Constitutional Pluralism in the European Union and Beyond 1849461252, 9781849461252

Citation preview

CONSTITUTIONAL PLURALISM IN THE EUROPEAN UNION AND BEYOND Constitutional pluralism has become immensely popular among scholars who study European integration and issues of global governance. Some of them believe that constitutionalism, traditionally thought to be bound to a nation state, can emerge beyond state borders – most importantly in the process of European integration, but also beyond that, for example, in international regulatory regimes such as the WTO, or international systems of fundamental rights protection, such as the European Convention. At the same time, the idea of constitutional pluralism has not gone unchallenged. Some have questioned its compatibility with the very nature of law and the values which law brings to constitutionalism. The critiques have come from both sides: from those who believe in the ‘traditional’ European constitutionalism based on a hierarchically superior authority of the European Union as well as from scholars focusing on constitutions of particular states. The book collects contributions taking opposing perspectives on constitutional pluralism – some defending and promoting the concept of constitutional pluralism, some criticising and opposing it. While some authors can be called ‘the founding fathers of constitutional pluralism’, others are young academics who have recently entered the field. Together they offer fresh perspectives on both theoretical and practical aspects of constitutional pluralism, enriching our existing understanding of the concept in current scholarship.

Studies of the Oxford Institute of European and Comparative Law Editor Professor Stefan Vogenauer Board of Advisory Editors Professor Mark Freedland, FBA Professor Stephen Weatherill Professor Derrick Wyatt, QC Volume 1: The Harmonisation of European Contract Law: Implications for European Private Laws, Business and Legal Practice Edited by Stefan Vogenauer and Stephen Weatherill Volume 2: The Public Law/Private Law Divide Edited by Mark Freedland and Jean-Bernard Auby Volume 3: Constitutionalism and the Role of Parliaments Edited by Katja Ziegler, Denis Baranger and AW Bradley Volume 4: The Regulation of Unfair Commercial Practices under EC Directive 2005/29: New Rules and Techniques Edited by Stephen Weatherill and Ulf Bernitz Volume 5: Human Rights and Private Law: Privacy as Autonomy Edited by Katja Ziegler Volume 6: Better Regulation Edited by Stephen Weatherill Volume 7: Forum Shopping in the European Judicial Area Edited by Pascal de Vareilles-Sommières Volume 8: The Reform of Class and Representative Actions in European Legal Systems: A New Framework for Collective Redress in Europe Christopher Hodges Volume 9: Reforming the French Law of Obligations: Comparative Reflections on the Avant-projet de réforme du droit des obligations et de la prescription (‘the Avant-projet Catala’) Edited by John Cartwright, Stefan Vogenauer and Simon Whittaker Volume 10: Performance-Oriented Remedies in European Sale of Goods Law Vanessa Mak Volume 11: Current Issues in European Financial and Insolvency Law: Perspectives from France and the UK Edited by Wolf-Georg Ringe, Louise Gullifer and Philippe Thery Volume 12: Article 82 EC: Reflections on its Recent Evolution Edited by Ariel Ezrachi Volume 13: Prohibition of Abuse of Law: A New General Principle of EU Law? Edited by Rita de la Feria and Stefan Vogenauer Volume 14: Constitutional Pluralism in the European Union and Beyond Edited by Matej Avbelj and Jan Komárek

Constitutional Pluralism in the European Union and Beyond Edited by

Matej Avbelj and Jan Komárek

OXFORD AND PORTLAND, OREGON 2012

Published in the United Kingdom by Hart Publishing Ltd 16C Worcester Place, Oxford, OX1 2JW Telephone: +44 (0)1865 517530 Fax: +44 (0)1865 510710 E-mail: [email protected] Website: http://www.hartpub.co.uk Published in North America (US and Canada) by Hart Publishing c/o International Specialized Book Services 920 NE 58th Avenue, Suite 300 Portland, OR 97213-3786 USA Tel: +1 503 287 3093 or toll-free: (1) 800 944 6190 Fax: +1 503 280 8832 E-mail: [email protected] Website: http://www.isbs.com © The editors and contributors severally 2012 The editors and contributors have asserted their right under the Copyright, Designs and Patents Act 1988 to be identified as the authors of this work. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission of Hart Publishing, or as expressly permitted by law or under the terms agreed with the appropriate reprographic rights organisation. Enquiries concerning reproduction which may not be covered by the above should be addressed to Hart Publishing Ltd at the address above. British Library Cataloguing in Publication Data Data Available ISBN: 978-1-84946-125-2 Typeset by Hope Services, Abingdon Printed and bound in Great Britain by TJ International Ltd, Padstow, Cornwall

PREFACE The essays published in this volume are the revised papers first presented at a conference that took place on 20–21 March 2009 in Oxford. The conference was organized under the auspices of the Institute of European and Comparative Law of the University of Oxford and sponsored by the Oxford/Stockholm Wallenberg Foundation Venture and the Faculty of Law, University of Oxford. We would like to thank Professors Stephen Weatherill, Stefan Vogenauer and Ulf Bernitz for their support of the conference and also during the time when the manuscript of this volume was being produced. We would also like to thank all those – rep­ resented in this volume or not – who contributed to the success of the conference as speakers and chairpersons, and Jenny Dix for her administrative support. With this volume we hope to further stimulate the debate on the legal nature of European integration and globalization, whether or not conceived in constitutional and/or pluralist terms. Ljubljana and London, May 2011 Matej Avbelj Jan Komárek

CONTENTS List of Contributors Table of Cases Table of Legislation 1 Introduction Matej Avbelj and Jan Komárek

ix xi xxiii

1

2 Constitutionalism and Pluralism in Global Context Neil Walker

17

3 Rethinking Constitutional Authority: On the Structure and Limits of Constitutional Pluralism Mattias Kumm

39

4 Three Claims of Constitutional Pluralism Miguel Poiares Maduro

67

5 Systems Pluralism and Institutional Pluralism in Constitutional Law: National, Supranational and Global Governance Daniel Halberstam

85

6 Multilevel Constitutionalism and Constitutional Pluralism Franz C Mayer and Mattias Wendel

127

7 The Fallacy of European Multilevel Constitutionalism René Barents

153

8 Federalism as Constitutional Pluralism: ‘Letter from America’ Robert Schütze

185

9 Out with the New, in with the Old – Neo-Roman Constitutional Thought and the Enigma of Constitutional Pluralism in the EU Ola Zetterquist

213

10 Institutional Dimension of Constitutional Pluralism Jan Komárek

231

11 Legal Pluralism and Institutional Disobedience in the European Union Julio Baquero Cruz 249 12 Constitutional Disagreement in Europe and the Search for Pluralism Gareth Davies

269

viii  Contents 13 The Silent Lamb and the Deaf Wolves Daniel Sarmiento

285

14 Constitutional Dialogues, Pluralism and Conflicting Identities Xavier Groussot

319

15 Monism: A Tale of the Undead Alexander Somek

343

16 Can European Integration be Constitutional and Pluralist – Both at the Same Time? Matej Avbelj 381 Index

411

LIST OF CONTRIBUTORS Matej Avbelj, Assistant Professor of European Law, Graduate School of Government and European Studies, Slovenia Julio Baquero Cruz, Member of the Legal Service of the European Commission; Visiting Professor at Sciences Po (Paris) René Barents, Director of Research and Documentation, Court of Justice of the EU; Professor of European Community Law, Maastricht University Gareth Davies, Professor of European Law, VU University Amsterdam Xavier Groussot, Associate Professor of European Union law, Lund University Daniel Halberstam, Eric Stein Collegiate Professor of Law, Director, European Legal Studies Program, University of Michigan Law School Jan Komárek, Lecturer in EU law, European Institute and Department of Law, London School of Economics and Political Science Mattias Kumm, Research Professor ‘Rule of Law in the Age of Globalization’ at the Social Science Research Center Berlin; Professor of Law at Humboldt University of Berlin; Professor of Law at the New York University School of Law Franz C Mayer, Professor of public law, European law, public international law, comparative law and law and politics, University of Bielefeld Miguel Poiares Maduro, Professor of European Law; Director of the Global Governance Programme, European University Institute Florence Daniel Sarmiento, Professor of EU Law at the University Complutense of Madrid; référendaire, Court of Justice of the European Union Robert Schütze, Reader, Durham Law School Alexander Somek, Charles E Floete Chair in Law, College of Law, University of Iowa Neil Walker, Regius Professor of Public Law and the Law of Nature and Nations, University of Edinburgh Mattias Wendel, Research Fellow, Walter-Hallstein Institute for European Constitutional Law, Humboldt University Berlin Ola Zetterquist, Associate Professor of European law, Gothenburg University; Justice at the Court of Appeal for Western Sweden

