certainly generalizable to this area. As such, our understanding of EU foreign policy-making will benefit if future researchers specify this logrolling model and distinguish its observable implications from those of Normative Institutionalism more fully than was possible here.

Implications for EU foreign policy-making Sceptics often deride talk of an EU common foreign policy as little more than wishful thinking. As one journalist put it, ‘national selfinterest will consistently trump the EU’s formation of unified policy’ (Vinocur, 2008). Similarly, The Economist declares, ‘Manners may change, but it is national interests that still shape the inner life of the European Union’ (Charlemagne, 2007). Glaring breakdowns in EU unity, such as that evident most recently regarding Kosovo’s declaration of independence, seem to support such scepticism. But the pursuit of a ‘common’ approach to foreign policy, to which the member states are formally committed, is not the same as a ‘single’ foreign policy based on unified institutions, to which only the most ardent Europhiles aspire. The distinction is critical. This study provides real reason to doubt the Intergovernmentalist assertion that EU unity in world affairs will progress no faster or further than the ‘slowest’ member state, which can always threaten to veto any initiative it finds unacceptable. In fact, the study shows that EU member states often reach agreement by pursuing their national interests in a manner that takes into account the values and commitments that have already articulated together. As the case studies demonstrate, it is often this tendency, and not any convergence (coincidental or deliberate) of member states’ policy preferences, that allows the EU to reach agreement on common policies and thus to exert influence on the world stage. How robust is this tendency? As shown by the utter collapse of EU unity over whether or not to support the US-led invasion of Iraq in early 2003, pre-existing norms do not prevent determined national governments in the EU from breaking ranks when they see fit. However,

Frank Schimmelfennig and Daniel C. Thomas 191

many of the case studies show member states reaching agreements on common policies that depart significantly from their particular policy preferences. In fact, as shown in the second Iraq case study, the EU’s substantive and procedural norms sometimes enable member states to re-establish unity even when policy preferences remain divergent. But EU governments that want the Union to exert an international influence commensurate with its vast population and wealth would be wise to remember that however tempting it may be to break ranks on a particular issue, the normative foundation of agreement on common policies is more easily broken than rebuilt.

Works cited Allen, D. (1998) ‘Who Speaks for Europe? The Search for an Effective and Coherent External Policy’, in J. Peterson and H. Sjursen (eds) A Common Foreign Policy for Europe? Competing Visions of the CFSP. Abingdon: Routledge, pp.41–58. Bretherton, C. and Vogler, J. (2006) The European Union as a Global Actor, Second Edition. London: Routledge. Charlemagne (2007) ‘The Polish Farewell’, The Economist, 1 December. Ginsberg, R.H. (2001) The European Union in International Politics: Baptism by Fire. Lanham, MD: Rowman and Littlefield. König, T. and Junge, D. (2009) ‘Why Don’t Veto Players Use Their Power?’, European Union Politics, 10(4): 507–34. Manners, I. (2002) ‘Normative Power Europe: A Contradiction in Terms?’, Journal of Common Market Studies, 40: 235–58. Manners, I. (2006) ‘Normative Power Europe Reconsidered: Beyond the Crossroads’, Journal of European Public Policy, 13(2): 182–99. McKibben, H.E. (2010) ‘Issue Characteristics, Issue Linkage, and States’ Choice of Bargaining Strategies in the European Union’, Journal of European Public Policy, 17(5): 696–709. Schimmelfennig, F. (2003) ‘Strategic Action in a Community Environment: The Decision to Enlarge the European Union to the East’, Comparative Political Studies, 36(1/2): 156–83. Rittberger, B. and Schimmelfennig, F. (2006) ‘Explaining the Constitutionalization of the European Union’, Journal of European Public Policy, 13(8): 1148–67. Vinocur, J. (2008) ‘A Chance for Europe to Face the New Truth’, International Herald Tribune, 23 June.

11 Norms and All That: Progress in Research on EU Foreign Policy Thomas Risse

‘Economic giant, political dwarf!’ – ‘The EU does not get its act together’ – ‘The capability-expectation gap is widening’ (on the latter see Hill, 1993; for a most recent criticism of the EU’s inability to develop a coherent foreign policy see Howorth, 2010). These are only a few of the many indictments of the European Union’s (EU) foreign and security policy to be found in the editorials, but also occasionally in the scholarly literature. And yet it moves! As Daniel Thomas writes in this volume (ch.1), there have been more than 1,000 common strategies, common positions, and joint actions under Common Foreign and Security Policy (CFSP) since 1993 and more than 2,000 foreign policy statements by the EU Council and Presidency between 1995 and 2008. In other words, cooperation on foreign policy matters is the rule rather than the exception in the EU – and the empirical case studies in this volume document it in detail. They show beyond doubt that the EU has emerged as a foreign policy actor which pursues rather coherent foreign policies – if it wants to and if the conditions are right. Thus, the EU’s actorness in foreign and security policy matters is no longer in question (see Sjursen, 2006). This is the first good news about this volume. It could be stated even more loudly and more clearly – particularly in light of the bad press that the EU’s foreign policy got recently in the context of Catherine Ashton’s appointment as EU ‘foreign minister’ (High Representative of the Union for Foreign Affairs and Security Policy, to be precise) and the turbulent negotiations surrounding the European External Action Service (EEAS). The second good news about this volume is that the authors stay clear of the rather confusing debate about whether or not the EU is a ‘normative’, ‘civilian’, or even ‘metrosexual’ power (see e.g. Manners, 2002; Sjursen, 2006; Diez, 2005; Duchêne, 1972; Khanna, 2004). This 192

Thomas Risse 193

debate has shown two things – surprise, surprise: First, the EU tries to externalize its own values and norms in its external relations. It is no different from any other great ‘normative power’, from the Roman Empire to Stalin’s USSR and the post-World War II United States. But, second, the EU does this inconsistently, that is, the promotion of values often conflicts with other foreign policy goals such as security or economic interests. What else should one expect? In contrast, this volume’s purpose is analytical and empirical. It uses standard social science theories – institutionalism in this case – to explain the variation in the extent to which EU member states cooperate in foreign policy matters. The editor develops a set of clearly specified hypotheses which are then subject to empirical evaluation. This alone represents huge progress in the study of EU foreign policy – a subfield of comparative foreign policy analysis which has suffered for a long time from a lack of theory-guided empirical research (on comparative foreign policy analysis in general, see Carlsnaes, 2002). But my job in this comment is not to write a praising blurb for the volume, but to offer some critical thoughts. I will do so by focusing, first, on the volume’s effort at theory-building and, second, on the empirical case studies and the degree to which they support or reject the volume’s main hypotheses. My bottom line is that I have yet to be persuaded by the strong claim in Frank Schimmelfennig’s and Daniel Thomas’s conclusions (ch.10, this volume) that ‘(n)ormative institutionalism has proven to be a robust theory of EU decisionmaking on foreign policy and external relations whose explanation for the policy-making process and outcome in actual cases is generally (though not universally) superior to that of competing theories’.

Normative institutionalism – Rationalist or sociological institutionalism in disguise? Ever since Peter Hall and Rosemary Taylor wrote their famous review essay (Hall and Taylor, 1996), political science scholars thought in terms of three institutionalisms: rationalist or neoliberal (the latter as the dominant notion among international relations scholars); historical, and sociological. Then Vivien Schmidt introduced ‘discursive institutionalism’ as a fourth one (Schmidt, 2008; Schmidt, 2002) and now Daniel Thomas et al. suggest ‘normative institutionalism’ as a fifth one borrowing from Guy Peters (Peters, 2005). I am not convinced, however, that we need five different institutionalisms, as I argue in the following.

194 Norms and All That

According to Thomas (ch.2), normative institutionalism ‘emphasizes the behavioural impact of standards of appropriateness established by the community’s normative and policy commitments’. He then suggests that EU member states have developed two meta-norms, namely joint action as an intrinsic value as well as consistency and coherence in EU policy-making. Moreover, EU member states have established a number of substantive (democracy, human rights, and so forth) as well as procedural norms, among them regular communication and consultation as well as decision-making by consensus (see particularly Lewis, 2000, 2005). These norms then lead to substantive and procedural entrapment in the decision-making process for those member states whose preferences are deemed inconsistent with them – under certain scope conditions (Thomas, ch.2, hypothesis one). They also lead to cooperative bargaining with the aim at minimizing differences and reaching consensus (Thomas, ch.2, hypothesis two). As Thomas and Schimmelfennig emphasize in the conclusions of this volume, normative institutionalism should not be confused with ‘normative power Europe’, even though the EU’s substantive norms specified above resemble the concept. At first glance, this description of what normative institutionalism is supposed to mean comes awfully close to standard accounts of sociological institutionalism, particularly of the March and Olsen variety (March and Olsen, 1989, 1998) who have introduced the ‘logic of appropriateness’ into the social science literature. Yet, normative institutionalism in the interpretation by Thomas and Schimmelfennig appears to contain an understanding of actors who are only very thinly socialized. Jeffrey Checkel calls this ‘Type I socialization’: ‘The key is agents knowing what is socially accepted in a given setting or community. Following a logic of appropriateness, then, means simply that conscious instrumental calculation has been replaced by conscious role playing’ (Checkel, 2005: 804). In contrast, Type II socialization would refer to actors who adopt the interests or even identity of the community in which they participate (ibid). At this point, the norms of the community become part of the social fabric of actors, they exert constitutive effects. The second of these is explicitly not what Daniel Thomas and Frank Schimmelfennig have in mind, since their normative institutionalism only affects the behaviour of actors, not their definition of interests and preferences. In their understanding, the preferences of EU member states are exogenously given, while the community norms only affect their behaviour. Otherwise, the entrapment hypothesis would make no

Thomas Risse 195

sense, since it posits that actors behave in a way that is inconsistent with their preferences, because they follow community norms. In other words, norms appear to become intervening variables between interests and preferences which is precisely the formula in Stephen Krasner’s conceptualization of rational choice or neoliberal institutionalism when he defines international ‘regimes as intervening variables’ (Krasner, 1983). Yet, rational choice institutionalism does not assume that actors behave contrary to their given interests (which is what the term ‘entrapment’ implies), only that actors have mixed motives, i.e., they are interested in cooperation, but would also like to pursue their egoistic interests. International institutions then constrain the behaviour in the sense that they privilege cooperation. In sum, I am not sure where Daniel Thomas and Frank Schimmelfennig would situate themselves between standard accounts of rational choice (or neoliberal) institutionalism, on the one hand, and a sociological institutionalism emphasizing thinly socialized actors, on the other. What I am sure about is that ‘normative institutionalism’ can be interpreted either way – and, thus, is less different from conventional institutionalisms than the authors in this volume seem to suggest.1 Most rational choice or neoliberal institutionalists would probably interpret the case studies as consistent with an understanding of actors as boundedly rational and exhibiting low discount rates for the future (long-time horizons) which then lead to diffuse reciprocity as a guiding norm (on the latter, see Keohane, 1986). In this interpretation, ‘entrapment’ means nothing else than actors foregoing short-term gains in the interest of long-term cooperation. Normative institutionalism adds content to this by focusing on substantive and procedural norms,2 but it appears to buy into the basic rational choice assumptions. As Daniel Thomas puts it (ch.2), ‘(t)hey thus compromise their preferences and “play along” with the norm-consistent policy because they expect the social rewards for doing so to exceed the costs of compromise’. Note that Thomas explicitly uses rational choice language of cost-benefit calculations in this quote. And whenever these long-term benefits are not obvious or particular actors hold very intensive and conflicting preferences (see Jeffrey Lewis’ chapter on the debate over the Iraq war and Frank Schimmelfennig’s chapter on the Turkish accession negotiations after 2006), there will either be no cooperation or lowest common denominator solutions. But most sociological institutionalists would probably also regard the case studies as consistent with type I socialization as defined above. Thinly socialized actors know what is expected of them and what is

196 Norms and All That

appropriate in a given situation as a result of which they ‘play along’. This is all about role-playing. The way in which ‘entrapment’ is introduced in this volume (ch.2) suggests such an understanding. Particularly conditions 1 (determinacy) and 2 (precedent) are consistent with role-playing. But whenever the behavioural norms are indeterminate or there is no precedent, the logic of consequentialism trumps the logic of appropriateness, since there is no clear role which actors can ‘play along’. As James March and Johan Olsen put it: ‘When preferences and consequences are precise and identities or rules are ambiguous, a logic of consequences tends to be more important. When identities and their implications are clear but the implications of preferences or expected consequences are not, a logic of appropriateness tends to be more important’ (March and Olsen, 1998: 952). Table 10.2 in Chapter 10 by Schimmelfennig and Thomas is consistent with this interpretation which has the added advantage that one does not have to resort to terms such as ‘disentrapment’. In sum, the theoretical status of normative institutionalism remains unclear throughout the volume. It can be regarded as rationalist institutionalism in disguise, but it can also be interpreted as a variant of sociological institutionalism emphasizing thinly socialized actors. ‘Entrapment’ of thinly socialized actors vs. constitutive norms of a community The volume uses the concept of ‘entrapment’ to denote that EU member states might comply with certain rules to which they have committed even if these norms contradict their preferences or current interests. It was Frank Schimmelfennig who first developed the concept of rhetorical entrapment: ‘In an institutional environment like the EU, political actors are concerned about their reputation as members and about the legitimacy of their preferences and behavior. Actors who can justify their interests on the grounds of the community’s standard of legitimacy are therefore able to shame their opponents into norm-conforming behavior and to modify the collective outcome that would have resulted from constellations of interests and power alone’ (Schimmelfennig, 2001: 48; see also Schimmelfennig, 2003).3 Daniel Thomas has to be applauded for pushing this concept further by suggesting scope conditions under which entrapment takes place. Schimmelfennig’s – and by extension Daniel Thomas’s – problem is that entrapment or shaming presupposes actors who, first, know about the community’s norms and values (thin socialization at least, see above) and who, second, for some reason, then decide to comply with these norms