TABLE OF CASES ECJ Court of Justice (By Case Name) Abrahamsson Case C-407/98 [2000] ECR I-5539.................................................322 Adria-Wien Pipeline Case C-144/99 [2001] ECR I-8365.....................................323 Advocaten voor de Wereld Case C-303/05 [2007] ECR I-3633............................312 Albatros Case 20/64 [1964] ECR 40.......................................................................309 Alpe Adria Energia Case C-205/08 [2009] ECR I-11525..............................239, 323 AM&S Case 155/79 [1982] ECR 1575....................................................................165 Arduino Case C-35/99 [2002] ECR I-1529............................................................293 Bartsch Case C-427/06 [2008] ECR I-7245...................................................... 337–8 Bickel and Franz Case C-274/96 [1998] ECR I-7637............................................280 Cartesio Case C-210/06 [2008] ECR I-9641......................................292, 302–5, 323 Centro Europa 7 Case C-380/05 [2008] ECR I-349..........................100, 293, 296–7 Chacón Navas Case C-13/05 [2006] ECR I-6467..................................................337 CILFIT and Others Case 283/81 [1982] ECR 3415.............14, 309–10, 313–17, 335 Ciola v Land Vorarlberg Case C-224/97 [1999] ECR I-2517................................280 Cipolla and Others Joined Cases C-94/04 and C-202/04 [2006] ECR I-11421........................................................................................................293 Comet Case 45/76 [1976] ECR 2043........................................................................43 Commission v Belgium Case 149/79 [1980] ECR I-3881.....................................332 Commission v Council Case C-176/03 [2005] ECR I-7879.................................291 Commission v Council Case C-440/05 [2007] ECR I-0000.................................291 Commission v Luxembourg and Belgium Case 90/63 [1964] ECR 625..............106 Commission v Luxembourg Case C-473/93 [1996] ECR I-3207.........................332 Commission v Poland Case C-165/08 [2009] ECR I-00000.................................340 Commission v Spain Case C-154/08 [2009], nyr..................................................315 Costa v ENEL Case 6/64 [1964] ECR 585............... 2, 10, 43, 45–7, 49, 54–5, 59–61, 98, 106, 162, 181, 214, 226, 287, 309, 393 Da Costa Joined Cases 28/62, 29/62 and 30/62 [1963] ECR 61............................314 De Coster Case C-17/00 [2001] ECR I-9445.........................................................331 De Geus Case 13/61 [1962] ECR 45...............................................................134, 162 Defrenne (II) Case 43/75 [1976] ECR 455.............................................................228 Dynamic Medien Case C-244/06 [2008] ECR I-505.............................................339 Factortame Case C-213/89 [1990] ECR I-2433.....................................................321 FIAMM and FIAMM Joined Cases C-120/06 P and C-121/06 P [2008] ECR I-6513..........................................................................................................336

xii  Table of Cases FII Group Litigation (Test Claimants) Case C-446/04 [2006] ECR I-11753.......165 Foglia v Novello Case 104/79 [1980] ECR 745......................................................293 Foglia v Novello (No 2) Case 244/80 [1981] ECR 3045........................................293 Fratelli Costanzo Case 103/88 [1989] ECR 1839.............................................. 238–9 Gambelli and Others Case C-243/01 [2003] ECR I-13031.........301–3, 308, 339–40 Garcia Avello Case C-148/02 [2003] ECR I-11613................................................280 Gaston-Schul Case C-461/03 [2005] ECR I-10513...............................................313 GB-Inno-BM Case C-362/88 [1990] ECR I-667...................................................294 Gebhard Case C-55/94 [1995] ECR I-4165...................................................282, 339 Giovanni Dell’Orto Case C-467/05 [2007] ECR I-5557.......................................322 Grado and Bashir Case C-291/96 [1997] ECR I-5531..........................................293 Greenpeace France Case C-6/99 [2000] ECR I-1651............................................165 Huber Case C-366/00 [2002] ECR I-7736.............................................................165 Humblet v Belgium Case 6/60 [1960] ECR 1167..................................................162 IN.CO.GE ’90 Joined Cases C-10/97 to C-22/97 [1998] ECR I-6307..........136, 178 International Transport Workers’ Federation and Finnish Seamen’s Union Case C-438/05 [2007] ECR I-10779 (Viking Line)...................................235, 300 Internationale Handelsgesellschaft Case 11/70 [1970] ECR 1125........................................................... 98, 106, 226–7, 233, 272, 279, 332 Job Centre Case C-55/96 [1997] ECR I-7119........................................................323 Kadi and Al Barakaat International Foundation v Council and Commission Joined Cases C-402/05 P and C-415/05 P [2008] ECR I-6351................................................... 7, 10, 24, 39, 41–3, 46–7, 53–5, 62–3, 96, 106, 148, 151, 228, 267, 319, 364, 368, 376 Köbler Case C-224/01 [2003] ECR I-10239.239, 304, 313–15, 320–1, 324, 328, 335 Konstantinidis Case C-168/91 [1993] ECR I-1191...............................................167 Kreil Case C-285/98 [2000] I-69....................................................................136, 295 Kremzow Case C-299/95 [1997] ECR I-2629........................................................293 Kurrer Case 33/67 [1968] ECR 179........................................................................163 Läärä Case C-124/97 [1999] ECR I-6067....................................................... 339–40 Laval un Partneri Case C-341/05 [2007] ECR I-11767.........................................300 Les Verts v Parliament Case 294/83 [1986] ECR 1339....................................44, 234 Liga Portuguesa Case C-42/07, nyr..................................................301, 333, 339–40 Lindorfer v Council Case C-227/04 P [2007] ECR I-6767...................................337 Mangold Case C-144/04 [2005] ECR I-9981.............................................291, 337–9 Marleasing Case C-106/89 [1980] ECR I-4135.....................................................178 Maruko Case C-267/06 [2008] ECR I-1757..................................................... 337–8 Maurin Case C-144/95 [1996] ECR I-2909...........................................................293 Mayr Case C-506/06 [2008] ECR I-1017...............................................................295 Michaniki Case C-213/07 [2008] ECR I-9999........................... 295, 297, 307, 331–3 Nationale Loterij Case C-525/06, nyr....................................................................303 Oleifici Borelli Case C-97/91 [1992] ECR I-6313..................................................165 Omega Case C-36/02 [2004] ECR I-9609......................................................333, 339 Österreichischer Gewerkschaftsbund Case C-195/98 [2000] ECR I-10497.........322

Table of Cases  xiii Parfum Christian Dior Case C-337/95 [1997] ECR I-6013..................................324 Pfeiffer Joined Cases C-397/01 to C-403/01 [2004] ECR I-8835.........................178 Placanica and Others Joined Cases C-338/04, C-359/04 and C-360/04 [2007] ECR I-1891................................................................ 298, 301–2, 304, 308, 339–40 Portugal v Commission Case C-88/03 [2006] ECR I-7115..........................280, 299 PreussenElektra Case C-379/98 [2001] ECR I-2099.............................................293 Pupino Case C-105/03 [2005] ECR I-5285...........................................................322 Rewe-Markt Steffen Case 158/80 [1981] ECR 1805..............................................165 Rewe-Zentral (Cassis de Dijon) Case 120/78 [1979] ECR 649.............................282 Rheinmühlen Case 166/73 [1974] ECR 33............................................................163 Rheinmühlen II Case 146/73 [1974] ECR 139......................................................308 Roda Golf & Beach Resort SL Case C-14/08 [2009] ECR I-00000.................. 321–3 Salgoil Case 13/68 [1968] ECR 453........................................................................309 Schindler Case C-275/92 [1994] ECR I-1039................................................. 339–40 Schmidberger Case C-112/00 [2003] ECR I-5659................................................339 Simmenthal II Case 106/77 [1978] ECR 629........................... 43, 136, 306, 321, 357 Société Arcelor Atlantique et Lorraine Case C-127/07 [2008] ECR I-09895................................................................................................327, 331 Society for the Protection of Unborn Children Ireland Case C-159/90 [1991] ECR I-4685 (Grogan)........................................................................ 293–6 Stauder ECR Case 29/69 [1969] 419......................................................................226 Steinicke Case C-77/02 [2003] ECR I-9027...........................................................339 Stork v High Authority Case 1/58 [1958] ECR 63................................................162 Telemarsicabruzzo and Others Joined Cases C-320/90 to C-322/90 [1993] ECR I-393................................................................................................293 Unibet Case C-432/05 [2007] ECR I-2271............................................165, 321, 329 Unión General de Trabajadores de La Rioja Joined Cases C-428/06 to C-434/06 [2008] ECR I-6747......................................... 280, 299–300, 308, 310 United Pan Europe Case C-250/06 [2007] ECR I-11135......................................339 Vaassen-Göbbels Case 61/65 [1966] ECR 377...............................................293, 331 Van der Weerd and Others Joined Cases C-222/05 to C-225/05 [2007] ECR I-4233..........................................................................................................293 Van Gend en Loos Case 26/62 [1963] ECR 1...........96, 106, 164, 181, 214, 225, 287 Victoria Film Case C-134/97 [1998] ECR I-7023..................................................322 Watson and Belmann Case 118/75 [1976] ECR 1185...........................................339 Wiener SI GmbH and Hauptzollamt Emmerich Case C-338/95 [1997] ECR I-6495..........................................................................................................313 Wilhelm Case 14/68 [1969] ECR I.........................................................................162 Zenatti Case C-67/98 [1999] ECR I-7289....................................301–2, 308, 339–40 By Case Number Case 1/58 Stork v High Authority [1958] ECR 63................................................162 Case 6/60 Humblet v Belgium [1960] ECR 1167..................................................162

xiv  Table of Cases Case 13/61 De Geus [1962] ECR 45...............................................................134, 162 Case 26/62 Van Gend en Loos [1963] ECR 1............96, 106, 164, 181, 214, 225, 287 Joined Cases 28/62, 29/62 and 30/62 Da Costa [1963] ECR 61............................314 Case 90/63 Commission v Luxembourg and Belgium [1964] ECR 625..............106 Case 6/64 Costa v ENEL [1964] ECR 585........................... 2, 10, 43, 45–7, 49, 54–5, 59–61, 98, 106, 162, 181, 214, 226, 287, 309, 393 Case 20/64 Albatros [1964] ECR 40.......................................................................309 Case 61/65 Vaassen-Göbbels [1966] ECR 377...............................................293, 331 Case 33/67 Kurrer [1968] ECR 179........................................................................163 Case 13/68 Salgoil [1968] ECR 453........................................................................309 Case 14/68 Walt Wilhelm [1969] ECR I.................................................................162 Case 29/69 Stauder ECR [1969] 419......................................................................226 Case 11/70 Internationale Handelsgesellschaft [1970] ECR 1125........................................................... 98, 106, 226–7, 233, 272, 279, 332 Case 146/73 Rheinmühlen II [1974] ECR 139......................................................308 Case 166/73 Rheinmühlen [1974] ECR 33............................................................163 Case 43/75 Defrenne (II) [1976] ECR 455.............................................................228 Case 118/75 Watson and Belmann [1976] ECR 1185...........................................339 Case 45/76 Comet [1976] ECR 2043........................................................................43 Case 106/77 Simmenthal II [1978] ECR 629........................... 43, 136, 306, 321, 357 Case 120/78 Rewe-Zentral (Cassis de Dijon) [1979] ECR 649.............................282 Case 104/79 Foglia v Novello [1980] ECR 745......................................................293 Case 149/79 Commission v Belgium [1980] ECR I-3881.....................................332 Case 155/79 AM&S [1982] ECR 1575....................................................................165 Case 158/80 Rewe-Markt Steffen [1981] ECR 1805..............................................165 Case 244/80 Foglia v Novello (No 2) [1981] ECR 3045........................................293 Case 283/81 CILFIT and Others [1982] ECR 3415.............14, 309–10, 313–17, 335 Case 294/83 Les Verts v Parliament [1986] ECR 1339....................................44, 234 Case 103/88 Fratelli Costanzo [1989] ECR 1839.............................................. 238–9 Case C-362/88 GB-Inno-BM [1990] ECR I-667...................................................294 Case C-106/89 Marleasing [1980] ECR I-4135.....................................................178 Case C-213/89 Factortame [1990] ECR I-2433.....................................................321 Case C-159/90 Society for the Protection of Unborn Children Ireland [1991] ECR I-4685 (Grogan)........................................................................ 293–6 Joined Cases C-320/90 to C-322/90 Telemarsicabruzzo and Others [1993] ECR I-393................................................................................................293 Case C-97/91 Oleifici Borelli [1992] ECR I-6313..................................................165 Case C-168/91 Konstantinidis [1993] ECR I-1191...............................................167 Case C-275/92 Schindler [1994] ECR I-1039................................................. 339–40 Case C-473/93 Commission v Luxembourg [1996] ECR I-3207.........................332 Case C-55/94 Gebhard [1995] ECR I-4165...................................................282, 339 Case C-144/95 Maurin [1996] ECR I-2909...........................................................293 Case C-299/95 Kremzow [1997] ECR I-2629........................................................293 Case C-337/95 Parfum Christian Dior [1997] ECR I-6013..................................324

Table of Cases  xv Case C-338/95 Wiener SI GmbH and Hauptzollamt Emmerich [1997] ECR I-6495..........................................................................................................313 Case C-55/96 Job Centre [1997] ECR I-7119........................................................323 Case C-274/96 Bickel and Franz [1998] ECR I-7637............................................280 Case C-291/96 Grado and Bashir [1997] ECR I-5531..........................................293 Joined Cases C-10/97 to C-22/97 IN.CO.GE ’90 [1998] ECR I-6307..........136, 178 Case C-124/97 Läärä [1999] ECR I-6067....................................................... 339–40 Case C-134/97 Victoria Film [1998] ECR I-7023..................................................322 Case C-224/97 Ciola v Land Vorarlberg [1999] ECR I-2517................................280 Case C-67/98 Zenatti [1999] ECR I-7289....................................301–2, 308, 339–40 Case C-195/98 Österreichischer Gewerkschaftsbund [2000] ECR I-10497.........322 Case C-285/98 Kreil [2000] I-69....................................................................136, 295 Case C-379/98 PreussenElektra [2001] ECR I-2099.............................................293 Case C-407/98 Abrahamsson [2000] ECR I-5539.................................................322 Case C-6/99 Greenpeace France [2000] ECR I-1651............................................165 Case C-35/99 Arduino [2002] ECR I-1529............................................................293 Case C-144/99 Adria-Wien Pipeline [2001] ECR I-8365.....................................323 Case C-17/00 De Coster [2001] ECR I-9445.........................................................331 Case C-112/00 Schmidberger [2003] ECR I-5659................................................339 Case C-366/00 Huber [2002] ECR I-7736.............................................................165 Case C-224/01 Köbler [2003] ECR I-10239...................................239, 304, 313–15, 320–1, 324, 328, 335 Case C-243/01 Gambelli and Others [2003] ECR I-13031.........301–3, 308, 339–40 Joined Cases C-397/01 to C-403/01 Pfeiffer [2004] ECR I-8835.........................178 Case C-36/02 Omega [2004] ECR I-9609......................................................333, 339 Case C-77/02 Steinicke [2003] ECR I-9027...........................................................339 Case C-148/02 Garcia Avello [2003] ECR I-11613................................................280 Case C-88/03 Portugal v Commission [2006] ECR I-7115..........................280, 299 Case C-105/03 Pupino [2005] ECR I-5285...........................................................322 Case C-176/03 Commission v Council [2005] ECR I-7879.................................291 Case C-461/03 Gaston-Schul [2005] ECR I-10513...............................................313 Joined Cases C-94/04 and C-202/04 Cipolla and Others [2006] ECR I-11421...293 Case C-144/04 Mangold [2005] ECR I-9981.............................................291, 337–9 Case C-227/04 P Lindorfer v Council [2007] ECR I-6767...................................337 Joined Cases C-338/04, C-359/04 and C-360/04 Placanica and Others [2007] ECR I-1891.................................................... 298, 301–2, 304, 308, 339–40 Case C-446/04 Test Claimants in the FII Group Litigation [2006] ECR I-11753........................................................................................................165 Case C-13/05 Chacón Navas [2006] ECR I-6467..................................................337 Joined Cases C-222/05 to C-225/05 Van der Weerd and Others [2007] ECR I-4233..........................................................................................................293 Case C-303/05 Advocaten voor de Wereld [2007] ECR I-3633............................312 Case C-341/05 Laval un Partneri [2007] ECR I-11767.........................................300 Case C-380/05 Centro Europa 7 [2008] ECR I-349..........................100, 293, 296–7