Thomas Risse 197

even if they contradict their short-term preferences. This is problematic on theoretical grounds, because one cannot have it both ways. On the one hand, Thomas and Schimmelfennig want to stick to pre-social rational actors whose interests and preferences remain exogenous to the institution, the European Union in this case. On the other hand, the community norms and values are supposed to exert some effects on behaviour – if only via ‘entrapment’ – which assumes social actors who have accepted the legitimacy of these norms. Once again, this argument sits uneasily between rationalist and sociological institutionalism. My own take at these issues is from a perspective that takes sociological institutionalism, constructivism, and the mutual constitutiveness of agents and structures more seriously (for details see Adler, 2002). In my view, the norms and rules emphasized by Thomas and Schimmelfennig are constitutive for the European Union in the sense that they define what it means to be a member of the EU. Values such as democracy, the rule of law, human rights, and market economy as well as procedural norms such as regular communication, consultation, and consensus-seeking do not just prescribe certain behaviour; they are also constitutive for the EU as a community. They are part of the EU’s social identity (for details see Risse, 2010). This is why ‘entrapment’ works. Strategic framing by some member states or by the European Commission is successful, because it connects to the social identities of member states. Does this mean that identity always trumps other goals or perceptions of interests? Does it mean that actors never violate constitutive norms? Not at all. One does not have to resort to terms such as ‘disentrapment’ to realize that, first, constitutive norms can be violated, since all rule-following is probabilistic rather than determinate (otherwise, we could forget about the mutual constitutiveness of agency and structure, see Wendt, 1987). Second, constitutive norms might contradict each other in particular situations. For example, my social identity as a German might conflict with my European identity with regard to the question of whether or not the EU should bail out Greece (this resembles condition 1 ‘determinacy’ in ch.2). Last not least, constitutive norms might contradict other goals which actors might pursue. This latter case is particularly significant with regard to foreign policy in which this is the rule rather than the exception. Democracy promotion, for example, which can be regarded as a constitutive norm of the EU (Magen et al., 2009) often conflicts with other goals such as security concerns. This is normal in foreign policy and no reason for indictments of EU democracy promotion as hypocritical. It is rarely the

198 Norms and All That

case that actors pursue one goal only in their external relations: balancing between often conflicting goals is normally the case. In the following, I present my own interpretation of the empirical case studies in this volume based on an understanding of sociological institutionalism which emphasizes the constitutive role of norms affecting actors’ identities. I particularly discuss those cases in which the EU was confronted with conflicting norms, interests, and preferences, and offer an alternative explanation of how these conflicts were managed. Conflicting norms and EU foreign policy First, the volume covers situations in which social identities themselves are contested (in Thomas’ language, condition 1 of determinacy is not met). In the case of Turkish membership in the EU (Schimmelfennig, ch.7, this volume), there are at least two constitutive, but nevertheless conflicting norms. According to the democracy principle, Turkey should be admitted into the union as long as it accepts liberal values and complies with them. But there is also the cohesion principle suggesting that the EU must have the institutional absorption capacity to take on new members (unfortunately, Schimmelfennig does not discuss this norm, which was adopted after the Copenhagen criteria of 1993). This latter principle at least challenges Turkish membership aspirations. On top of it, there is also a very powerful identity discourse in many member states according to which a predominantly Muslim country does not belong in an EU that they define as a community of white Christians. Given how contested Turkish membership has been in the EU all along, it is all the more puzzling that membership negotiations have been opened at all. I agree with Schimmelfennig that the shifting positions of the German and Greek governments explain to a large degree the ups and downs in these negotiations. I also agree that domestic politics in key member states explains the variation in EU behaviour to a large extent. But I am not convinced that ‘entrapment’ does a lot of work in situations of deep contestation (precisely because Thomas’ first scope condition is not met). Second, the volume covers situations in which overlapping memberships in different communities lead to conflicting constitutive norms. The almost complete breakdown of the EU’s constitutive norms in the run-up to the Iraq war in 2003 seems to resemble this case (Lewis, ch.5, this volume). In this case, fundamental norms of the transatlantic community and of the EU clashed head-on leading to a temporary breakdown of both communities. Great Britain, Spain, and other member states as

Thomas Risse 199

well as accession candidates joined the US-led ‘coalition of the willing’ thereby violating CFSP consultation norms, while a majority of EU members led by Germany and France publicly opposed the war and even organized a counter-coalition against the US in the UN Security Council. As Jeffrey Lewis points out in his chapter, the most striking point of this EU as well as transatlantic crisis is how quickly member states agreed on a joint EU position with regard to rebuilding Iraq which not only restored the EU community but also the transatlantic one. Such attempts at reconstructing a community are precisely what is to be expected after severe violations of constitutive norms which not only regulate behaviour but also define the fundamentals of a community. Once again, constitutive norms as well as other norms only affect behaviour in a probabilistic fashion; one should, therefore, expect them to be violated occasionally. However, one would expect different behaviour and different discourses to follow cases of norm violations. Actors are then likely to engage in all kinds of symbolic, discursive and other behavioural action in order to restore the community. This is precisely what happened in the EU in the aftermath of the 2003 Iraq war. Last not least, the volume covers several cases in which EU foreign policy simply tried to find compromises among conflicting goals, whether constitutive or not. This is how I see both the Ukraine case and the EU reaction to the US attempt to ‘undo’ the Rome Statute of the International Criminal Court (ICC). In the case of Ukraine’s Orange Revolution (Youngs, ch.3, this volume), the EU’s constitutive norm of promoting democracy and human rights worldwide clashed with the goals of some member states to keep relations with Russia on a stable footing. Both goals are legitimate, but the further the Orange Revolution proceeded, the less sustainable became a policy not to support the Ukrainian opposition. In other words, a constitutive norm took precedent over a conflicting goal. One can call this ‘entrapment’, but a much simpler account would argue that a constitutive norm prevailed over a conflicting, but less well entrenched foreign policy goal. The two ICC cases (Thomas, ch.4, this volume) exhibit multiple dynamics, but the first of them can also be interpreted as a clash between a constitutive norm and a conflicting foreign policy goal. Member states had invested a lot of effort in negotiating the Rome Statute and the EU had publicly committed itself to the ICC, which was seen as a concrete manifestation of the Union’s long-standing commitment to promote human rights. But when the Bush administration tried to use the UN Security Council as well as bilateral agreements with every state in the international system to achieve ICC immunity for its citizens and armed

200 Norms and All That

forces and thereby to undo the Rome Statute, the EU was faced with a dilemma. Some member states wanted to accommodate the US but all member states were legally committed to the ICC, which made it difficult to accomplish the first goal. The EU’s decision not to raise the issue of immunity for UN peacekeepers in CFSP forums, which enabled the UK to satisfy the US within the Security Council, can be interpreted as a pragmatic solution to this dilemma. The second sub-case, in which the UK and Italy ignored domestic opinion and threatened to veto any EU position on bilateral agreements that failed to satisfy the US, but then back-tracked and supported EU guidelines that de facto prohibited any agreements that the US would be willing to sign, appears to show member states prioritizing a constitutive norm over conflicting foreign policy goals. Only this second case could be interpreted as ‘entrapment’.

Conclusions Let me summarize my comments on this volume: 1. I agree with the authors’ attempts to use standard international relations theories – institutionalism in this case – to explain EU foreign policy decisions. In doing so, the volume demonstrates in detail that EU foreign policy is much more cohesive and cooperative than many analysts have it. The volume also pays attention to alternative explanations, particularly to a more realist or intergovernmentalist account.4 However, I would have preferred a bit more attention to the institutional settings of the various cases, namely whether decision rules of unanimity or qualified majority voting (QMV) made a difference at all. 2. My main criticism concerns the theoretical account, however. I am not convinced that normative institutionalism provides much analytical leverage over rationalist institutionalism, on the one hand, and sociological institutionalism, on the other. The authors’ theoretical claims are consistent with a version of rationalist institutionalism emphasizing boundedly rational actors with long-time horizons and information constraints. But they are also consistent with a version of sociological institutionalism focusing on thinly socialized actors who know what is appropriate under specific circumstances. I do not see the ‘value added’ of yet another institutionalism. 3. I have also raised some concerns about the concept of ‘entrapment’. I have argued in this comment that ‘entrapment’ via naming and shaming or other discursive devices only works if the actors consider

Thomas Risse 201

the norms legitimate. This implies, however, that actors consider themselves members of a social community, i.e. that the norms are constitutive. I then offer an alternative reading of some of the empirical case studies based on the notion of conflicting norms including conflicting social identities. In conclusion, however, I would like to repeat my impression that the volume as a whole represents a large step forward toward theoryguided empirical research on EU foreign policy. The emerging actorness of the EU in external affairs is too important not to subject it to rigorous testing of competing explanations. This is precisely where the field of EU foreign policy studies should be.

Notes 1 Unfortunately, there is no discussion in the volume of how normative institutionalism is related to Hall’s and Taylor’s ‘three institutionalisms’. 2 I thank Daniel Thomas for alerting me to this. 3 In my earlier work, I have also used the term ‘argumentative entrapment’ to describe situations in which actors accept arguments rhetorically at first and then cannot back-track without being accused of insincerity in a discourse (see e.g. Risse, 2000). 4 In contrast, none of the empirical chapters do a serious job with regard to policy learning and normative suasion. This would have required covering the various negotiations in much more detail in order to evaluate their deliberative quality. In addition, it is hard to see how the volume can study policy learning, since the empirical cases cover rather short time-periods, as Daniel Thomas and Frank Schimmelfennig admit in the conclusions.

Works cited Adler, E. (2002) ‘Constructivism in International Relations’, in W. Carlsnaes, B. Simmons and T. Risse (eds) Handbook of International Relations. London: Sage, pp.95–118. Carlsnaes, W. (2002) ‘Foreign Policy’, in W. Carlsnaes, T. Risse and B. Simmons (eds) Handbook of International Relations. London: Sage, pp.331–49. Checkel, J.T. (2005) ‘International Institutions and Socialization in Europe: Introduction and Framework’, International Organization, 59(4): 801–26. Diez, T. (2005) ‘Constructing the Self and Changing Others: Reconsidering “Normative Power Europe”’, Millennium, 33(3): 613–36. Duchêne, F. (1972) ‘Europe’s Role in World Peace’, in R. Mayne (ed.) Europe Tomorrow: Sixteen Europeans Look Ahead. London: Fontana/Collins, pp.32–47. Hall, P.A. and Taylor, R.C.R. (1996) ‘Political Science and the Three New Institutionalisms’, Political Studies, 44: 952–73. Hill, C. (1993) ‘The Capability Expectation Gap, or Conceptualizing Europe’s International Role’, Journal of Common Market Studies, 31(4): 305–28.

202 Norms and All That Howorth, J. (2010) ‘The EU as a Global Actor: Grand Strategy for a Global Grand Bargain?’, Journal of Common Market Studies, 48(3): 455–74. Keohane, R.O. (1986) ‘Reciprocity in International Relations’, International Organization, 40(1): 1–27. Khanna, P. (2004) ‘The Metrosexual Superpower’, Foreign Policy, July–August. Krasner, S.D. (1983) ‘Structural Causes and Regime Consequences: Regimes as Intervening Variables’, in S.D. Krasner (ed.) International Regimes. Ithaca NY: Cornell University Press, pp.1–21. Lewis, J. (2000) ‘The Method of Community in EU Decision-Making and Administrative Rivalry in the Council’s Infrastructure’, Journal of European Public Policy, 7(2): 261–89. Lewis, J. (2005) ‘The Janus Face of Brussels: Socialization and Everyday Decision Making in the European Union’, International Organization, 59(4): 937–71. Magen, A., Risse, T. and McFaul, M. (eds) (2009) Promoting Democracy and the Rule of Law. American and European Strategies. Houndmills, Basingstoke: Palgrave Macmillan. Manners, I. (2002) ‘Normative Power Europe: A Contradiction in Terms?’, Journal of Common Market Studies, 40(2): 235–58. March, J.G. and Olsen, J.P. (1989) Rediscovering Institutions. The Organizational Basics of Politics. New York, London: The Free Press. March, J.G. and Olsen, J. P. (1998) ‘The Institutional Dynamics of International Political Orders’, International Organization, 52(4): 943–69. Peters, G. (2005) Institutional Theory in Political Science. London: Routledge. Risse, T. (2000) ‘“Let’s Argue!” Communicative Action in International Relations’, International Organization, 54(1): 1–39. Risse, T. (2010) A Community of Europeans? Transnational Identities and Public Spheres. Ithaca NY: Cornell University Press. Schimmelfennig, F. (2001) ‘Liberal Norms, Rhetorical Action, and the Enlargement of the EU’, International Organization, 55(1): 47–80. Schimmelfennig, F. (2003) The EU, NATO, and the Integration of Europe: Rules and Rhetoric. Cambridge: Cambridge University Press. Schmidt, V.A. (2002) ‘Does Discourse Matter in the Politics of Welfare Adjustment?’, Comparative Political Studies, 35(2): 168–93. Schmidt, V.A. (2008) ‘Discursive Institutionalism: The Explanatory Power of Ideas and Discourse’, Annual Review of Political Science, 11: 303–26. Sjursen, H. (2006) ‘What Kind of Power?’, Journal of European Public Policy, 13(2): 169–81. Wendt, A. (1987) ‘The Agent-Structure Problem in International Relations Theory’, International Organization, 41(3): 335–70.

12 Interests, Power and the EU’s Role in International Security: A Sceptical Response to Normative Institutionalism Anand Menon

I very much enjoyed reading the chapters in this volume, and appreciate the opportunity to be able to offer something in the way of a critical commentary on the theory of normative institutionalism. As its title implies, the focus of this chapter is very much on the author’s own area of interest – security policy. Consequently, the focus here is on the chapters dealing with Ukraine, the International Criminal Court, Iraq, and particularly with Operation Artemis in the Democratic Republic of Congo. The overarching goal of this project was to develop and test a general theory of EU decision-making on foreign affairs. Normative Institutionalism (Thomas, ch.2, this volume) stresses twin mechanisms that exert a profound impact on foreign policy decision-making within the EU. First, ‘entrapment’ – whereby member state behaviour is shaped by ‘shared perceptions regarding which policy options are consistent or inconsistent with pre-existing EU norms and commitments’. Second, ‘cooperative bargaining’ implies that intra-EU bargaining is ‘characterized by a great deal of give-and-take in the context of an intensive search for solutions that are acceptable… for the greatest number of member states’. The various authors succeed admirably in achieving their objective. The coherence between the empirical chapters and the theoretical outline is impressive and testifies to an editorial job well done. More substantively, the theory of Normative Institutionalism offers a useful conceptual framework through which to analyse and compare the different elements of EU foreign policy. All this being said, I have my reservations, which revolve as much around those issues the volume does not address as those it does. The following discussion focuses on three of them. First, power, which seems lacking from many of the accounts in this volume. Apparent consensus 203

204 Interests, Power and the EU’s Role in International Security

can often mask the deployment of power to secure agreement. This is related to a second point, namely case selection. For obvious reasons, it makes sense to focus attention on specific missions or actions undertaken by the EU, but doing so overlooks those cases where either discussion was prevented or no decision was reached, which in turn may mask the exercise of power. Finally, process. There is both too little and too much in the various chapters concerning decision-making processes. Too little in that it would be useful to have some broader idea of how ‘cooperative bargaining’ works in practice. Too much in that the various chapters do not spend much time at all considering the outcomes produced by the bargaining processes they investigate. At a minimum, this leaves grounds for some scepticism concerning the ‘normative’ aspect of consensus policy-making within the Union.

Power Largely missing from the accounts in this volume is a conception of power and its importance. Decisions do not always indicate the emergence of consensus, but can, rather, reflect the wielding of power. For example, there was no clear consensus in 1998 on the need for the EU to be given any kind of military capacity, yet the French and British government, via their ambitious Saint Malo declaration, managed to bounce the issue onto the agenda and thereafter effectively to cajole even their most reluctant partners into accepting the creation of the European Security and Defence Policy (ESDP). Indeed, a striking feature of much of the activity undertaken under the aegis of ESDP has been the need for those missions undertaken – particular the larger military operations – to enjoy strong support from one of the larger member states. Julian King, then British Ambassador to the Political and Security Committee, was a central player in drumming up political will and resources for the EU intervention in Aceh. The French have been similarly entrepreneurial when it comes to cajoling partners to participate in missions in Africa, as Simon Duke (ch.6, this volume) makes clear. As former Commissioner for External Relations Chris Patten has put it, ‘there is no European foreign policy on a big issue unless France, Germany and Britain are on side’ (cited in Toje, 2008a: 134). Jeffrey Lewis cites an EU official as pointing out that there is no CFSP mechanism to overcome divergence between these ‘big three’ (Lewis, ch.5, this volume). Leadership, in other words, can be as important as consensus building when it comes to garnering support – however equivocal – in the

Anand Menon 205

Council. President Sarkozy’s vigorous activity following the Russian invasion of Georgia on 7 August 2008 was viewed in many quarters, and not least by the French President himself (Le Figaro, 18 August 2008), as an example of effective EU action. Certainly the diplomacy he conducted, and the eventual dispatch within 20 days of an EU monitoring mission numbering some 200, smacked of decisive (if limited) action. Yet much of this action was carried out in the teeth of opposition on the part of several member states. Estonia, Latvia, Lithuania and Poland were all critical, for instance, of Sarkozy’s willingness to sign up to a ceasefire agreement between Moscow and Tbilisi that did not mention the inviolability of Georgian territorial integrity. At the end of the day, however, openly challenging the actions of the French Presidency was not seen as possible. Power, of course, is not simply important in terms of relations between member states: it also plays an important role in shaping the Union’s actions towards the outside world. The Union has often struggled to arrive at clear policies in its dealings with both the United States and China. Indeed, one observer has argued that the reason why Africa has been the location of so many EU military missions is precisely because of the possibility of deploying without trespassing on the interests of the great powers (Toje, 2008b: 208). In this volume, Richard Youngs’ (ch.3) account of the vacillations that marked policies towards Ukraine in the lead up to the Orange Revolution illustrates all too clearly the difficulties member states encounter when attempting to coordinate policies with implications for relations with Russia. The point here is not that Normative Institutionalism should simply add another variable to its account – clearly the whole point of theory is to privilege certain explanations over others. Rather, it would be a useful exercise to include a power-based explanation as an alternative to test. The exercise of power within the Union can lead to non-lowest common denominator outcomes and thus entails more than simply competitive bargaining. Comparative studies of the EU and NATO (within which the United States, of course, plays a hugely influential role) might be an interesting way to tease out the role that unequal member state power plays in shaping decision-making within international institutions. Certainly in the area of ESDP (where, clearly, large disparities between member states in terms of military capabilities are fundamental) it would be interesting to see how a power-based explanation compared to Normative Institutionalism. An initial hypothesis could perhaps be that consensus for potential missions will only

206 Interests, Power and the EU’s Role in International Security

be possible if either France or the United Kingdom strongly favours such an intervention.

Cases Whilst such instances of the role of unequal state power are illustrative, they are far from unique. Yet observers sometimes struggle to come to grips with their implications because of the cases they select for study. It has become commonplace to assess ESDP (or Common Security and Defence Policy (CSDP), as it is now known) in terms of the number of missions that have been carried out under its aegis (Menon, 2009: 229). In this collection, Simon Duke (ch.6) focuses on one such mission – Operation Artemis, carried out in the Democratic Republic of Congo (DRC) in 2003. Certainly there is much to be learned from a case study of this sort, and space constraints mean that the author could hardly provide such an in-depth analysis of more than one mission. Nevertheless, a focus on those missions deployed provides an unrepresentative view of the ability of member states to forge consensus over CSDP decisions. Arguably, as much could be learned from studies of those instances where the member states fail to achieve consensus, and hence the EU doesn’t act at all. Take a couple of examples. In 2003, as fighting escalated in Darfur, member states considered the deployment of a military force. The nature of the crisis seemed tailor made for EU intervention, with real potential for a small scale mission managing to stop the fighting and provide significant humanitarian benefits. Yet member state reluctance meant that no force was deployed. However congruent with EU norms such a mission might have been, the Union proved unable to act. Crucially (referring back to the argument made above about power) the reluctance of both Britain and France to countenance military intervention proved a deal breaker (Toje, 2008a: 136–7). Similarly, when fighting erupted in the DRC in late 2008, the UN Secretary-General himself appealed directly to the Union for an intervention force. This was hardly surprising, as the Union had spent the years since Operation Artemis developing capabilities designed to facilitate precisely this kind of intervention – notably much-heralded EU battlegroups (Lindstrom, 2007). Given explicit UN sanction (which had been lacking in Darfur five years previously), and the humanitarian suffering on the ground, the occasion seemed tailor-made for an EU military deployment. Yet consensus, once again, proved elusive. Such failures to reach agreement between member states clearly challenge the rather

Anand Menon 207

positive picture of consensus decision-making painted by many of the contributors to this volume. In other cases, rather than trying but failing to decide on a common policy, the member states have failed to reach the decision-making stage. Such instances represent classic cases of ‘nondecision-making’, or the ‘practice of limiting the scope of actual decision-making to “safe” issues by manipulating the dominant community values, myths, and political institutions and procedures’ (Bachrach and Baratz, 1963: 632). Although the study of ‘nondecisions’ certainly poses serious methodological challenges, they are important components of any institutional setting. Thus, while the Political and Security Committee has frequently been portrayed as ‘cooperative and consensus seeking’ (Howorth, 2010: 10; see also Meyer, 2006), this is sometimes because member states conspire to prevent contentious issues from reaching its agenda: the 15 Ambassadors never formally discussed the Iraq crisis.

Process In fairness, the case studies in this volume purport to provide a theoretical explanation of the cases they do consider, rather than of those they do not. Central to this explanation is a view of how EU decisionmaking works, and of the importance of the particular ways in which member states interact. A combination of the consensus norm, consultation reflex, and member state identification with Europe is said to spawn cooperative bargaining within the Council (Thomas, ch.2). This emphasis on bargaining style as opposed to bargaining rules is not new. Over two decades ago, Fritz Scharpf distinguished between problem solving, bargaining and confrontational negotiating styles – though the contemporary obsession with rational choice meant that these insights were largely overlooked in favour of arguments concerning the role of decision rules in shaping the ‘joint decision trap’ (Scharpf, 1988). Others still have focused more squarely on EU security policies in coining terms such as ‘supranational intergovernmentalism’ to capture the ‘normative socialization processes’ that characterize the workings of the Political and Security Committee (Howorth, 2010; Alecu de Flers et al., 2011). There is much to be said for focusing on more than simply formal rules when investigating decision-making. After all, an emphasis on formal rules in much Institutionalist literature has led to an emphasis on path dependence and the change resistant nature of institutions, which hardly provides a compelling account of the rapid development of ESDP (Menon, forthcoming).

208 Interests, Power and the EU’s Role in International Security

The various contributions to this volume certainly provide extensive evidence related to the prevalence or otherwise of different negotiating tactics. Thus, Thomas (ch.2) clearly distinguishes between competitive and cooperative bargaining in terms of the willingness of member states to resort to either veto threats or consultation and consensus seeking. The individual case studies build impressively on this analysis. In his analysis of the ICC case, Thomas (ch.4) focuses extensively on the UK’s initial use and then withdrawal of the veto threat, followed by engagement in negotiations clearly oriented toward median compromise, once the issue had been publicized and reframed. Lewis (ch.5) tells a clear and detailed story of the way in which member states moved from outright disagreement over the Iraq towards consensus over the need for EU involvement in reconstruction efforts. This evidence is compelling and serves to underline the explanatory potential of Normative Institutionalism. My comments in this regard relate more to further work that could be done to ‘flesh out’ the approach, rather than being criticisms of a failure to include more in this single volume. First, it would be interesting to know how it is that member states come to accommodate the positions of their partners. By what process and for what reasons do national governments come to value consensus more than the staunch defence of narrow national interests? What aspects of the EU institutional system promote such developments, and how? Is there a stronger impulsion towards consensus in the Union than in other international organizations? Answering these questions would clearly require a detailed analysis of the evolution of national bargaining positions and the role of interaction in the Council in fostering agreement. Such details are crucial in order that we can discount alternative hypotheses. Without them, for instance, it becomes very difficult to exclude the null hypothesis that the EU itself plays little or no role in bringing about the reconciliation of member state negotiating positions. Thus it is not clear to what extent member states, which were themselves committed in both constitutional texts and by their own rhetoric to norms such as democracy, would have acted any differently in the case of Ukraine’s Orange Revolution had the EU not existed. Moreover, it would be equally useful to have slightly more detail and clarity when it comes to distinguishing Normative Institutionalism from alternative Institutionalist approaches that stress, not consensus, but preference convergence. On the surface, the distinction seems fairly clear. Thus, the focus of Normative Institutionalism (Thomas, ch.2)

Anand Menon 209

is on how member states ‘overcome divergent preferences to reach agreement’. National preferences are thus exogenous to institutional setting, and not liable to be affected by it. The issue here is one of tactics – sacrificing short-term benefits in the expectation of longerterm gains. Yet elsewhere in the same chapter, Thomas suggests that a habit of consultation has ‘produced a partial convergence in the strategic cultures and identities of the member states’. Where, then, does consensus stop and convergence begin? Are the habits of consultation that lead to consensus bargaining the same as those that might lead to shifts in preferences?

Outcomes Finally, and perhaps most importantly, the whole point of EU foreign policies is to exert influence over the world outside the Union’s borders. Consequently, no analysis of the way in which decisions are arrived at is complete without a consideration of the impact of these processes on the outcomes produced. This is all the more so given that an element of normative institutionalism is substantive. The notion of ‘entrapment’ encapsulates a supposed desire on the part of member states to ensure that policy options are consistent with prevailing EU norms and commitments. Several issues emerge when this aspect of the theory is considered. First, it is not clear precisely how ‘normative’ Normative Institutionalism really is. Daniel Thomas (ch.2) makes relatively limited claims regarding a ‘normative’ element to member state behaviour, underlining that member states do not necessarily actively seek to abide by EU norms for their own sake: Although the policy preferences of EU Member States may diverge on particular issues, they value ‘coherence and consistency’ in EU foreign policy and value being seen as acting in accordance with the community’s normative and policy commitments. As a result… the policymaking behavior of Member States is shaped significantly by shared perceptions regarding which policy options are consistent or inconsistent with pre-existing EU norms and commitments. Those Member States whose policy preferences are seen as inconsistent with the EU’s substantive norms or policy commitments are less willing to insist on their preferences and more acquiescent to those with norm-consistent preferences. They thus compromise their preferences and ‘play along’ with the

210 Interests, Power and the EU’s Role in International Security

norm-consistent policy because they expect the social rewards for doing so to exceed the costs of the compromise. Portrayed as such, normative entrapment represents an instrumental rather than a properly ‘normative’ phenomenon. But one may therefore question how different it is from the notion of ‘specific reciprocity’ (Keohane, 1986) more commonly associated with a competitive bargaining approach. It is, after all, the expectation of future reward, rather than the norms themselves, that account for member state willingness to compromise. Second, in order to take the claim that there is a ‘normative’ element to EU policies seriously, we must have evidence that their intended impact is in conformity with the Union’s norms. That is to say, there is a big difference between appearing to abide by EU norms, and acting in such a way as to actually do so. Again, Thomas makes the point, emphasizing that member states value ‘being seen’ to act in accordance with these norms. Understandably owing to space constraints, the authors cannot spend too much time analysing the outcomes of the decisions they focus on. Yet to claim a ‘normative’ aspect to policies based merely on how they are negotiated seems to me to be somewhat unconvincing. Moreover, several of the case studies actually underline the limits of any ‘normative’ basis to member state actions. Richard Youngs’ (ch.3) excellent and detailed analysis of EU policies towards Ukraine makes clear that the Union’s initial indecision and reluctance to intervene was overcome only when, with events proceeding rapidly on the ground, member states had no real choice but to take a stance. It is in the chapter on CSDP that Simon Duke (ch.6) makes perhaps the most ambitious claims regarding the ‘normative’ aspect of EU policies – claims which possibly go beyond the rather nuanced approach taken in Thomas’ theoretical chapter. Duke argues that member state actions have been influenced by the ‘prevailing normative values’ and ‘collective understanding of “appropriate” action’. In particular, he emphasizes the normative commitments made by member states in the Treaty on European Union and European Security Strategy to preserve peace and strengthen security. Yet a brief analysis of the outcome even of Operation Artemis itself reveals the shortcomings of such claims. The mission was the object of much criticism from humanitarian groups, described by one as ‘totally insufficient’ to meet challenges on ground, not least because of its limited duration in terms of both space and time (International Crisis