xvi  Table of Cases Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat International Foundation v Council and Commission [2008] ECR I-6351................................................... 7, 10, 24, 39, 41–3, 46–7, 53–5, 62–3, 96, 106, 148, 151, 228, 267, 319, 364, 368, 376 Case C-432/05 Unibet [2007] ECR I-2271............................................165, 321, 329 Case C-438/05 International Transport Workers’ Federation and Finnish Seamen’s Union [2007] ECR I-10779 (Viking Line).................................235, 300 Case C-440/05 Commission v Council [2007] ECR I-0000.................................291 Case C-467/05 Giovanni Dell’Orto [2007] ECR I-5557.......................................322 Joined Cases C-120/06 P and C-121/06 P FIAMM and FIAMM [2008] ECR I-6513..........................................................................................................336 Case C-210/06 Cartesio [2008] ECR I-9641......................................292, 302–5, 323 Case C-244/06 Dynamic Medien [2008] ECR I-505.............................................339 Case C-250/06 United Pan Europe [2007] ECR I-11135......................................339 Case C-267/06 Maruko [2008] ECR I-1757..................................................... 337–8 Case C-427/06 Bartsch [2008] ECR I-7245...................................................... 337–8 Joined Cases C-428/06 to C-434/06 Unión General de Trabajadores de La Rioja [2008] ECR I-6747..........................................280, 299–300, 308, 310 Case C-506/06 Mayr [2008] ECR I-1017...............................................................295 Case C-525/06 Nationale Loterij, 24 March 2009, nyr..........................................303 Case C-42/07 Liga Portuguesa, nyr..................................................301, 333, 339–40 Case C-127/07 Société Arcelor Atlantique et Lorraine [2008] ECR I-09895................................................................................................327, 331 Case C-213/07 Michaniki [2008] ECR I-9999........................... 295, 297, 307, 331–3 Case C-14/08 Roda Golf & Beach Resort SL [2009] ECR I-00000.................. 321–3 Case C-154/08 Commission v Spain [2009], nyr..................................................315 Case C-165/08 Commission v Poland [2009] ECR I-00000.................................340 Case C-205/08 Alpe Adria Energia [2009] ECR I-11525..............................239, 323 Opinions Opinion 1/78 Rubber [1979] ECR 287..................................................................163 Opinion 1/91 EEA I [1991] ECR I-6079..........................................................96, 181 General Court/Court of First Instance Case T-289/03 BUPA [2008] ECR II-741...............................................................339 Case T-315/01 Kadi v Council and Commission [2005] ECR II-3649....................7

European Court of Human Rights Case 45036/98, judgment of 30 June 2005, Bosphorus v Ireland [2006] 42 EHRR 1..................................................................................... 62, 319, 327, 368

Table of Cases  xvii

International ICJ Case concerning the United States Diplomatic and Consular Staff in Tehran (United States of America v Iran) [1980] ICJ Rep 3.............................102 Corfu Channel Case (UK v Albania) (Merits) [1949] ICJ Rep 4..........................102 Reservations to the Convention on the Preservation and Punishment of the Crime of Genocide [1951] ICJ Rep 15........................................................102 PCIJ Case of the SS Wimbledon Series A No 1 (1923) PCIJ Rep 25.............................102

National Austria Constitutional Court Decision B 2251, 2594/97 Adria-Wien Pipeline....................................................324 Belgium Constitutional Court Case No 12/94, judgment of 3 February 1994, European Schools [1994] BS 6137..................................................................................................................59 Cour d’arbitrage belge Case No 6/97, judgment of 19 February 1997.......................................................323 Court of Appeal of Antwerp Judgment of 8 November 2007..............................................................................304 Cyprus Supreme Court Civil Appeal No 294/2005, judgment of 7 November 2005, European Arrest Warrant.........................................................................................................73, 312 Czech Republic Constitutional Court judgment of 8 March 2006, Pl. ÚS 50/04, Sugar Quota Regulation III....... 139, 147–8 judgment of 3 May 2006, Pl. ÚS 66/04..................................................................312

xviii  Table of Cases judgment of 28 November 2008, Pl. US 19/08, Lisbon I.........................................74 judgment of 3 November 2009, Pl. US 29/09 Lisbon II..................................74, 148 Denmark Supreme Court Carlsen v Rasmussen [1998] UfR 800, I 361/1997................................................146 Carlsen v Prime Minister [1999] 3 Common Market Law Review 854...............328 France Conseil d’Etat judgment of 7 February 2007 No 287110 (Arcelor)....................................74, 326–7 Constitutional Council judgment of 2 September 1992 No 92-312 DC (Maastricht II).............................73 judgment of 23 September 1992 No 92-313 DC (Maastricht III)........................147 judgment of 19 November 2004 No 2004-505 DC...............................................332 Germany Federal Constitutional Court Case 2 BvG 1/51 Südweststaat [1951] BVerfGE 1, 14............................................136 Case 1 BvR 248/63, 216/67, EWG-Verordnungen [1967] BVerfGE 22, 293.................................................................................................................57, 134 Case 2 BvR 255/69 Milchpulver [1971] BVerfGE 31, 145...............................57, 134 Case BvL 52/71 Solange I [1974] BVerfGE 37, 271...................................21, 57, 134 Case 2 BvR 499/74, 1042/75 Rückwirkende Verordnungen [1977] BVerfGE 45, 142..................................................................................................134 Case 2 BvL 6/77 Vielleicht-Beschluss [1979] BVerfGE 52, 187.............................134 Case 2 BvH 1, 2/82, 2 BvR 233/82 Startbahn West [1982] BVerfGE 60, 175........136 Case 1 BvR 1025/82, 1 BvL 16/83, 10/91 Nachtarbeitsverbot [1992] BVerfGE 85, 191..................................................................................................134 Case 2 BvR 197/83 Solange II [1986] BVerfGE 73, 339...........................57, 134, 165 Case 1 BvR 1257/84 Herrnburger Bericht [1987] 77 BVerfGE 240......................112 Case 2 BvR 687/85 Kloppenburg [1987] BVerfGE 75, 223...................................134 Case 1 BvR 1215/87 Nationalhymne [1990] 81 BVerfGE 298..............................112 Case 2 BvL 12, 13/88, 2 BvR 1436/87 Absatzfonds [1990] BVerfGE 82, 159........134 Case 1 BvR 1087/91, [1995] BVerfGE 93, 1 (Crucifixes in Bavarian Classrooms)...........................................................................................................72 Case 2 BvR 2134, 2159/92 Maastricht [1993] BVerfGE 89, 155; [1994] 1 CMLR 57...................................... 2, 12, 21, 49, 51, 53, 58–9, 73, 96, 127–8, 132, 134, 142–3, 145, 155, 182, 186, 215, 231–2, 236, 275, 287, 289–90, 320, 328, 345–6, 356–7, 381 Case 2 BvR 1481/04 Görgülü I [2004] BVerfGE 111, 307.....................................144

Table of Cases  xix Case 2 BvR 2236/04, European Arrest Warrant [2005] BVerfGE 113, 273...............................................................................................266, 312, 356 Case 2 BvR 2661/06, judgment of 6 July 2010, Honeywell............... 58, 74, 148, 251 Case 2 BvE 2/08 et al, judgment of 30 June 2009, Lisbon............................................................7, 49, 53, 58, 73–4, 96, 128, 132, 134, 143–4, 146, 148–9, 251, 274, 319, 334, 408 Case 1 BvR 256/08, 1 BvR 263/08, 1 BvR 568/08, judgment of 2 March 2010, Data Retention......................................................................................................59 Greece State Council Katsarou v KI.KATSA, 25 September 1998............................................................307 Hungary Constitutional Court decision of 22 June 1998, 30/1998 (VI 25) AB, Europe Agreement, [1998] MK 1998/55.........................................................................................................135 Ireland Attorney General (at the relation of the Society for the Protection of Unborn Children Ireland Ltd) v Open Door Counselling Ltd and Dublin Wellwoman Centre Ltd [1988] Irish Reports 593............................................................. 293–4 Italy Corte Costituzionale Decision No 183/1973 of 27 December 1973........................................................330 Decision No 170/1984 of 8 June 1984....................................................................330 Decision No 826/1988 of 13 July 1988.....................................................................72 Decision No 420/1994 of 5 December 1994............................................................72 Decision No 536/1995 Messaggero Servizi............................................................324 Decision No 28/2001 General Electoral System....................................................324 Decision No 466/2002 of 20 November 2002..........................................................72 Decision No 275/2003 of 8 July 2003 Federfarma.....................................305–6, 315 Decision No 103/2008............................................................................................330 Corte suprema di cassazione Decision No 111/04 of 26 April 2004 Gesualdi.....................................................302 Lombardy Regional Court Decision No 4195/2004 of 24 June 2004 Federfarma................................305–6, 315

xx  Table of Cases Netherlands Supreme Court A 13V v Public Prosecutor, 2 November 2004, NJ (2005) No 80, LJN No AR 1797...................................................................................................96 Poland Constitutional Tribunal Case K 18/04 judgment of 11 May 2005, K 18/04, Poland’s membership in the European Union (The Accession Treaty)......................................21, 138, 147 Case P 1/05 judgment of 27 April 2005, European Arrest Warrant............................................................................... 59, 73–4, 232, 312, 322 Portugal Constitutional Court Decision of 1 February 1989, Case No 184/89 (ERDF)..........................................73 Spain Constitutional Court Case 1236/92, judgment of 1 July 1992, Maastricht [1992] 177 BOE 5...........59, 73 Declaration of 13 December 2004, 1/2004 [2005] 3 BOE 5.............. 7, 130, 135, 266 Supreme Court Judgment of 9 December 2004 (recurso de casación No 7893/1999)..................308 Sweden Supreme Administrative Court Barsebäk case, RÅ 1999, ref. 76...............................................................................328 Supreme Court Data Delecta Case NJA 1996, 668...........................................................................328 Volvo Service Case NJA 1998, 474..........................................................................328 United Kingdom Salomon v Commissioners of Customs and Excise [1967] 2 QB 116..................215 United States Agostini v Felton, 521 U.S. 203 (1997)...................................................................115 Alden et al v Maine, 527 U.S. 706 (1998).......................................................... 208–9 Brown v Allen, 344 U.S. 443, 540 (1953)...............................................................242