Anand Menon 211

Group, 2006). Médecins Sans Frontières pointed out that European forces had managed only to guarantee civilian safety only in very limited areas (Médecins sans Frontières, 2003). Two academic observers were more forthright still, describing the mission as ‘limited, brief, riskaverse and ultimately ineffective’ (Haine and Giegerich, 2006). It is hard to square such analyses with the Union’s own ambitious and ubiquitous declarations of high-minded principle. Artemis, moreover, set the tone for several subsequent EU interventions into Africa. For all the rhetoric about EU values, member states have been able to agree only on small scale missions that fail to tackle the real issues on the ground. Thus, African specialists criticized the Security Sector Reform mission in the DRC because of its small size, limited budget and lack of helicopters (Gegout, 2007). EUFOR Congo in 2006 was similarly criticized for being too short to achieve much (Gegout, 2007). Once the Germans had been persuaded by Paris of the need to intervene in the DRC in 2006, Berlin insisted on the application of a series of restrictive caveats that restricted its 100 troops to Kinshasa (Alecu de Flers et al., 2011). The one norm with which member states have seemed truly anxious to comply is that of ensuring the viability of the EU as a security actor. Thus, as one senior official put it, ‘Artemis became an EU operation because it might prove the value of an EU military capability for peacekeeping, not because of what it could do for Africa’ (Interview, senior Council official, Brussels, 2004). Contrary to what is claimed, or at least implied by some contributions to this volume, therefore, the ‘normative’ element of EU security policy is somewhat self-regarding, rather than revolving round ends-orientated norms concerning human rights, democracy promotion or conflict prevention. In the case of reconstruction aid to Iraq, it seems, to this observer at least, that an (interest-based) desire to illustrate that the Union was capable of deploying the instruments of ‘soft power’ weighed as heavily in the minds of member states as any concern with EU norms relating to promoting stability or development in Iraq. So much for the operations actually carried out. Equally, those operations the Union has declined to take on also have real world implications. Lewis (ch.5) makes a compelling case about the way member states moved from discord to (relative) harmony over Iraq, but the point remains that, on the big issue – the conflict itself – the former proved decisive. There could, moreover, perhaps have been no mission more in keeping with self-professed EU norms than that which some proposed for Darfur in 2003, yet sending such a mission was never

212 Interests, Power and the EU’s Role in International Security

formally discussed within the Council. Similarly, no formal vote was taken when the Union decided against intervention in the DRC in December 2008. Ironically, Normative Institutionalism suggests that member states may prefer nondecisions to unsuccessful bargaining in cases like these because they provoke less public interest than do negative decisions, shielding member states from the kinds of publicity that make normative entrapment more likely. The evidence from ESDP suggests that EU norms are deployed by member states both instrumentally and flexibly. National governments are certainly keen to be seen to be complying with their oft-repeated principles. And when the tension between action and value becomes too severe (as in the case of the Orange Revolution) they will act. On other occasions, however, norms seem to serve more as ‘window dressing’ than as inspiration for EU actions (see ch.6, for example). Actual outcomes thus often seem to reflect a desire to be seen to comply with norms rather than any actual desire to comply. Having considered and then decided against intervention in Darfur in 2003, EU officials were quick to begin questioning whether the situation in that country really accounted to ‘genocide’ (Toje, 2008a: 137). Perhaps more insidiously, there is evidence that some member states have come to see the Union as a useful instrument for the pursuit of their own narrow interests. Dressing up missions in a Union flag serves to provide legitimacy and to shield the member state concerned from searching questions regarding its motives. Thus France supplied 1785 out of 2200 troops for Operation Artemis, which ‘was more a French operation with an EU cover, than an EU operation led by the French’ (Grignon, 2003). As one critical observer (Bailes, 2008: 120–1) pungently put it: The Union’s ‘more militarily active members are deliberately casting the ESDP in the role of the “nice cop”: as the shop front where they advertise themselves as friendly and “safe” interveners, with bloodied hands discreetly held behind backs. This kind of double-think may be highly utilitarian in terms of maintaining the EU’s coherence self-confidence and image: but this author would hesitate to call it ethical behaviour on the part of either the states that practice it or the EU that lends itself to the ploy.’ In this sense, rather than allowing for the pursuit of EU normative ambitions, CSDP might actually allow member states to provide themselves with cover whilst pursuing more traditional neo-imperialist

Anand Menon 213

policies – a tactic referred to in the rationalist institutionalist literature as ‘dirty laundering’ (Abbott and Snidal, 1998: 19). Some evidence for this is provided by the changing attitudes of member states towards CSDP missions, which in raises questions relating to the limited conception of ‘learning’ outlined by Thomas (ch.2) as an alternative to Normative Institutionalism and highlights the importance of change over time. Thomas focuses in particular on the way member states may learn from each other when confronting complex, technical issues. An alternative conceptualization relates to the way member states change their beliefs on the basis of their own experience (rather than via negotiations with partners). This has been all too evident in development of ESDP, most particularly when it comes to suspicions on the part of some member states that certain of their partners are using the EU to pursue narrow national interests. Simon Duke (ch.6) argues that France’s use of the Union for Operation Artemis helped assuage any concerns on the part of its partners that this was ‘solely a French operation in EU guise’. Yet those concerns have become more marked over time, as member states learned from experience with repeated EU missions in sub-Saharan Africa. German officials in particular are prone to expressing their concern lest Germany be ‘instrumentalized’ by states such as France that have interests in Africa. Such doubts were one of the reasons for the hesitations and delays that preceded the deployment of EUFOR RD Congo (Alecu de Flers et al., 2011). Finally, whilst consensus building might have positive connotations for those interested in the workings of the EU itself, its practical impact is far less sanguine. The consensus requirement explains the risk-averse nature of EU foreign policy decisions (Laïdi, 2010) and the Union’s tendency to act reactively rather than proactively. Thus, member states hesitated before taking a clear stand on the Orange Revolution, (reacting far more slowly than, say, the United States), and failed to offer a membership perspective to Ukraine, which would have implied real costs. ESDP missions have tended to be small scale and not to address the real problems on the ground (particularly for missions outside the Balkans, where member states find it easier to agree on more far-reaching interventions). Related to this, there is a potential tension between the cooperative bargaining and ‘entrapment’ elements of the theory. In the case of ESDP, rather than attempting to act in accordance with the Union’s normative principles, member states are usually more focused on doing what is necessary to secure agreement. The need to convince often

214 Interests, Power and the EU’s Role in International Security

reluctant partners militates against the rapid deployment of coercive power (Toje, 2008b: 207; Rynning, 2003: 487). As a consequence, they tend to select those issues most conducive to consensus. And these issues conform to a somewhat depressing pattern, in that ‘the lower the level of commitment, the higher the likelihood of achieving consensus’ (Toje, 2008a: 132).

Conclusion It would clearly be misleading to attempt to derive general theoretical insights from the focus in this short essay on a single EU policy area. After all, this discussion suffers from some of the shortcomings it identifies in the project at a whole, not least in terms of the limited number of cases on which it draws. ESDP is probably not a representative policy sector. For one thing, ‘cheap talk’ is always a serious possibility given that member states can approve ESDP missions without needing to contribute assets to them. Unlike in, say, first pillar matters, therefore, it is not a question of legally binding decisions that impose adjustment costs on all member states. Moreover, and profiting from my role as a discussant, I have drawn somewhat selectively on examples from this policy area, providing a doubtless somewhat partial account. Yet the case of ESDP serves, at the least, to pose some interesting questions about the claims of Normative Institutionalism. In particular, the foregoing analysis has questioned the extent to which it is genuinely either institutionalist or normative, given the (relative) lack of attention paid to the processes via which member states move towards agreement on foreign policy decisions, and the – more striking – absence of congruence between the Union’s actions and its declared normative principles. I have also attempted in this short discussion to question what seems to me a bias in favour of ‘consensus’ as something to be valued. Experience suggests that the search for consensus leads to a highly restrictive set of policy outcomes, with normative values abandoned in favour of decision-making convenience. The quest for consensus, in other words, is not necessarily a quest for effective action. As Margaret Thatcher put it, in typically unequivocal terms, ‘consensus seems to be the process of abandoning all beliefs, principles, values and policies. So it is something in which no one believes and to which no one objects’. This is hardly a recipe for truly effective EU foreign policies. All this being said, the authors, and particularly the editor, are to be congratulated for their ambition in attempting to devise a new

Anand Menon 215

theoretical framework for explaining EU foreign policy. My argument is that Normative Institutionalism might overlook certain key issues, and cannot, itself, provide a convincing explanation for certain crucial elements of EU actions. However, this is an argument based on only a partial view of those actions, and as such hardly counts as an adequate refutation of a general theory of how EU member states overcome divergent preferences to arrive at common policies across the full range of foreign policy issues. Happily, we have here the basis for a rewarding dialogue between different theoretical approaches that can only enhance our understanding of the foreign policies of the European Union.

Works cited Abbott, K.W. and Snidal, D. (1998) ‘Why States Act Through Formal Institutions’, The Journal of Conflict Resolution, 42(1): 3–32. Alecu de Flers, N., Chappell, L. and Müller, P. (2011) ‘The EU’s Foreign and Security Policy: Incremental Upgrading of Common Interests and the Effects of Institutionalised Cooperation’, in Falkner, G. (ed.) The EU’s Decision Traps: Comparing Policies. Oxford: Oxford University Press. Bachrach, P. and Baratz, M. (1963) ‘Decisions and Nondecisions: An Analytical Framework’, The American Political Science Review, 57(3): 632–42. Bailes, A.J.K. (2008) ‘The EU and a “Better World”: What Role for the European Security and Defence Policy?’, International Affairs, 84(1): 115–30. Gegout, C. (2007) ‘The EU and Security in the Democratic Republic of Congo: Unfinished Business’, CFSP Forum, 4(6): 5–9. Grignon, F. (2003) ‘The Artemis Operation in the Democratic Republic of Congo: Lessons for the Future of EU Peace-keeping in Africa’, The Challenges of EuropeAfrica Relations: An Agenda of Priorities (Tivoli Tejo Hotel, Lisbon). Haine, J.-Y. and Giegerich, B. (2006) ‘In Congo, a Cosmetic EU Operation’, International Herald Tribune, 12 June. Howorth, J. (2010) The Political and Security Committee: A Case Study in ‘Supranational Intergovernmentalism’. Paris: Sciences Po, Centre d’Études Européennes. International Crisis Group (2006) Congo Crisis: Military Intervention in Ituri. Keohane, R.O. (1986) ‘Reciprocity in International Relations’, International Organization, 58(1): 137–74. Laïdi, Z. (2010) ‘Europe as a Risk Averse Power: A Hypothesis’, Garnet Policy Brief. Lindstrom, G. (2007) Enter the EU Battlegroups. Paris: European Union Institute for Security Studies. Médecins sans Frontières (2003) Ituri: Unkept Promises? A Pretence of Protection and Inadequate Assistance. Menon, A. (2009) ‘Empowering Paradise? ESDP at Ten’, International Affairs, 85(2): 227–46. Menon, A. (forthcoming) ‘Power, Institutions and the CSDP: The Promise of Institutionalist Theory’, Journal of Common Market Studies. Meyer, C.O. (2006) The Quest for a European Strategic Culture: Changing Norms on Security and Defence in the European Union (Basingstoke: Palgrave Macmillan).

216 Interests, Power and the EU’s Role in International Security Rynning, S. (2003) ‘The European Union: Towards a Strategic Culture?’, Security Dialogue, 34(4): 479–96. Scharpf, F.W. (1988) ‘The Joint-Decision Trap: Lessons from German Federalism and European Integration’, Public Administration, 66: 239–78. Toje, A. (2008a) ‘The Consensus Expectations Gap: Explaining Europe’s Ineffective Foreign Policy’, Security Dialogue, 39(1): 121–41. Toje, A. (2008b) ‘The European Union as a Small Power, or Conceptualizing Europe’s Strategic Actorness’, European Integration, 30(2): 199–215.

13 Norms, Institutions and EU Foreign Policy: Advancing the Research Programme Michael E. Smith

The question of whether the EU provides a tangible ‘value-added’ to European foreign policy (EFP) actions beyond the aggregated contributions of its member states has inspired a large literature. In other words, are there endogenous mechanisms at work within EU institutions that lead EU member states to pursue foreign policies they otherwise might not adopt? If so, what are these mechanisms and how do they impact foreign policy-making processes among EU member states and EU institutions? The more orthodox (i.e., realist or intergovernmental) answer to these questions rejects the possibility of an independent EU influence; in this view, EU foreign policy actions are merely a product of bargains among EU member states based on their pre-existing policy preferences. The contributions to this volume add to this debate by considering the role of institutional factors in determining EU foreign policy actions, which may help to moderate or even condition the pursuit of national positions in EFP deliberations on major global problems. Moreover, by focusing on actual EU decisionmaking procedures across a range of foreign policy issue-areas, and by examining both more-successful and less-successful examples of EU deliberations with a single set of hypotheses, the volume has done a major service to the study of EFP. Indeed, the literature lacks a consensus on what might be called a unified theory of European foreign policy-making, largely because the empirical content of EFP – decisions, norms, and policies/actions – is produced by multiple actors and institutional procedures depending on the issue-area or problem at hand (Smith, 2008). This fact creates very difficult conceptual and methodological problems that must be clarified before undertaking any theoretically-informed analysis of EFP. Although the contributions to this volume address some of these 217

218 Norms, Institutions and EU Foreign Policy

problems, and thus – taken as a whole – make a plausible prima facie case regarding the importance of normative factors in the negotiations of the EFP outcomes analysed in their case studies, there is considerable scope for refinement and clarification of the overall argument. However, it must also be said that the ambitions of the volume are quite considerable in view of the range of issues and outcomes covered within, and the contributors deserve much credit for attempting to take on this challenge. This essay is written in hopes of driving the debate forward and building onto what we think we already know about EFP, particularly in terms of normative/institutional factors. In the sections that follow, I first examine the notion of ‘normative institutionalism’ as a theoretical device for explaining EU foreign policy decisions. I then take a closer look at the methodological implications and challenges of adopting such a perspective, as reflected in the arguments found in the various case studies. One critical issue here is the fact that an analysis of negotiation dynamics (as attempted in this volume) is different from, and more complicated than, an analysis of foreign policy-making or decision-making. Assuming one is not using formal modelling or game theory (as with the cases in this volume), negotiation analysis requires a high degree of knowledge about the intensity and ranking of preferences held by various actors involved in the process. As we shall see below, this inside information varies widely depending on the case at hand. In the third and final section, I set out some guidelines for conducting future research on the making of EU foreign policy.

Conceptual issues: What is ‘normative institutionalism’? There is little doubt today about the potential of the EU to wield great influence over certain problems in world politics. This power is most apparent in international economic affairs, and is now expanding into the security realm with various judicial, policing, and even military operations conducted under the auspices of the Common Foreign and Security Policy (CFSP) and the Common Security and Defence Policy (CSDP). The progressive expansion of these capabilities since the 1960s and 1970s, when the EU could barely even discuss many difficult problems in world politics, raises a number of interesting theoretical questions related to foreign policy analysis and international cooperation/ organization. Most fundamentally, we must consider the extent to which membership of, and regular participation within, various EU institutions directly

Michael E. Smith 219

conditions EU member states to adopt certain foreign policies even when they might hold initially divergent policy preferences (obviously if national preferences are convergent before institutionalized deliberations occur, then there is little scope for independent institutional effects on state behaviour). In my view, there are at least three basic positions on this point, ranging from the ‘minimalist’ to the ‘maximalist’. The minimalist position is based on realism and asserts that EU institutions can have no independent impact on either the preferences or policies of EU member states; all EFP outcomes are the result of (typically lowestcommon-denominator, or ‘LCD’) competitive intergovernmental bargains, particularly among the EU’s major powers. An intermediate position, which is based on functional or instrumental institutionalism, asserts that EU institutions can help EU member states find common policies by pooling information (i.e., learning or suasion) and providing opportunities for side-payments, logrolling, and other joint decision mechanisms, especially when general preferences must be pooled into specific common policies. The maximalist position, which is generally consistent with social constructivism, asserts that not just member state policies but also their basic preferences can be modified, if not inspired in the first place, thanks to their intensive participation in EU institutions (particularly in terms of the discourse on the content/purpose of EFP). One immediate concern, then, is the conceptual difference between ‘normative’ institutional effects and policy outcomes, as compared to other such effects and outcomes. In other words, what is meant by ‘normative’ and how do we know such behaviour when we encounter it in world politics? According to the conceptual framework advanced in this volume (Thomas, ch.2), normative effects can be distinguished from competitive/LCD bargaining and from two other types of institutional effects (learning and suasion). Thus, even when LCD bargains and/or learning/suasion are not evident or operative, EU member states still might be able to find consensus due to norm-conditioned processes of ‘cooperative bargaining’ (i.e., finding a more median position rather than deadlock or an LCD outcome), and/or because of normative ‘entrapment’. This is a condition whereby EU member states are ‘constrained’ in their foreign policy actions once they have already committed themselves to a particular set of norms, such as the promotion of democracy or the protection of human rights, and have taken an initial discursive step in defining a policy problem in terms of those norms; these two conditions are labelled as ‘determinacy’ and ‘relevance’ (Schimmelfennig and Thomas, ch.10, this volume). Further, such entrapment is more likely when the public pays greater attention to the issue

220 Norms, Institutions and EU Foreign Policy

(the condition of ‘publicity’); prior EU policy commitments are in place (‘precedent’) and the issue is negotiated in a setting in which EU norms are salient (‘forum’). Thus, entrapment is most likely when the conditions of determinacy, relevance, publicity, precedent, and forum are in perfect alignment, and least likely when most or all of these conditions do not hold. The case of supporting democratization in Ukraine (Youngs, ch.3, this volume) represents a perfect example of entrapment, while the case of allowing immunity for UN peacekeepers despite the norms of the International Criminal Court (ICC) represents a perfect example of ‘disentrapment’ (Thomas, ch.4, this volume), meaning that EU member states are largely free to determine their own policies on an issue. All of the other cases in the volume fall somewhere in between these extremes but generally suggest an important role for norms during the negotiation of specific EU foreign policies; moreover, under these conditions (i.e., neither full ‘entrapment’ nor ‘disentrapment’) the possibility of ‘cooperative bargaining’ may exist and allow EU member states to adjust their positions in the service of a common policy even though the five basic conditions (determinacy, relevance, publicity, precedent, and forum) may not hold. This argument begs two questions: how to distinguish normativelyinspired effects on EU foreign policies from ‘non-normative’ effects, and the extent which EU member states really feel ‘entrapped’ by such normatively-inspired policies. As I have argued elsewhere (Smith, forthcoming), all foreign policies can be seen as ‘normative’ if they serve some basic value shared by the community (such as security or economic prosperity); as used in this volume, however, ‘normative’ seems to refer to any a priori standard of appropriate behaviour or policy goal have been adopted by EU member states to guide their collective decision-making when dealing with any specific foreign policy problem. The contributors to the volume also seem to be fairly agnostic regarding whether such norms help identify the EU as a ‘civilian’ or ‘normative’ or ‘civilizing’ power (Manners, 2002), which is in fact a refreshing way of approaching this question. However, the question of why EU member states actually comply with pre-existing norms (regardless of their source or justification) may deserve more attention than it receives in this volume, for it forces us to critically examine the point raised above: what is the real value-added of the EU in the realm of foreign policy-making? In other words, does the EU merely encourage its member states to live up to the promises they make (which seems to be the approach favoured in this volume), or does the EU in fact represent and reinforce a new set of norms/goals (i.e., cosmopolitan or post-Westphalian; see Linklater, 2005)

Michael E. Smith 221

that EU member states ‘discover’ when attempting to collectively determine the fundamental EU interest or value at stake in dealing with a certain problem? Although many analysts (though not necessarily in this volume) still debate the role of ‘strategic/rationalist’ goals versus ‘normative/ ideational’ goals, this is a false and misleading distinction. Major powers in world politics often attempt to assert their goals according to general or universal principles, and it is certainly possible for a so-called ‘civilian power’ (such as the EU) or for a so-called ‘liberal hegemon’ (such as the US) to pursue illiberal or even illegal/aggressive foreign policy goals. Moreover, to the extent that the EU protects, upholds, exports, or projects such goals in its EFP actions, such ‘aspirational’ goals (Botcheva and Martin, 2001), at least in principle, cannot easily be traced to a material interest (such as expanding foreign markets or securing supplies of natural resources), though they are still rational or strategic goals. As others have pointed out, however (Youngs, 2004; Hyde-Price, 2006), distinguishing between such goals (i.e., materialist vs. non-materialist, or ‘strategic’ vs. ‘normative’) as the ‘true’ rationale for certain EFP behaviours is far easier said than done, especially when such goals might in fact reinforce each other. Finally, as the chapters in this volume focus on both ‘norms’ and ‘policies’ as part of the overall causal chain, it is not entirely clear whether we should pay attention to the EU’s justification of its policies (i.e., the norms supposedly served by the policies), or the actual empirical content/effects of those policies, when determining whether a priori ‘normative’ factors are operative. Similarly, the supposed link between normative causes and effects is the process of ‘framing’ a policy in reference to a pre-existing norm, yet the only way to identify successful ‘framing’ is when EU member states are convinced by it (and then agree to a common policy) when deliberating a certain foreign problem. In other words, there may be an element of tautological reasoning going on here as the supposed cause (the ‘power’ of preexisting norms) is confused with its supposed effects (agreement on EU foreign policies). Moreover, the volume defines the dependent variable – EFP ‘outputs’ – in a very wide fashion, when in fact the commitment and resource costs of these outputs (i.e. Council Conclusions versus a CSDP military operation) could vary quite widely and call into question the extent to which EU member states really feel ‘entrapped’ or not. These facts raise important concerns about causality and the overall ‘entrapment’ hypothesis, as I shall elaborate in the next section.

222 Norms, Institutions and EU Foreign Policy

The contributors are on more solid ground, however, in attempting to distinguish between substantive and procedural norms, although these precise terms are not used throughout the volume. Although both the ‘entrapment’ and ‘cooperative bargaining’ hypotheses rely heavily on the power of procedural norms (i.e., consultations, confidentiality, coherence, consistency, consensus), most of the case study contributions focus instead on the role of substantive norms and normatively-consistent policy commitments, which are far more problematic to analyse in a causal fashion for reasons to be discussed below. Indeed, only Lewis’ case study (ch.5, this volume) on the Iraq War focuses primarily on the consequences of violating the EU’s fundamental procedural norms as a prelude to examining the extent to which such a clear violation then undermined the prospects for a substantive EU policy on the issues. Assuming, therefore, that we focus on procedural norms before examining actual substantive norms or policies, we must again confront the question of why EU member states would follow these norms even if doing so appears to undermine or contradict their individual policy preferences. The theoretical chapter to the volume (Thomas, ch.2) offers five hypotheses regarding the conditions and mechanisms of EU agreement on common policies: normative entrapment, cooperative bargaining, competitive (i.e., LCD) bargaining, policy learning, and normative suasion. However, there is clearly some overlap in the definition of these hypotheses (i.e., two types of ‘bargaining’ and two options – policy learning and suasion – where ‘learning’ is involved); the difference between rhetoric-bound ‘normative entrapment’ and learning-based ‘normative suasion’ in particular seems far too slight to justify distinct hypotheses unless it can be proven that all or most EU member states in fact held ‘fixed’ preferences in the former situation or ‘malleable’ preferences in the latter one. As all EU foreign policy outcomes are predicated on some degree of ‘give and take’ when discussing options, this distinction seems difficult to sustain as an analytical tool. Even more important, these are not all mutually exclusive hypotheses. Indeed, it is entirely plausible that some aspects of all of these dynamics are at work during specific EFP negotiations (or are at work on some EU member states); if so, a more synthetic approach to the analysis of EFP policymaking is in order rather than one which attempts to determine which single hypothesis or causal mechanism offers the most convincing explanation.

Michael E. Smith 223

Methodological issues These concerns can be appreciated more clearly by examining the seven empirical chapters, all but one of which analyse two more subcases within the context of a single foreign policy problem or issuearea, for a total of 14 negotiation outcomes coded in various ways in the Conclusion to the volume (Schimmelfennig and Thomas, ch.10). As the Conclusion claims that the overall case study outcomes are more consistent with the two normative institutionalist hypotheses (i.e., the ‘entrapment’ and ‘cooperative bargaining’ hypotheses) than with any of the alternatives, my purpose in this section is to raise questions about these claims based on my reading of the cases and the Conclusion to the volume. One immediate concern in a volume composed of case studies involves the mechanism for case study selection. As suggested above, the concept of a clear and distinct normative/ideational (or constitutive/ aspirational) basis for certain EU foreign policy behaviours (which is in turn predicated on the collective support of EU member states) must lead to some kind of typology to distinguish such goals from other, more materialist, objectives. Then one would select some cases based on the presence (or strength) of normative factors and other cases on the absence (or weakness) of such factors (i.e., choosing cases based on variation on the primary independent variable). In my view this is a necessary ‘first cut’ in determining the possibility of normative institutional effects, and such a typology is unfortunately lacking in this volume. However, if the goal of the volume is to determine whether there is some consistency between previously stated normative goals and actual EU foreign policies, then the cases should at least vary according to the presence and absence of such norms to avoid selection bias. The contributors have done an admirable job here, as eight cases out of 14 indeed claim the presence of a pre-existing norm (i.e. the condition of ‘determinacy’ noted above), while the six others claim the lack of such a norm; moreover, five of those eight ‘norm determinacy’ cases also show consistency with the entrapment and/or competitive bargaining hypotheses. Two other cases also show ‘weak’ entrapment, plus some degree of competitive bargaining, when a pre-existing norm was lacking but a clear EU policy precedent was in place (see Table 10.2 in Schimmelfennig and Thomas, ch.10). Finally, two cases – both on Economic Partnership Agreements (EPAs) – demonstrate ‘weak entrapment’ in the absence of both conditions of ‘determinacy’ and

224 Norms, Institutions and EU Foreign Policy

‘precedent’; this outcome raises some interesting questions about the overall normative thesis guiding the volume, as we shall see below. For additional variance, the case-study contributions in this volume also fall under two general headings: ‘foreign and security policy’ (Youngs, ch.3, Thomas, ch.4, Lewis, ch.5, and Duke, ch.6) and ‘external relations’ (the chapters by Schimmelfennig, ch.7, Elgström, ch.8, and Vogler, ch.9). These two categories are based on the traditional division between intergovernmental versus supranational methods of making EFP (i.e., the CFSP/ESDP method vs. the Community method), which is the most prominent way to distinguish between two sets of procedural norms in the making of any EU policies, whether domestic or foreign. However, Schimmelfennig’s study (ch.7) of Turkey’s possible accession to the EU does raise a flag; in my view decisions regarding accession to the EU should not be treated as foreign policy decisions given the stakes involved and the generally irreversible nature of allowing a new state to join the EU. And given their drawn-out nature, especially in the case of Turkey, it is also difficult to isolate one or two (or three, as in this example) ‘cases’ as the distinct moments when an important decision or negotiation took place. Instead, accession negotiations/decisions are far closer to grand history-making decisions, somewhat similar to the periodic re-negotiations of EC/EU treaties over the years. This point is reinforced by the threat of some EU member states to veto Turkish membership (or subject it to a national referendum as part of the ratification process), a critical phase of policymaking that obviously lies outside the scope of the three ‘case studies’ in the chapter. In my view, most foreign policy goals can be grouped under three general headings: security, economic prosperity, and value projection (Smith, forthcoming) and all of the ‘normative’ EFP goals examined in this volume would lie within at least one of these categories. However, even this simple typology is confounded by the actual nature of EFP actions, which comprise a very wide range of issue-areas almost equal to the foreign policy behaviours of a major power such as the US. This fact leads us to another methodological problem: how do we distinguish an actual EFP ‘action’ or ‘decision’ as an important dependent variable, and how should these actions/decisions be treated as individual or discrete ‘case studies’? Based on the discussion above, it seems we have at least four issue-areas represented by the case studies, each with its own normative/institutional dynamics and stakes: 1) foreign/ security policy (Lewis, Duke); 2) economic prosperity (Elgström, Vogler); 3) value projection (Youngs, Thomas); and 4) accession (Schimmelfennig). Vogler’s