Table of Cases  xxi Brown v Board of Education of Topeka, 347 U.S. 483 (1954)..............................203 Chisholm v State of Georgia, 2 U.S. 419 (1793)....................................................191 City of Boerne v Flores, 521 U.S. 507 (1997).........................................................114 Cooper v Aaron, 358 U.S. 1, 18 (1958)...................................................................114 Dred Scott v Sandford 60 U.S. 393 (1857).............................................................244 Fairfax’s Devisee v Hunter’s Lessee, 11 U.S. 603 (1813)........................................186 Foster v Neilson, 27 U.S. 253 (1829)......................................................................108 Harper v Virginia State Board of Elections, 383 U.S. 663 (1966).........................115 Hunter v Martin, 18 Va. 1 (1815)...........................................................................186 Lassiter v Northampton County Board of Elections, 360 U.S. 45 (1959)............115 Marbury v Madison, 5 U.S. 137 (1803)............................................................91, 116 Martin v Hunter’s Lessee, 14 U.S. 304 (1816)........................................................186 Merryman, Ex parte17 F. Cas. 144 (1861).............................................................115 Myers v United States, 272 U.S. 52 (1926).............................................................115 Northwest Austin Mun. Utility Dist. No. One v Holder, 129 S. Ct. 2504 (2009)..................................................................................................................115 Plessy v Ferguson, 163 U.S. 537 (1896)..........................................................203, 245 Roe v Wade, 410 U.S. 113 (1973)............................................................................246 U.S. Term Limits, Inc. v Thornton, 514 U.S. 779 (1995).................................. 206–8

TABLE OF LEGISLATION European Union Charter of Fundamental Rights (CFR).................................... 61, 162, 172, 279, 338    Art 21...................................................................................................................338    Art 52(4)..............................................................................................................332 Directive 76/207/EEC.............................................................................................295 Directive 89/391/EEC    Art 16(1)..............................................................................................................295 Directive 92/85/EEC...............................................................................................295 Directive 96/71/EC..................................................................................................300 Directive 2000/78....................................................................................................338 Directive 2003/87....................................................................................................326 EC Treaty (former)................................................................... 96, 106, 163, 214, 298   Preamble..............................................................................................................225    Art 13...................................................................................................................337    Art 28...................................................................................................................237    Art 43...........................................................................................................300, 302    Art 49............................................................................... 293–4, 296, 300, 304, 306    Art 87...........................................................................................................299, 308    Art 87(1)..................................................................................................... 299–300    Art 88...........................................................................................................299, 308    Art 149.................................................................................................................307    Art 149(1)............................................................................................................307    Art 220.........................................................................................................161, 234    Art 230...................................................................................................................58    Art 234......................................................................................... 165, 266, 303, 307    Art 234(3)............................................................................................................328    Art 307...........................................................................................................52, 106 Framework Decision 2002/584/JHA..............................................................232, 237 Lisbon Treaty..................................................... 7, 21, 61, 98, 100, 111, 130, 134, 153, 158–9, 162, 172–5, 177, 229, 251, 266, 274, 322, 334, 407   see also Treaty on European Union; Treaty on the Functioning of the   European Union    Declaration No 17...........................................................................................60, 98    Declaration No 115...............................................................................................46    Protocol No 2......................................................................................................100

xxiv  Table of Legislation Maastricht Treaty................................................................................ 21, 56, 215, 294   see also Treaty on European Union    Art F(1)................................................................................................................331 Single European Act..........................................................................................45, 287 Treaty on European Union (TEU).........................46, 49, 52–3, 56, 73, 98, 100, 160, 173–4, 181, 215, 233–4, 271, 275, 287, 334    Art 1................................................................................................................ 173–4    Art 2.....................................................................................................................172    Art 4.....................................................................................................................309    Art 4(2)..................................................................................61, 139, 148–9, 331–2    Art 4(3)................................................................................................................274    Art 5(1)................................................................................................................160    Art 6.......................................................................................60, 161–2, 225, 296–7    Art 6(3) (former)................................................................................................331    Art 7.....................................................................................................................166    Art 10(1)..............................................................................................................174    Art 19...........................................................................................................161, 311    Art 48.............................................................................................................98, 174    Art 50...............................................................................................................96, 98 Treaty on the Functioning of the European Union (TFEU)...................98, 100, 165    Art 5.......................................................................................................................99    Art 14...................................................................................................................339    Art 49...........................................................................................................300, 302    Art 56........................................................................... 293, 296, 300, 302, 304, 306    Art 107.........................................................................................................299, 308    Art 108.........................................................................................................299, 308    Art 165.................................................................................................................307    Art 194(1)(d).......................................................................................................134    Art 194(2)............................................................................................................134    Art 220.................................................................................................................105    Art 235...................................................................................................................99    Art 257.................................................................................................................291    Art 258.................................................................................................................315    Art 267.............. 165, 266, 286, 291–2, 298–9, 303–5, 307, 309, 311, 314, 323, 335    Art 267(3)..............................................................................315–16, 324, 328, 331    Art 347.................................................................................................................105    Art 348.................................................................................................................106    Art 351.................................................................................................................106    Art 352.................................................................................................................290

Table of Legislation  xxv

International Constitution of the WHO......................................................................................128 European Convention on Human Rights..............................................................338    Art 6.....................................................................................................................323 Montevideo Convention of 1933    Art 1.....................................................................................................................216 United Nations   Charter........................................................................................ 26, 39, 46, 65, 142     Art 25.................................................................................................................52     Art 103..................................................................................... 39, 42, 46, 52, 101    Security Council Resolution S/RES/1441 of 8 November 2002.......................110    Security Council Resolution S/RES/1730 of 19 December 2006........................62    Security Council Resolution S/RES/1735 of 22 December 2006........................62    Security Council Resolution S/RES/1822 of 30 June 30 2008............................62    Security Council Resolution S/RES/1904 of 17 December 2009................... 62–3 Vienna Convention on the Law of Treaties (VCLT)    Art 27...................................................................................................................274    Art 30.....................................................................................................................52    Art 30(2)................................................................................................................53    Arts 31-33..............................................................................................................52

National Austria Basic Law    Art 133(4)............................................................................................................323 Federal Constitutional Law    Arts 10-15..............................................................................................................94 Cyprus Constitution    Art 182.................................................................................................................147 Czech Republic Constitution of the Czech Republic    Art 1(1)................................................................................................................148    Art 10a.................................................................................................................139    Arts 10a and 10b...................................................................................................98

xxvi  Table of Legislation France Constitution of the French Republic....................................... 98, 100, 147, 326, 332    Art 55...................................................................................................................327    Art 61-1...............................................................................................................325    Art 88-1...............................................................................................................327    Arts 88-1 to 88-7...................................................................................................98   Title XV...............................................................................................................100 Constitutional Law No 2008-724 of 23 July 2008.................................................325    Art 29...................................................................................................................325 Déclaration des droits de l’Homme et du Citoyen (1789)....................................105 Decree 2004-832.....................................................................................................326 Germany Act Approving the Treaty of Lisbon.......................................................................134 Basic Law............................................................ 56–7, 95–7, 100, 120–2, 139, 144–7, 165, 203, 206, 251, 275, 295, 334, 388   Preamble..............................................................................................................150    Art 12a(4)............................................................................................................295    Art 23.......................................................................................................56, 97, 100    Art 23(1)..............................................................................................139, 146, 150    Art 24.....................................................................................................................56    Art 59 II.................................................................................................................56    Art 72(3)................................................................................................................95    Art 79...................................................................................................................122    Art 79(3)..............................................................................................................146    Art 93(1)(1).........................................................................................................334    Art 93(1)(2).........................................................................................................334    Art 93(1)(3).........................................................................................................334    Art 94(2)..............................................................................................................121    Art 100(1)............................................................................................................334    Art 101.................................................................................................................165    Art 146.................................................................................................................388 Federal Constitutional Court Law.........................................................................121    Art 2(3)................................................................................................................123    Art 3.....................................................................................................................123    Art 6.....................................................................................................................123    Art 7.....................................................................................................................123 Greece Constitution............................................................................................................307    Art 14(9)..............................................................................................................332