Michael E. Smith 225

study (ch.9) of climate change could be interpreted as a security issue if one defines ‘security’ in broad terms; however, since it mainly focuses on the intra-EU distributional aspects of reducing carbon emissions, it is most appropriate to treat this as an economic issue. Overall, I believe the cases covered by Youngs and Thomas (on the democratic transition in Ukraine and the International Criminal Court, respectively) represent the most obvious examples of the EU’s pursuit of constitutive/ aspirational goals as reflected by EFP decisions/actions, although each of the other cases sheds some light on this type of EFP as discussed further below. To begin with the foreign/security policy chapters, Duke’s study (ch.6) of the ESDP operation in the Democratic Republic of the Congo (DRC) is the one analysis covering a single ‘case’: the EU’s decision to launch Operation Artemis in mid-2003. Here Artemis is judged as a case of entrapment based on the presence of the ‘determinacy’ condition: the general support of the EU for UN peacekeeping operations, and the specific emphasis on such operations in the EU’s own Security Strategy, which was being formulated at the time. Yet the actual analysis indicates a much wider range of variables – not ‘normative’ factors – at work: the desire to mount a (non-NATO) ‘Europe only’ operation, the desire to demonstrate EU resolve in the aftermath of its failure to find a common position on the Iraq War (also in operation in the Lewis chapter; see below), a need to build operational capacity/experience in the ESDP, and, above all, the absolutely critical role of French leadership in organizing and conducting the actual operation. The fact that the DRC possesses a high degree of mineral wealth, where EUheadquartered multinational firms operate, also cannot be discounted in explaining EU support in general, and French leadership in particular, for Artemis. The overall impression in this case, then, is that Artemis had much less to do with ‘pre-existing’ EU procedural and substantive norms than it did with other factors, including material interests, so the entrapment hypothesis seems somewhat overstated here. Regarding the Lewis study (ch.5) of the EU’s divisions over Iraq, we see theoretical claims regarding two sub-cases: initial disagreement over the war itself, and subsequent agreement over the provision of EU post-invasion reconstruction aid to Iraq. According to Lewis, a preexisting norm was absent in the first case, leading to LCD bargaining and deadlock, but present in the second case (in the form of a ‘humanitarian aid’ norm), which permitted some degree of entrapment and cooperative bargaining to influence the outcome. The most striking aspect of this case is indeed the extent to which certain EU member

226 Norms, Institutions and EU Foreign Policy

states (as well as some prospective member states) were so willing to publicly violate EU norms regarding confidentiality and consultation by publishing their opinions in the media; these norms are the bedrock on which all EFP decisions must be made. Lewis uses the terms ‘deactivated’ and ‘activated’ to describe whether EU member states were acting under the assumption of EFP procedural norms, as if such norms can be turned off and then on again depending on the circumstances. In my view, however, the simple fact that some EU member states can ‘get away with’ violating them so brazenly in this case strongly undermines the overall ‘entrapment’ hypothesis as an explanation of negotiated outcomes. Equally, the lack of a clear moment or decision point when the procedural norms were ‘activated’ again (to enable the provision of post-invasion humanitarian aid) also calls into question the idea that EU member states suddenly felt like they had no option but to agree to such a policy. Like the Artemis case, the number of arguments in favour of an EU aid policy on Iraq – building the EU as a soft power, repairing transatlantic relations, supporting the UN, etc. – makes it difficult if not impossible to point to a distinct collective normative consensus that led reluctant EU member states into supporting the policy through the specific processes of entrapment and cooperative bargaining. The two chapters on economic issues – those by Elgström on EPAs (ch.8) and Vogler on climate change (ch.9) – offer other insights but also involve methodological challenges. According to Table 10.2 in the Conclusion (Schimmelfennig and Thomas, ch.10), the EPAs were not negotiated under the shadow of a single, dominant pre-existing norms (the ‘determinacy’ condition); nor did they fall under some kind of policy precedent to help facilitate a normative consensus (the ‘precedent’ condition). Under these conditions, the outcome in both subcases (the negotiation mandate and final offer) was ‘weak entrapment’. Yet the actual narrative of Elgström’s chapter seems to suggest otherwise: there was a clear pro-free trade aspect to the talks, as institutionalized by the Commission’s role, and there was a long legacy of prior policies that attempted to balance trade and development aid goals in the EU’s relations with less developed countries (the Lomé/Cotonou frameworks). In fact, of all issues covered in the volume I would argue that the normative/policy consensus was highest on the EPA sub-cases given the historical legacy of EU relations with less developed countries and the highly institutionalized role of the Commission, so in my reading, the EPA ‘final offer’ sub-case represents an attempt by certain EU member states to hijack the talks for their own purposes after a clear

Michael E. Smith 227

mandate had been agreed; further, these states had multiple avenues to assert their preferences (while tapping into the EU’s large repertoire of norm-based arguments), even including the threat of a veto, and they were able to force a change in the final offer that reflected their concerns. Finally, and given this outcome, I don’t understand what is meant by ‘weak entrapment’ in these two sub-cases, a term applied (though not defined) in the Concluding chapter but not by Elgström himself. Does ‘weak’ mean that only certain aspects of the decision produce ‘entrapment’ effects, or does it mean that only certain EU member states were entrapped (or both)? Or does it mean that certain EU member states had discretion over only part of the negotiation process? Vogler’s study of climate change (ch.9) examines two sub-cases: the Berlin mandate and the post-2012 climate regime. According to the volume’s Conclusion (ch.10, Table 10.2), both climate change cases reflect a lack of normative consensus (no ‘determinacy’) but do follow on from earlier policy precedents. Under these conditions, the outcome in both cases was a combination of ‘weak entrapment’ and cooperative bargaining. However, it is difficult to analyse the actual stakes of EU’s talks on the post-2012 regime with any confidence since the EU’s final position would be contingent on those of other leading players (as ultimately reflected by the negative outcome of the December 2009 Copenhagen talks on climate change); moreover, the distinctions between ‘competitive’ (i.e., LCD) and ‘cooperative’ bargaining are not very clear. In this case, the combined outcome of ‘weak entrapment’ and ‘cooperative bargaining’ can easily be open to multiple interpretations regarding the role of norms and/or policy precedents. Finally, as with the EPA cases, there is a disconnect here between Vogler’s narrative – where he notes the role of both competitive and cooperative bargaining, as well as some degree of normative suasion – and Schimmelfennig and Thomas’s concluding essay (ch.10), which codes this as a case of ‘weak’ entrapment and cooperative bargaining. These conflicting interpretations, by contributors to the same volume, clearly indicate the difficulties involved in defining and measuring, in an a priori fashion, which norms might exert causal effects on various EU member states regarding a certain policy problem, especially one as complex and drawn-out as the global climate change talks. Two chapters examine what I would call ‘value projection’ goals: EU policies toward Ukraine (Youngs, ch.3) and EU support for the ICC (Thomas, ch.4). In the two Ukraine sub-cases (support for democratization and EU accession, respectively), both involved ‘determinacy’ effects but only the democratization issue enjoyed a clear

228 Norms, Institutions and EU Foreign Policy

policy precedent. It also benefited from positive ‘relevance’, ‘forum’, and ‘publicity’ effects, making it a textbook example of entrapment effects (Schimmelfennig and Thomas, ch.10, Table 10.2). However, based on the narrative in the democratization case study, the EU’s common policy – which clearly took some time to coalesce after much hesitation – was quite contentious and seemed to involve a high degree of bargaining rather than simple entrapment effects (it is coded by the editors as entrapment only, rather than entrapment plus cooperative bargaining). Numerous exogenous factors, such the views of Russia and other non-EU actors, seemed to have played a major role as well. Conversely, the accession case resulted in ‘disentrapment’ and then LCD bargaining (i.e., competitive, not cooperative, bargaining). Again, it is not clear to me how ‘disentrapment’ can be measured against ‘weak entrapment’; further, there is a major normative difference between supporting pro-democracy movements and supporting accession to the EU (as with the Turkish case below), so it may be inappropriate to code the accession case as one operating under a clear normative frame. The ICC chapter involves two sub-cases: immunity for UN peacekeepers and bilateral immunity deals with the US. The former lacked ‘determinacy’ and ‘precedent’ conditions and led to ‘disentrapment’ and then LCD bargaining (like the Ukraine accession case), while the latter enjoyed ‘determinacy’ but not ‘precedent’ effects and resulted in entrapment and cooperative bargaining effects. However, the critical variable in bilateral immunity case seems to be the ‘forum’ effects rather than pre-existing norms or policy precedents: the issue of bilateral agreements fell under the scope of the CFSP (rather than the UN). The peacekeeping immunity case involved additional conflicts as two important EU member states (France and the UK) are permanent members of the UN Security Council; the UK in particular also had to play a mediating role (between the US and the EU) in its role as president of the Security Council. All EU member states understood that failing to find a common position on the issue would make them highly vulnerable to US ‘divide and conquer’ tactics, while the fact that this issue also fell under the shadow of prior EU disagreement over the US-led Iraq War also cannot be discounted. In other words, exogenous factors seems to be equally if not more important here than endogenous (i.e., normative) ones. Finally, Schimmelfennig’s study (ch.7) of Turkish accession to the EU has been divided into three sub-cases (accession negotiations in 1997, 2004, and 2006) spanning nearly ten years of intra-EU debate. All three sub-cases involve initial ‘determinacy’ conditions (i.e., Turkey clearly

Michael E. Smith 229

has a right to apply for membership) but result in three different outcomes (LCD bargaining, entrapment, and disentrapment respectively) depending on other factors at work. The overall narrative argues that EU member states were unbound in their positions on Turkey in 1997, then ‘entrapped’ into re-opening talks in 2004, then ‘disentrapped’ in 2006. The primary reason for the critical change between 2004 and 2006 was Turkey’s behaviour towards Cyprus, which gave the EU a convenient excuse to shut down the talks. Although this analysis seems more straight-forward than some of the outcomes discussed above, there are still some questions. First, given the very long time period involved in Turkey’s membership bid, it is somewhat problematic to choose three ‘decision points’ as core barometers of the entire process, especially when the primary outcome at stake – Turkey’s membership of the EU – is still highly contested. Second, and following from that point, it is questionable whether EU member states who opposed Turkish membership really felt so ‘entrapped’ in 2004 when they made a credible threat to put Turkish membership to a national referendum if the accession talks ultimately produce an offer of membership – which of course hasn’t happened yet. Third, the main reason that talks broke down in 2006 was Turkey’s refusal to extend customs union rights to Cyprus; this requirement, however, could be viewed as a kind of ‘poison pill’ introduced in the previous phase of talks (2004/05) because opponents of Turkish membership clearly knew that the Erdogan government would find it difficult if not impossible to fulfil the condition. In other words, proponents of Turkish membership might have won the occasional battle (as in 2004/05) but they are clearly losing the war, and Turkish membership of the EU will by no means be decided by Turkey’s prima facie compliance with various EU norms and/or by the ‘entrapment’ of EU member states who still oppose such membership. In sum, we can see the presence of multiple causal factors at work – and not just normative ones – during the various case studies under consideration; some of these factors are exogenous to the EU while others are more endogenous. Beyond this problem, an equally important concern is that there is no direct evidence that EU member states really felt so ‘normatively entrapped’ to the degree argued by some of the cases in this volume. Instead, normative causality is one-step removed in terms of evidence: the ‘missing link’ in the entrapmentconfirming cases (on Ukraine, the ICC, reconstruction aid to Iraq, Operation Artemis, and the 2004 Turkish accession talks) is clear evidence that officials from EU member states felt or perceived or calculated some real cost if they did not conform to the decision-consensus at

230 Norms, Institutions and EU Foreign Policy

hand. Instead, the analysis skips this crucial evidence and focuses on whether certain policies were agreed in the absence of an a priori consensus of preferences. If such policies were agreed, then the contributors merely assume or infer that various officials from EU member states (or EU institutions, such as the Commission) felt ‘entrapped’ enough to go along with the prevailing discursive position at the time. As Jeffrey Lewis (ch.5) puts it, there is never any ‘smoking gun’ evidence whereby EU officials admit they were ‘entrapped’ into following a common EU policy, even though several contributors did manage to interview EU officials during the course of their investigations. In other words, there is too much confusion about the true dependent variable here: it is not the final agreed policy, but rather the negotiating space as experienced by various actors. Since the volume focuses on the former as a proxy for measuring the latter, we cannot be truly confident yet about the actual degree of causality here (i.e., the presence of pre-existing norms/policies causes a restriction of the negotiating space as measured by various preference-outliers or veto players, which then causes them to agree on common EU policies they might not otherwise prefer).