Table of Legislation  xxvii    Art 16.......................................................................................................306–7, 315    Art 16(5)..............................................................................................................307    Art 110.................................................................................................................147 Ireland Constitution Art 40, s 3............................................................................................................ 293–4 Italy Constitution............................................................................................................306    Art 11.....................................................................................................................97    Art 127.................................................................................................................330    Art 137.................................................................................................................330    Art 139.................................................................................................................147 Sardinia Regional Law no 4/2006, Art 4...............................................................................330 Netherlands Constitution    Art 91(3)................................................................................................................96 Poland Constitution............................................................................................................233    Art 55(1)..............................................................................................................232 Portugal Constitution    Art 288.................................................................................................................147 Romania Constitution    Art 152(1) and (2)...............................................................................................147 Spain Constitution....................................................................................................299, 324    Additional Provision 1........................................................................................299

xxviii  Table of Legislation Sweden Instruments of the Government   Ch 10:5................................................................................................................329 United Kingdom European Communities Act 1972   s 68.......................................................................................................................100 United States Alien and Sedition Acts...............................................................................188–9, 210 Authorization for the Use of Military Force against Iraq Resolution 2002 (House Joint Resolution 114).............................................................................110 Bill of Rights....................................................................................................114, 188 California    Constitution, Art III, § 1.......................................................................................94 Constitution........................................................... 12, 41, 94–5, 104–5, 108, 113–20, 122, 125, 180, 186–92, 195–207, 209–10, 244, 246   Preamble..............................................................................................................389    Art I, s 2, cl 2........................................................................................................206    Art I, s 4, cl 1........................................................................................................207    Art I, s 5, cl 1........................................................................................................207    Art III...........................................................................................................191, 246    Art IV.....................................................................................................................95    Art V............................................................................................................122, 195    Art VI.............................................................................................................94, 186    Art VII............................................................................................................ 207–8   Eleventh Amendment.............................................................................191, 208–9   Fourteenth Amendment...........................................................114–15, 201, 203–4 Declaration of Independence.................................................................................104 Religious Freedom Restoration Act (RFRA) 42 U.S.C. § 2000bb-2000bb-4........114 Tenure in Office Act................................................................................................114 Voting Rights Act 42 U.S.C. § 1973-1973aa-6................................................ 114–15

1 Introduction MATEJ AVBELJ AND JAN KOMÁREK Human creativity knows no limits, but nor do the challenges that people face; they constantly fight hermeneutical battles with the surrounding world. New concepts are frequently born to provide tools for grasping and at the same time potentially changing the social reality. However, it is not known how the new concepts would fare in practice at the time of their inception. Some are more successful than others. What all bring along, however, is a claim that something about the ‘old’ is wrong and needs to be redone. Constitutional pluralism, the concept that this book explores, has emerged from hermeneutical clashes of this kind and seems to have gained a remarkable success. Since the mid 1990s, it has grown from a concept that was only unconsciously used to an idea that has drawn remarkable attention and even resulted, according to some, in a ‘movement’.1 The concept of such a rapid ascendancy merits attention for that reason alone, but it does so even more when it manages to capture the attention and mobilise some of the leading academics in a given legal field – be it in constitutional theory, European integration, or international and global relations. As an edited volume, this book’s purpose is not to provide a single valid exegesis of the evolution and the present state of the idea of constitutional pluralism. Its objective rather lies in demonstrating how different authors have understood and used constitutional pluralism in different contexts. For that purpose we invited a group of scholars who are engaging in this debate to a conference, which took place in Oxford in March 2009.2 What some of those invited present as ‘constitutional pluralism’ can eventually seem not be ‘pluralist’ or ‘constitutional’ at all. The book therefore offers a critical perspective on the current state of the idea of constitutional pluralism. In order to help the reader through the maze of different theories and to allow a better understanding and critical engagement with the contents of this book, this introduction provides a short sketch of what can be considered the essentials of constitutional pluralism.

1   J Baquero Cruz, ‘The Legacy of the Maastricht-Urteil and the Pluralist Movement’ (2008) 14 European Law Journal 389. 2   Constitutional Pluralism in the European Union and Beyond, Oxford, 20–21 March 2009.

2  Matej Avbelj and Jan Komárek

I.  The Birth of an Idea The idea of constitutional pluralism first emerged in the context of European integration. Neil MacCormick can be named as its ‘founding father’. In a very short article3 published in the aftermath of the German Federal Constitutional Court’s Maastricht decision,4 MacCormick broke apart from the then conventional wisdom. According to a then common view, Community law was supposed to have absolute supremacy over the Member States’ legal orders and the pronunciations of the European Court of Justice were to be respected unconditionally. Any reservations on the side of the national courts were deemed, almost a priori, unfounded and short-lived.5 However, as the German Federal Constitutional Court’s Maastricht decision – and the reaction which this decision provoked on the part of other actors, including other national highest courts – showed, a reversal of the national courts’ attitude was nowhere in sight. To the contrary, the courts even reinforced their stance. MacCormick thus sincerely engaged with their concerns and concluded that despite all criticism, the Maastricht decision had a ‘sound basis in legal theory’.6 The basis consisted in a pluralistic legal theory,7 following which the most appropriate analysis of the relations of legal systems is pluralistic rather than monistic, and interactive rather than hierarchical. The legal system of Member States and their common legal system of EC law are distinct but interacting systems of law, and hierarchical relationships of validity within criteria of validity proper to distinct systems do not add up to any sort of all-purpose superiority of one system over another.8

MacCormick’s article effectively marked the beginning of a theory of constitutional pluralism, save for calling it by this name. However, even before the notion of ‘constitutional pluralism’ was first employed, other authors too had started unveiling a more intricate internal legal structure of European integration9 and were trying to devise a set of principles that could enable the national courts to legitimately deviate from the supremacy of European Community law.10   Neil MacCormick, ‘The Maastricht Urteil: Sovereignty Now’ (1995) 1 European Law Journal 259.   Case 2 BvR 2134, 2159/92 Maastricht [1993] BVerfGE 89, 155, published in English as [1994] 1 CMLR 57. 5   For a classical voice in this respect see Eric Stein, ‘Toward Supremacy of Treaty-Constitution by Judicial Fiat: On the Margin of the Costa Case’ (1965) 63 Michigan Law Review 491. Baquero Cruz, n 1, provides perhaps the most forceful current expression of these ideas. 6   MacCormick, above n 3, at 265. 7   Formulated in the spirit of legal institutionalism developed earlier by MacCormick, first in a book written together with Ota Weinberger, An Institutional Theory of Law: New Approaches to Legal Positivism (Kluwer Academic Publishers, Dordrecht, 1986). See also N MacCormick, Questioning Sovereignty: Law, State and Nation in the European Commonwealth (Oxford, Oxford University Press, 1999) 1–15 (Chapter ‘The Legal Framework: The Normative Institutional Order’). 8   MacCormick, above n 3, at 265. 9   See particularly JHH Weiler, The Constitution of Europe: ‘Do the New Clothes Have an Emperor?’ and Other Essays on European integration (Cambridge, Cambridge University Press, 1999). 10   Mattias Kumm, ‘Who is the Final Arbiter of Constitutionality in Europe? Three Conceptions of the Relationship between the German Federal Constitutional Court and the European Court of Justice’ (1999) 36 Common Market Law Review 351. 3