Future directions: Norms, rules, and policies The discussion above should not lead one to believe that norms are unimportant in analysing EFP outcomes. But there is a distinct difference in using norms to describe or define such outcomes (in the form of a typology, perhaps) versus using norms as direct causal variables to explain (or predict) such outcomes. Moreover, in treating norms as causes there is also a difference between analysing normative effects over a long time period (as in the form of socialization or learning effects) versus their use in an instrumental fashion as a source of negotiating leverage. As it is difficult to measure the clarity and salience of norms separately from their use or interpretation during specific decision-making settings, there is a clear danger of tautology here. To overcome these problems, future work on the role of norms in EFP negotiations/decisions must address the following problems, at a minimum. First, we need a clearer and more robust typology of EFP norms, in both procedural and substantive terms. I have focused on procedural norms in much of my work (see especially Smith, 2001; and Smith, 2003), arguing that norms in general can be measured as independent variables in terms of their clarity, formality/legality, and bindingness; a

Michael E. Smith 231

dichotomous measure for ‘salience’ (or ‘relevance’) can then be added to determine whether they also apply to a specific foreign policy problem (i.e., a transition from a general norm to a specific rule). Now the debate over norms is paying more attention to substantive EFP norms given the much larger body of EFP policy outputs as compared to the 1970s–80s. Yet even these substantive norms can involve material versus non-material (i.e., aspirational or constitutive) goals, as well as varying degrees of clarity/legality/bindingness in terms of whether they are viewed as norms or rules (or even EU laws). Second, and in a similar fashion, there needs to be some more systematic attention to what EFP ‘outputs’ should ‘qualify’ as EU foreign policies, at least for the purpose of rigorous theoretical analysis. As a starting point, I would suggest that analysts focus on outputs that require more than a rhetorical commitment on the part of the EU – that is, on actions (especially those required material resources) rather than declarations/conclusions. Similarly, these EFP outputs, once defined, can be classified under at least four major issue-areas as noted above: security, economic prosperity, value projection, and accession/enlargement (or ‘constitutive’ goals). At a minimum, this exercise would go a long way towards showing the true coherence/consistency between EU foreign policy claims and actions, and hopefully take us past the sterile ‘normative power Europe’ versus ‘strategic/rationalist Europe’ debate conducted over the past decade (though not in this volume). Third, once we have a clearer picture of more normatively-inspired outcomes as compared to more materialist ones, we can take a closer look at active normative causality (as opposed to passive correlations between normative goals and policy outputs), both over time and during specific negotiating settings. As this volume is concerned primarily with the latter, we have to find more persuasive evidence that various EU actors changed their positions on certain EFP issues because certain norms, whether procedural or substantive (or both) were clearly in effect (or ‘activated’). Personally, I believe the ‘entrapment’ hypothesis is somewhat overstated in this volume (and difficult to distinguish from ‘weak entrapment’ and ‘cooperative bargaining’) when applied to single policy decisions or outcomes. Moreover, a clear a priori normative consensus is neither necessary nor sufficient for the EU to take foreign policy actions. And since there are no clear punishments (reputational or otherwise) for violating most EFP norms, other factors must be at work in explaining the wide range of outcomes examined in the case studies herein. Finally, and given the secretive nature of many EFP negotiations, it seems that norms can easily be invoked after a decision has

232 Norms, Institutions and EU Foreign Policy

been taken for other reasons to justify the appropriateness of the policy after the fact. Fourth, given the highly contingent nature of normative effects, such effects can be appreciated or analysed only in the context of other factors discussed (but not prioritized analytically) in this volume. These would include exogenous variables such as the influence of major actors like the UN, the US, and Russia, as well as endogenous variables such as leadership by certain EU member states, changes of government in EU member states, and feedback/demonstration effects from similar policies. Indeed, in my reading the various cases in this volume confirm a basic point about the making of EFP: it is too complex in terms of issues, policy outputs, actors, and institutions to reduce to one or two key variables. If it were possible to quantify these factors then the obvious solution would be some form of multiple regression to measure the relationships between norms and other causal variables. However, since we are dealing with qualitative data we must adopt a more deductive approach and infer relationships by using other sources of evidence, such as documents but also interviews with insiders. Moreover, we must consider that different actors in the EFP system – especially officials from EU member states – might be motivated to adopt common policies for different reasons. Fifth and finally, even while contextualizing the role of norms in the manner suggested above, we can still ask whether the EU really represents a new kind of global political actor, in terms of its values and how they are defended/projected. For example, one longstanding and very crucial motivation behind the development of a common EU foreign policy is a desire to pursue policy positions that differ from the US; this goal is confirmed in several of the empirical chapters in this volume. Yet is this view a result of a simple transatlantic conflict of interests, or does it mean that EU officials really believe that their approach to global problems (which stresses institutionalized, multilateral, ‘soft power’ solutions) might be more effective than the more unilateral and ‘hard power’-oriented approach of the US? In other words, should we analyse EU norms in more narrow/ rhetorical terms (i.e., the EU merely wants to live up to its stated goals, the view adopted in this volume), or in broader and more cosmopolitan terms (i.e., the EU believes it has a real capacity for unique and principled global leadership to confront modern policy problems)? I believe a real debate over the latter role has been materializing in terms of how the EU feels distinct from the US (and NATO) in general, and how the EU applies its power in particular; moreover, interviews with various EU insiders across its main institutions confirm this view (Smith, forthcoming).

Michael E. Smith 233

This question has become even more salient in light of the EU’s willingness to use force or to capture/punish those who violate its interests (as in the EU’s Operation Atalanta against piracy; see Germond and Smith, 2009). In fact, the EU is losing its reputation as a wholly civilian power, yet it is unclear how this change will impact the EU’s normative justification of its ambitions, which are as great as any other major power. Unlike during the 1980s–90s, the EU’s debate over the use of force to protect its interests is no longer a theoretical one, so normative justifications (if not actual normative causes) will become increasingly important as the EU expands this capability, especially now that it is implementing the long-overdue measures of the Lisbon Treaty (such as the European External Action Service). The possibility of a new principled leadership role for the EU in security affairs, as with its already existing economic roles, could also help demonstrate a real social value-added at work in the making of EU foreign policy, and further undermine the more atomistic view of EFP outcomes as argued by the realist/intergovernmentalist position.

Works cited Botcheva, L., and Martin, L. (2001) ‘Institutional Effects on State Behavior: Convergence and Divergence’, International Studies Quarterly, 45(1): 1–26. Germond, B. and Smith, M.E. (2009) ‘Re-thinking European Security Interests and the ESDP: Explaining the EU’s Anti-Piracy Operation’, Contemporary Security Policy, 30(3): 573–93. Hyde-Price, A. (2006) ‘“Normative Power Europe”: A Realist Critique’, Journal of European Public Policy, 13(2): 217–34. Linklater, A. (2005) ‘A European Civilising Process?’, in C. Hill and M. Smith (eds) International Relations and the European Union (Oxford: Oxford University Press). Manners, I. (2002) ‘Normative Power Europe: A Contradiction in Terms?’, Journal of Common Market Studies, 40(2): 234–58. Smith, M.E. (2001) ‘Diplomacy by Decree: The Legalization of EU Foreign Policy’, Journal of Common Market Studies, 39(1): 79–104. Smith, M.E. (2003) Europe’s Foreign and Security Policy: The Institutionalization of Cooperation. Cambridge: Cambridge University Press. Smith, M.E. (2008) ‘Researching European Foreign Policy: Some Fundamentals’, Politics, 28(3): 177–87. Smith, M.E. (forthcoming 2011) ‘A Liberal Grand Strategy in a Realist World? Power, Purpose, and the EU’s Changing Global Role’, Journal of European Public Policy. Youngs, R. (2004) ‘Normative Dynamics and Strategic Interests in the EU’s External Identity’, Journal of Common Market Studies, 42(2): 415–36.

Index Note: The notation ‘f’ indicates continued discussion on a following page; ‘ff’ indicates continued discussion on two following pages. Names that also appear within a chapter’s ‘Works Cited’ are not included in the Index. Actorness (of EU) 8, 14, 70, 80, 87, 138, 151, 162, 187f, 192, 201 Adamkus, Valdas 36 Africa 92–110, 131, 133, 134, 135, 137, 141, 204, 205, 211, 213 See also under names of African states African, Caribbean and Pacific countries (ACP) 131–48 Albania 75 Alliot-Marie, Michèle 101 Ankara Agreement 123, 125 American Service-members Protection Act (ASPA) 53, 54 Amnesty International 61, 99 Angola 97 Annan, Kofi 93, 95, 98, 99, 100, 101, 104, 124 Appropriateness 14, 23, 31, 38, 76, 77, 92, 107, 129, 194, 196, 200, 210, 220, 232 Arab states 43, 44 Ashton, Catherine 87, 192 Asia 46, 111 See also under names of Asian states Association agreement (with the EU) 41, 44, 113, 118, 119, 120, 121, 122, 123, 125 Australia 162, 165 Austria 36, 46, 60, 100, 112, 115, 116, 117, 126, 128, 153, 154, 155, 159, 161, 165 Bakenende, Jan Peter 77 Bargaining 10, 11, 21, 48, 132, 147, 160, 169, 171, 207, 208, 212

Competitive bargaining 7, 8, 20–2, 25, 31, 39, 47, 50, 51, 52, 61, 65, 93, 105, 106, 113, 132, 145, 146, 151, 159, 168, 179, 185, 186, 189, 205, 208, 210, 219, 222, 223, 225, 227, 228, 229 Cooperative bargaining 6, 18–20, 25, 31, 40, 47, 51, 66, 72, 77, 79, 83–4, 85, 92, 104, 105, 122, 159, 177–90, 194, 203, 204, 207–9, 213, 219, 220, 222, 223, 225, 226, 227, 228, 231 Integrative bargaining 160, 171 Barroso, José Manuel 43, 44, 89 Berlusconi, Sylvio 75, 156, 166, 167 Blair, Tony 56, 75, 76, 104, 155, 161, 165 Borrell, Josep 41 Bot, Bernard 77, 88 Brazil 100, 170 Bulgaria 75, 120, 121, 154, 165, 168, 171 Bush, George W. 53, 54, 56, 74, 103, 155, 162, 199 Canada 100, 108, 162 Central African Republic 101 Chad 107 China 3, 167, 170, 205 Chirac, Jacques 37, 73, 74, 75, 76, 100, 101, 116, 117, 124, 125, 129, 170 Christian Democrat parties 77, 115, 116, 121, 125 234

Index 235 Clinton, Bill 53 Coalition for the International Criminal Court 65 Cohesion countries (of EU) 156, 160, 161, 164, 169 Comité Juridique (‘Cojur’) 61, 62, 63, 65, 66 Committee of Permanent Representatives (‘Coreper’) 23, 63, 74, 77, 92, 152 Congo, Democratic Republic of 7, 92–110, 178, 181, 183, 203, 206, 211, 212, 213, 225 Congress (US) 53, 54 Consensus norm 6, 7, 8, 9, 15, 18, 19, 56, 70, 71, 77, 78, 79, 83, 84, 85, 86, 89, 132, 138, 139, 144, 145, 146, 147, 177, 194, 197, 204, 207, 208, 213, 214, 222 Constitutional Treaty (of the EU, proposed) 117, 125 Consultation norm 6, 15, 18, 19, 72, 78, 105, 107, 177, 194, 197, 199, 207, 209, 222, 226 Conflict prevention 15, 81, 82, 83, 95, 211 Constructivism 5, 7, 23, 25, 40, 197, 219 Cooperative bargaining see under Bargaining Copenhagen see Copenhagen summits, Copenhagen criteria, Denmark Copenhagen criteria 118, 120, 122, 123, 124, 198 Copenhagen summits 95, 123, 150, 151, 168, 170, 227 COPS see Political and Security Committee Coreper see Committee of Permanent Representatives Coreu 59 Cotonou Partnership Agreement 134, 135, 139, 142, 144, 226 Council of Europe 17, 62 Croatia 75, 120

Cyprus 112, 114, 115, 116, 117, 122, 124, 125, 126, 127, 154, 165, 184, 229 Darfur 206, 211, 212 de Hoop Scheffer, Jaap 77 de Villepin, Dominique 101, 126 Deadlock 7, 21, 22, 50, 63, 113, 178,179, 181, 183, 219, 225 Democracy 7, 14, 34, 35, 36, 37, 38, 40, 41, 43, 44, 46, 53, 72, 80, 81, 85, 86, 88, 95, 112, 113, 118, 120, 123, 124, 125, 176, 182, 184, 185, 194, 197, 198, 199, 208, 211, 219, 228 Denmark 56, 60, 62, 63, 64, 67, 89, 104, 108, 115, 117, 132, 136, 139, 141, 143, 154, 155, 159, 161, 165, 169 Determinacy 16, 180, 181, 182, 183, 185, 186, 187, 196, 197, 198, 219, 220, 223, 225, 226, 227, 228 Development 3, 8, 10, 12, 15, 41, 78, 131–148, 153, 156, 182, 184, 187, 226 See also European Commission Directorate General for Development Dimas, Stavros 166 Dini, Lamberto 116 Disentrapment 17, 178, 179, 181, 182, 186, 196, 197, 220, 228, 229 Economist 167, 190 ECJ see European Court of Justice Economic Partnership Agreements (EPAs) 131–49, 178, 181, 184, 187, 189, 223, 226, 227 emissions trading 8, 150, 152, 162, 165, 171 Enlargement 5, 10, 12, 20, 31–49, 96, 111–13, 164, 184, 231 See also European Commission Directorate General for Enlargement

236 Index Energy policy 7, 39, 44, 45, 46, 150–73 See also European Commission Directorate General for Energy Entrapment see normative entrapment, disentrapment Environment 8, 10, 12, 15, 150–73, 185 Erdogan, Recep Tayyip 124, 125, 228 Estonia 75, 154, 165, 205 European Commission 5, 8, 11, 33, 35, 39, 41, 43, 45, 53, 54, 56, 58, 59, 60, 61, 67, 78, 81, 84, 112, 113, 114, 117, 119, 120, 121, 123, 124, 125, 126, 127, 128, 129, 131, 132, 134, 136, 137, 138, 139, 140, 141, 142, 143, 145, 146, 147, 151, 152, 153, 155, 156, 164, 167, 168, 169, 171, 184, 187, 190, 197, 226, 230 Directorate General for Development 138, 140, 142, 143 Directorate General for Energy 166 Directorate General for Enlargement 58, 121, 129 Directorate General for Environment 151, 152, 157, 161, 166, 169, 170 Directorate General for External Relations (Relex) viii, 53, 81, 204 Directorate General for Trade 131, 135, 136, 138, 139, 140, 142, 143, 144, 145, 147 Legal Service 58, 66 European Council 3, 33, 42, 71, 78, 79, 88, 94, 95, 96, 112, 113, 117, 120, 121, 122, 123, 125, 152, 162, 163, 166f, 168 European Court of Justice 11, 164 European People’s Party 115 European External Action Service (EEAS) 3, 192, 233 European Neighbourhood Policy (ENP) 34, 35, 41, 42, 44