4

Introduction  3 A theory of constitutional pluralism was first explicitly addressed by its name, again by Neil MacCormick, in his book entitled Questioning Sovereignty.11 The language used there, however, was still slightly ambiguous, as constitutional pluralism was at times assimilated with other notions such as normative, legal or juridical pluralism.12 Nevertheless, this work already provided for a vague definition of constitutional pluralism that was portrayed as a situation of plurality of institutional normative orders, each with a functioning Constitution, whereby each acknowledges the legitimacy of every other within its own sphere, while none asserts or recognises constitutional superiority over another.13 Constitutional pluralism as applied to European integration, MacCormick later averred, entails a strongly decentralised conception of the Union whose constitutional architecture is much closer to confederation than to federation.14 The theory of constitutional pluralism was therefore initially limited to European integration. However, in 2002 Neil Walker suggested that constitutional pluralism is a wider phenomenon, which extends beyond it.15 Walker contended that we are facing a transition away from the old one-dimensional Westphalian world. The state is no longer an exclusive unit of legal and political organisation. The link between the autonomy and territorial sovereignty has been severed and the exclusivity of the territorially limited claims towards ultimate legal authority have given way to the competition of various functionally underpinned but equally plausible claims towards ultimate legal authority. Alongside the state, new sectorally and functionally oriented polities have been emerging. The European Union was one of them, perhaps the most developed, but certainly not the only one. New forms of legal and political community can be found on the sub-state, trans-state, supra-state and on other non-state levels.16 The theory of constitutional pluralism has an important role to play in this post-Westphalian world. It carries a promise of a better explanation of the new world order: ‘We can only begin to account adequately for what is going on . . . if we posit a framework which identifies multiple sides of constitutional discourse and authority’.17 It contains a normative vision of the new world order, which could be a better place than the old Westphalian configuration, built on mutual recognition and respect between the many emerging authorities. However, for that purpose these newly emerging authorities have to be taken for what they are: different epistemic sites that have in common no neutral perspective from which different representational claims to authority could be reconciled.18   MacCormick, above n 7, at 104.   Ibid 102.   Ibid 104. 14   Neil MacCormick, ‘A Comment on the Governance Paper’, Jean Monnet Working Paper No 6/01, ‘Symposium: Mountain or Molehill? A Critical Appraisal of the Commission White Paper on Governance’, http://centers.law.nyu.edu/jeanmonnet/papers/01/012501.html. 15   Neil Walker, ‘The Idea of Constitutional Pluralism’ (2002) 65 Modern Law Review 317. 16   Ibid 320. 17   Ibid 337. 18   Ibid 338. 11 12 13

4  Matej Avbelj and Jan Komárek With Walker constitutional pluralism is thus no more limited to European integration only. It can also account for, explain and normatively guide other socio-political phenomena, trends and developments characteristic of the postWestphalian world. However, beside the change of the scope of the theory of constitutional pluralism, which is hereinafter converted from the European-integration specific to a general theory, its original objective is altered too. If MacCormick largely saw constitutional pluralism as a device that can provide a sound(er) basis for thinking of, practising, and therefore functioning of the European integration, Walker primarily conceives of constitutional pluralism as a means of rehabilitation of the constitutional language.19 The diagnosis is that the changing world of non-exclusivity, fragmentation, heterarchy and of increasingly permeable boundaries between different sites and authorities has debased the traditional statist foundations of constitutionalism and therefore subjected it to an unprecedented strain. Constitutionalism has come under the risk of marginalisation, of losing its relevance, precisely at the stage when its values and social-­ engineering capacities are needed perhaps more than ever before. The world pervaded by plurality also requires a minimum degree of coherence and, more importantly, it calls for a meta-language through which the actors situated at different (epistemic) sites could reflexively engage with each other by recognising their differences with a simultaneous commitment to a certain shared framework of coexistence. Constitutionalism has proved successful as such a language in the statist environment and absent other alternatives, which are still nowhere emerging on the horizon, it should also continue to play the same role in the new pluralist world order beyond the state, provided it undergoes some pluralist transformation.20

II.  The Spread and Diversification of the Idea However, as a result of the launch of the process of adopting a formal Constitution of the EU (‘documentary constitutionalism’), the focus of constitutional pluralism again reverted to Europe, whereas its broader, potentially more general theoretical implications have for now remained far less developed. Inside the European Union’s debate constitutional pluralism has, however, drawn new supporters, but in so doing it has also become a more diversified and heterogeneous idea. Different authors are still referring to the nominally identical concept, but effectively they often harbour a different, sometimes also very different, understanding of what the latter concretely entails. Consequently, several different conceptions of European constitutional pluralism have gradually emerged. Six of them have been more prominent, developed and therefore influential.21   Ibid 317.   See, eg, N Walker, ‘Beyond the Holistic Constitution?’ in P Dobner and M Loughlin (eds), The Twilight of Constitutionalism? (Oxford, Oxford University Press, 2010) 291–308. 21   This classification follows M Avbelj, ‘Questioning EU Constitutionalisms’ (2008) 9 German Law Journal 1. 19 20

Introduction  5 The first is socio-teleological constitutionalism developed by Joseph Weiler. His special brand of European constitutionalism has three dimensions. Following the first, formal dimension, the European Union already has a Constitution, developed in the interaction between the national judicial as well as political branches, and therefore does not need a special, documented one, resembling the statist Constitution. Pursuant to the second, normative dimension, which is also the key pluralist component in this conception, the Union is founded on the principle of constitutional tolerance, which sends a deeply normative message of necessity and desirability of mutual recognition between the self-reflexive individuals and the Member States in their eternal pursuit of a decent life. Third, there is a sociological dimension, whereby constitutional tolerance is being exercised on a daily basis between all the actors of the integration: from the lowest-ranked official, to the highest judicial authority.22 The epistemic meta-constitutionalism, suggested by Neil Walker, offers constitutionalism, redefined in a pluralist way, as a meta-framework above and beyond the constituent entities of European integration. It similarly insists on the need of fostering dialogue, mutual-learning and cross-fertilisation between them,23 but it admits that due to a distinct epistemic status of each of the entities involved the bridging-mechanisms between them cannot be unlimited.24 Also, in contrast to Weiler’s conception, the epistemic meta-constitutionalism distances itself more from the classical hierarchical constitutional structure, while at the same time, largely because of the potentially positive effects a Constitution-making process could engender for the integration, remains in favour of the European Union’s own ‘documented’ Constitution.25 This is something that the best fit universal constitutionalism, recently dubbed cosmopolitan constitutionalism,26 would be also willing to concede to. This brand of constitutional pluralism also recognises the pluralist structure of European integration, but it considers its plurality, different as the epistemic meta-constitutionalism, to be situated in a universal framework of substantively homogeneous shared 22   JHH Weiler, ‘In Defence of the Status Quo: Europe’s Constitutional Sonderweg’ in JHH Weiler and M Wind (eds), European Constitutionalism Beyond the State (Cambridge, Cambridge University Press, 2003) 23. The idea of constitutional tolerance has won broad support among scholars; see, eg, S Oeter, ‘Federalism and Democracy’ in A von Bogdandy, J Bas (eds), Principles of European Constitutional Law (Oxford, Hart Publishing, 2006) 95; P Eleftheriadis, ‘The Idea of a European Constitution’ (2007) 1 Oxford Journal of Legal Studies 1, 16; C Offe and UK Preuss, ‘The Problem of Legitimacy in the European Union: Is Democratization the Answer?’ in C Crouch and W Streech (eds), The Diversity of Democracy: Corporatism, Social Order and Political Conflict (Cheltenham, Edward Elgar, 2006). 23   Neil Walker, ‘Flexibility within a Meta-Constitutional Frame: Reflections on the Future of Legal Authority in Europe’ in G de Búrca and J Scott (eds), Constitutional Change in the EU: From Uniformity to Flexibility? (Oxford, Hart, 2000). 24   Walker, n 15, 338. 25   N Walker, ‘Europe’s Constitutional Engagement’ (2005) 18 Ratio Juris 387, 398. 26  M Kumm, ‘The Cosmopolitan Turn in Constitutionalism: On the Relationship between Constitutionalism in and beyond the State’ in J Dunoff and J Trachtman (eds), Ruling the World: Constitutionalism, International Law and Global Governance (Cambridge, Cambridge University Press, 2009).