European Parliament (EP) 41, 54, 61, 62, 66, 74, 111, 115, 126, 152, 184 European Political Cooperation (EPC) 75, 105 European Security Strategy 72, 79, 84, 94, 95, 96, 107, 210, 225 Europeanization 12, 18, 40 Fillon, Francois 46 Financial Times 60, 75, 142 Finland 43, 60, 127, 154, 155, 158, 159, 165 Fischer, Joshka 37, 63, 64, 65, 103 Framing 16, 60, 62, 66, 70, 72, 73, 79, 80, 82, 83, 93, 94, 95, 100, 102, 105, 106, 113, 115, 132, 135, 136, 137, 138, 140, 141, 142, 145, 147, 148, 160, 168, 169, 197, 208, 221 France 4, 32, 35, 37, 39, 41, 42, 43, 44, 45, 46, 47, 55, 56, 57, 60, 66, 70, 71, 73, 75, 76, 78, 81, 83, 85, 87, 93, 94, 95, 96, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 112, 115, 116, 117, 124, 125, 126, 132, 136, 137, 141, 144, 146, 153, 154, 155, 159, 165, 167, 168, 170, 184, 199, 204, 205, 206, 211, 212, 213, 225, 228 GATT see General Agreement on Tariffs and Trade Gazprom 39 GDP 3, 111, 129, 154, 156, 164, 167, 168 G8 142, 154, 155, 165 General Affairs and External Relations Council (GAERC) 62, 63, 64, 73, 79, 92, 103 General Agreement on Tariffs and Trade (GATT) 133 Genocide 52, 56, 100, 212 Germany 32, 35, 37, 39, 41, 42, 43, 44, 45, 56, 58, 60, 62, 63, 64, 65, 66, 70, 71, 73, 76, 77, 78, 81, 83, 85, 96, 100, 102, 103, 104, 108,

Index 237 111, 115, 116, 117, 122, 124, 125, 127, 136, 137, 144, 146, 153, 154, 157, 158, 159, 160, 161, 165, 166, 167, 170, 197, 198, 199, 204, 211, 213 Greece 60, 72, 75, 78, 83, 88, 103, 115, 116, 117, 122, 126, 154, 156, 158, 159, 165, 197, 198 Green parties 116, 117, 125, 153, 170 Greenhouse gases (ghg) 150–73 Hallstein, Walter 119 Helsinki Final Act 94 High Representative for Foreign Affairs and Security Policy 3, 72, 74, 84, 87, 99, 192 House of Lords (UK) 64 House of Representatives (US) 54 Human rights 15, 50, 53, 56, 60, 80, 85, 95, 116, 118, 119, 123, 124, 133, 134, 194, 197, 211, 219 Humanitarian assistance 7, 10, 71, 78, 79, 81, 225, 226 Humanitarian intervention 3, 92–110, 183, 206, 210 Hungary 41, 87, 100, 154, 165, 166 Ideas (policy-making role of) 16, 22, 23, 46, 137, 221, 223 See also Consensus norm, Consultation norm, Identity Identity (of actors) 23, 40, 111, 112, 118, 133, 151, 162, 194, 197, 198 India 170 Institutionalism Discursive Institutionalism 193 Normative Institutionalism 4–7, 10–28, 38, 47f, 50ff, 64ff, 71f, 92, 107, 113, 128, 132, 139, 145, 147f, 177–233 Rational Institutionalism 14, 193–200, 213 Sociological Institutionalism 14, 193–200 Intergovernmentalism 5, 7, 20ff, 25, 50, 65f, 122, 179, 189f, 207, 233

International Criminal Court (ICC) 7, 50–69, 178, 179, 180, 181, 183, 187, 199f, 208, 220, 227, 228, 229 International Crisis Group 100, 210f International law 52, 54, 58, 59, 60, 61, 64, 126 Iraq 3, 7, 45, 58, 61, 70–91, 93, 96, 97, 99, 101, 103, 104, 105, 106, 107, 178, 179, 181, 183, 184, 185, 190f, 195, 198f, 203, 307, 208, 211, 222, 225f, 228, 229 Ireland 55f, 57, 60, 66, 100, 105, 115, 141, 154, 156, 158, 159, 165, 171 Islam 111, 115, 116, 198 Israel 59 Issue-linkage 166, 187, 189f Italy 32, 35, 60, 61, 64, 65, 66, 100, 115, 116, 117, 154, 155, 156, 159, 163, 165, 166, 167, 169, 200 Kaul, Hans-Peter 63 Kissinger, Henry 4 Klaus, Vaclav 164 Kosovo 3, 104, 191 Kuchma, Leonid 32, 33, 34, 35, 36, 37, 38, 39, 41, 43 Kwasniewski, Aleksander 36 Kyoto Protocol 8, 150–66, 169, 170 Labour parties and members 141, 155 Latvia 75, 154, 165, 205 LCD see Lowest common denominator Less Developed Countries (LDCs) 131, 132, 133, 135, 136, 139 Le Touquet summit 95, 104 Learning 7, 22f, 62, 107, 113, 115, 118, 122, 132, 146, 147, 171, 180, 201, 213, 219, 222, 230 Lindh, Anne 61 Lisbon Treaty 2, 23, 25, 233 Lithuania 34, 35, 36, 41, 43, 75, 154, 165, 205 Lomé conventions 133, 136, 137, 226

238 Index Lowest common denominator (LCD) 6, 7, 8, 21, 22, 24, 25, 31, 39, 47, 50, 51, 52, 57, 61, 66, 70, 71, 72, 74, 76, 78, 85, 93, 106, 113, 122, 137, 145, 146, 160, 178, 179, 181, 183, 184, 185, 189, 195, 205, 219, 222, 227, 228, 229 Luxembourg 60, 71, 76, 83, 85, 96, 105, 158, 163, 165 Maastricht Treaty see Treaty on European Union Macedonia, former Yugoslav Republic of (fYRoM) 75, 95 Malta 154, 165 Mandelson, Peter 135, 140, 142 McCreevy, Charlie 166 Meciar, Vladimir 35 Médecins Sans Frontières 211 Mediterranean 32, 43, 116, 121, 136, 137 Merkel, Angela 42, 117, 125, 127, 153 Michel, Louis 76, 142 Middle East 72, 79, 80, 83, 85 Mission des Nations Unies en République Démocratique du Congo (MONUC) 98, 100, 101 Moller, Per Stig 61 Moratinos, Miguel Angel 128 Multilateralism 15, 50, 55, 78, 79, 80, 82, 86, 100f, 102, 183, 232 Muslim beliefs or identity see Islam Namibia 97 NATO see North Atlantic Treaty Organization Netherlands see The Netherlands New Zealand 34 NGO see non-governmental organisation Non-governmental organisations 16, 37, 52, 53, 56, 60, 61, 65, 66, 132, 134, 136, 140, 141, 143, 183, 185, 187 Normative entrapment 6, 7, 8, 15–18, 25, 31, 38, 39, 40, 45, 47,

51, 66, 72, 79, 80–3, 92, 93, 104f, 107, 108, 111, 113, 114, 119f, 122, 126, 127, 128, 129, 138, 139, 144, 145, 147, 151, 161, 162, 168, 169, 174, 177, 178, 179, 180–7, 188f, 194–8, 199, 200, 203, 209–13, 219–31 See also Disentrapment Normative Institutionalism see Institutionalist theories North Atlantic Treaty Organization 44, 46, 60, 77, 84, 95, 96, 97, 101, 102, 103, 105, 122, 184, 205, 225, 232 Northern Cyprus see Cyprus Obama, Barack 4 Oil for Food Programme (UN) 71, 72, 79, 80 133 Committee 23, 138, 143, 145 Operation Artemis 92–110, 178, 181, 183, 184, 203, 206, 210, 211, 212, 213, 225 Operation Atalanta 233 Operation Concordia 95, 96 Operation Turquoise 101 Orange Revolution 7, 31–49, 182, 199, 205, 208, 212, 213 Organization for Security and Cooperation in Europe (OSCE) 36, 37 Oxfam 141, 143 Pangalos, Theodore 116 Papandreou, George 116 Papantoniou, Yiannos 103 Paris Charter 94 Peacekeeping 53, 55ff, 95, 97, 101, 178, 179, 181, 183, 187, 200, 211, 220, 225, 228 Piebalgs, Andris 166 Plassnik, Ursula 128 Poland 34, 35, 36, 37, 41, 43, 44, 46, 87, 115, 117, 132, 154, 165, 166, 168, 171, 205 Political and Security Committee (PSC) 58, 62, 63, 81 Portugal 60, 65, 87, 100, 136, 137, 154, 156, 158, 159, 160, 161, 165

Index 239 Powell, Colin 59 Precedent 17, 58, 128, 180–9, 196, 199, 220, 223, 224, 226, 227, 228 Presidency (of EU Council of Ministers) 12, 23, 54, 56, 60, 62, 63, 75, 78, 79, 80, 103, 127f, 152, 155, 157, 158, 159f, 163, 165f, 168, 171, 183, 192, 205 Prodi, Romano 34, 58, 116 PSC see Political and Security Committee Publicity 17, 20, 180–7, 212, 220, 228 Putin, Vladimir 37, 45 Qualified majority voting (QMV) 11, 200 See also Voting rules Rasmussen, Fogh 117 Realist theory 106, 146, 200, 217, 219, 233 Reciprocity 18, 21, 133, 134, 195, 210 Romania 44, 46, 48, 59, 60, 75, 120, 121, 154, 165, 168, 171 Rule of law 15, 50, 71, 72, 80, 81, 85, 86, 118, 123, 197 Rumsfeld, Donald 57, 74, 96 Russia 32, 34, 35, 36, 37, 38, 39, 42, 43, 44, 45, 46, 103, 163, 169, 199, 205, 228, 232 Rwanda 97, 98, 99, 101, 102 Scheffer, David 57 Schröder, Gerhard 33, 39, 73, 103, 122, 124 Senate (US) 53, 54 Security Council see United Nations Serbia 38 Slovakia 35, 37, 41, 75, 88, 154, 165, 166 Slovenia 75, 88, 154, 165 Simitis, Costas 75 Social Democrat parties 116, 125 Socialization 5, 10, 24, 75, 161, 194, 195, 196, 200, 207, 230

Solana, Javier 36, 39, 72, 74, 75, 79, 84, 99 South Africa 100, 108, 170 Southern Africa Development Community (SADC) 131, 134 Spain 32, 35, 37, 41, 42, 43, 45, 60, 65, 74, 76, 87, 100, 115, 117, 128, 136, 137, 154, 156, 161, 163, 165, 166, 198 St. Malo summit 94, 95, 101, 104, 105, 204 Steinmeier, Frank-Walter 42 Stubb, Alexander 4 Sweden 35, 37, 60, 61, 100, 103, 108, 117, 132, 136, 139 154, 155, 159, 163, 165, 166 Suasion 7, 10, 13, 23f, 25, 51, 62, 107, 113, 115, 118, 122, 132, 142, 143, 146, 147, 180, 188, 201, 219, 222, 227 Tajikistan 59 Tariffs 144, 146 See also General Agreement on Tariffs and Trade; Trade Tbilisi 205 Thatcher, Margaret 214 The Netherlands 37, 41, 42, 47, 54, 60, 75, 76, 77, 88, 100, 102, 115, 117, 125, 132, 136, 141, 153, 154, 155, 157, 158, 159, 160, 161, 165, 166, 171 Third World see Less Developed Countries Timor Leste 55, 59 Timoshenko, Julia 32 Trade 5, 8, 10, 12, 15, 41, 43, 44, 131–49, 187, 226 See also European Commission Directorate General for Trade; General Agreement on Tariffs and Trade; Tariffs Treaty on European Union 4, 11, 92, 94, 105, 108, 118, 210 See also Lisbon Treaty Turkey 8, 38, 111–30, 178, 179, 181, 184, 185, 198, 224, 228, 229

240 Index Uganda 97, 98, 99, 100, 102, 108 Ukraine 7, 31–49, 169, 178, 179, 181, 182, 185, 199, 203, 205, 208, 210, 213, 220, 225, 227, 228, 229 Uncertainty 22, 24, 45, 77, 146 United Kingdom (UK) 34, 35, 39, 42, 55, 56, 57, 60, 61, 62, 64, 65, 66, 94, 95, 96, 100, 102, 103ff, 108, 115, 116, 117, 121, 132, 136, 137, 139, 141, 142, 144, 154, 155, 158, 159, 160f, 165, 166, 200, 205f, 208, 228 United Nations (UN) 3, 7, 12, 50, 52, 53, 54, 55–7, 66, 70f, 72, 73, 74, 75, 76, 78, 79, 80, 81, 84, 85, 86, 88, 89, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 104, 105, 106, 108, 124, 150, 152, 153, 155, 162, 163, 164, 170, 178, 181, 183, 187, 199, 200, 220, 225, 226, 228, 232 United States (US) 3, 37, 38, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 70, 72, 74, 75, 81, 82, 83, 84, 95, 96, 97, 103, 105, 117, 133, 150, 151,

155, 156, 162, 165, 166, 170, 171, 182f, 190, 193, 199f, 205, 213, 221, 228, 232 Uruguay 98 Van Rompuy, Herman 4 Vedrine, Hubert 81 Verheugen, Günter 121, 166 Veto 5, 6, 10, 11, 13, 16, 18, 19, 20, 21, 22, 24, 55, 61, 65, 66, 71, 81, 85, 106, 112, 122, 137, 145f, 167f, 169, 177, 179, 190, 200, 228, 224, 227, 230 See also Voting rules Voting rules 5, 11, 21, 102, 126, 128, 138, 200 Wall Street Journal 74, 76 Wallstrom, Margot 162 World Trade Organization (WTO) 134, 135, 142, 147 Yanukovich, Viktor 32, 36, 27, 38, 39, 42, 43, 44, 45 Yushchenko, Viktor 32, 34, 36, 37, 38, 40, 41, 43, 44, 46 Zimbabwe