6  Matej Avbelj and Jan Komárek principles and values of political liberalism, which lie at the base of the modern constitutionalism. The latter provides a universal language against which the competing claims of the entities constituting European integration can be measured and balanced so as to find the best fit solution for the integration as a whole.27 Another version of European constitutional pluralism, harmonious discursive constitutionalism, closely resembles the best fit universal constitutionalism in welcoming and preserving plurality in the European Union, but only if this stays within certain manageable bonds that still ensure harmony of a system.28 The two differ, however, on the one hand as to the intensity of the plurality and its recognised implications, as well as, on the other hand, to the degree of universalism. The harmonious discursive constitutionalism is slightly more disposed in favour of plurality. It refrains from making (strong) claims about the actual substantive universality of principles and values, and insists only on the procedural dimension of universalisability of arguments through which actors across different entities of the integration justify their claims to authority.29 Constitutional pluralism in this sense therefore provides for a shared discursive, for example procedural framework, rather than for universally shared substantive foundations. Another branch of constitutional pluralism, defended mainly by German scholars and in particular by Ingolf Pernice, has been the so-called multilevel constitutionalism.30 It also proceeds from the presumption of a considerable degree of substantive unity and homogeneity of values between the constituent entities of the integration, which can therefore function as a composite of two independent constitutional lawyers, national and supranational, that are nonetheless forming part of a single European Constitution. Multi-level constitutionalism presupposes one European sovereign as well as a single answer in any constitutional conflict that might arise.31 In so doing, it most closely approaches the classical, for example non-pluralist, constitutional account. Exactly the opposite is attempted by the last of the six forms of EU constitutional pluralism: the pragmatic constitutionalism. This argues that the classical constitutional paradigm should be abandoned in its entirety, along with its sovereignty conundrum, pursuit of universality, coherence and integrity. The constitutional language should switch from the whole to the particular, from the constitutionally holistic to the constitutionally atomistic approach. The European integration should be, accordingly, completely reconstructed and established as a directly deliberative polyarchy,32 characterised by a pragmatic, experimentalist 27   M Kumm, ‘The Jurisprudence of Constitutional Conflict: Constitutional Supremacy in Europe before and after the Constitutional Treaty’ (2005) 11 European Law Journal 262, 292. 28   M Poiares Maduro, ‘Contrapunctual Law: Europe’s Constitutional Pluralism in Action’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 502. 29   Ibid 525. 30   I Pernice, ‘Multilevel Constitutionalism in the European Union’ (2002) 27 European Law Review 511, 514. 31   Ibid 518–19. 32   O Gerstenberg and CF Sabel, ‘Directly-Deliberative Polyarchy: An Institutional Ideal for Europe?’ in C Joerges, R Dehousse (eds), Good Governance in Europe’s Integrated Market (Oxford, Oxford University Press, 2002) 292.

Introduction  7 approach to governance involving a range of private and public actors entangled into a whole array of policy networks.33 The idea of constitutional pluralism has not remained tied to the ambit of scholarship only; its presence can also be increasingly traced in judicial practice, both national and supranational. Two examples should serve to illustrate this point. The German Constitutional Court in its much criticised Lisbon decision recognised the legal autonomy of EU law,34 but it also reserved for itself the right to exceptionally disapply EU law, albeit under special and narrow conditions.35 The Court, inter alia, argued that this is justified in the contexts of political order which are not structured according to a strict hierarchy.36 The same approach was adopted by the Court of Justice in relation to international law. In Kadi the Court, contrary to the conspicuously monist approach to international law followed by the Court of First Instance (renamed by the Lisbon Treaty as ‘General Court’),37 asserted the autonomy of the Union legal system and declined the possibility of hierarchical supremacy of international law over it. It treated Union law and international law as two distinct legal orders. In their relationship international law in principle enjoys primacy, but it must satisfy certain basic requirements of the Union legal order, such as protection of fundamental human rights, in order to be granted that effect.38 These and other examples of national and supra­ national judicial practice39 further illustrate that courts also to a certain extent subscribe to constitutional pluralism.

III.  The Theoretical Issues and Dilemmas In the last 10 years or so the idea of constitutional pluralism has therefore apparently made substantial progress: from a reaction to what was deemed a problematic practice in the mid 1990s it was transformed into a theory which has subsequently grown more sophisticated as well as diversified and is now influ­ encing the practice. Constitutional pluralism has thus emerged in the form of a virtuous hermeneutical circle, which is inspired by practice and via theoretical refinement influences it back. Yet, this hermeneutical circle has never been 33   G Marks, L Hooghe and K Blank, ‘European Integration from the 1980s: State-Centric v MultiLevel Governance’ (1996) 34 Journal of Common Market Studies 341; G Marks (ed), Governance in the European Union (London, Sage, 1996). 34   Case 2 BvE 2/08 et al Lisbon [2009], paragraph 333. 35   Ibid, paragraph 340. 36   Ibid. 37   Case T-315/01 Kadi v Council and Commission [2005] ECR II-3649. 38   Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat International Foundation v Council and Commission [2008] ECR I-6351, in particular paragraphs 282, 307–9, 317, 326, 327. 39  See, in particular, Spanish Constitutional Court, Declaration of 13 December 2004, 1/2004, [2005] 3 BOE 5. See also See F Castillo de la Torre, ‘Tribunal Constitucional (Spanish Constitutional Court), Opinion 1/2004 of 13 December 2004, on the Treaty Establishing a Constitution for Europe’ (2005) 42 Common Market Law Review 1169.

8  Matej Avbelj and Jan Komárek without problems, let alone criticism. It has raised strong theoretical and philosophical dilemmas even among its proponents. Many of them continue to await, if not a resolution, than at least a better understanding of the challenges they spur. Constitutional pluralism is widely regarded as presenting a paradigm shift; however, absent agreement on whether this shift is for better or for worse. The sceptics have blamed constitutional pluralists for justifying national constitutional courts’ diversions from the clear and precise requirements of European Union law, assisting them in turning the ‘real world into a fable’.40 The opposite side reacted that the so-called ‘real world’ has, perhaps, never been what it was thought to be and that it is only now with the help of constitutional pluralism that we are pulling our heads out of the sand,41 being faithful to ourselves in recognising that the classical story does not fit the actual practice and that it even fails to be convincing on its own terms.42 However, this paradigm shift has also been seen as going beyond the context of European integration. Constitutional pluralism is said to be introducing postmodernity into the sphere of law, which has been one of the last strongholds of modernity. This could be self-defeating for law and for lawyers.43 Fluidity, fragmentation, and law in a constant flux could come with too high a price in terms of clarity, certainty and effectiveness.44 Recognising this, however, the proponents of constitutional pluralism do not necessarily conceive of it as a post-modern enterprise. It is in fact the constitutional pluralism’s constitutional side that preserves the modernist virtues and tries to make them fit our changing world in which, unlike in the past, the overlap between different sites, authorities, jurisdictions and peoples has become endemic.45 Constitutional pluralism in that way serves as a balanced approach, as a via medium between the chaos and fragmentation of post-modernity, which is feared coming, and the slowly fading watertight certainty, coherence, uniformity and effectiveness of modernity.46 Constitutional pluralism is furthermore torn between a number of methodological and philosophical binary choices. Is it a descriptive or a normative theory? Do we merely describe the world and discover plurality, or do we also justify it as something that is good and ought to be preserved?47 This relates to a distinction, recognised by some and disregarded by others, between plurality, as a social fact of existence of a multitude of sites and authorities, and pluralism, as a normative fabric which recognises and accepts this plurality and tries to galvanise it into an operational whole.48 To the extent that constitutional pluralism lays emphasis   Baquero Cruz, n 1, 389.   N Walker in M Avbelj, J Komárek (eds), ‘Four Visions of Constitutional Pluralism’ (2008) 2 European Journal of Legal Studies 325, 334. 42   M Kumm in Avbelj and Komárek, above n 40, at 355. 43   J Baquero Cruz, in Avbelj and Komárek, above n 40, at 332. 44   Ibid 333. 45   N Walker in Avbelj and Komárek, above n 40, at 334; M Poiares Maduro in Avbelj and Komárek, above n 40, at 337. 46   Ibid 345. 47   For these queries and attempted answers, see Avbelj and Komárek, above n 40, at 328. 48   N Walker in Avbelj and Komárek, above n 40, at 336. 40 41

Introduction  9 on the plurality or on the common whole, that is, to what is different and what is shared, scholars have tried to assimilate it both in the universalist-particular and in the communitarian-cosmopolitan dichotomy. In so doing, they would occupy a different position on the pluralist-monist continuum, which ranges from radical pluralism on the one end, via moderate pluralism, to monism on the other end, with a number of pluralist or monist variants in between. As will be apparent, contributions to this book are no different in this respect.

IV.  What this Book Offers There is perhaps no better person to open the discussion on constitutional pluralism than Neil Walker. His Chapter presents three challenges that constitutional pluralism faces today and discusses how these challenges materialise in the European and global context. The first concerns the question of whether the normative assumptions of constitutionalism, its emphasis on hierarchically ordered authority, uniformity and finality, do not deny pluralism. In Walker’s words, ‘constitutional pluralism may, on closer inspection, simply mutate and settle into a new form of constitutional monism or singularity’.49 The second challenge contends that there is just a plurality of distinct and unconnected, bounded and hierarchically ordered constitutional entities while ‘we lack a constitutional code that operates independently of [them]’.50 Constitutional pluralism, according to this challenge, brings nothing but a series of reductions. The third challenge suggests that, in order to develop a meaningful framework that would enable us to grasp the pluralist reality, we must ‘dispense with the constraining and increasingly anachronistic language of constitutionalism as an appropriate characterization of such entities’.51 We should therefore speak of pluralism without the adjective ‘constitutional’. In the rest of the Chapter Walker discusses these three challenges in the European and global context. While he suggests that the former offers ‘a kind of regional comfort zone for the ideas of constitutional pluralism’, the latter context poses the challenges ‘more sharply and insistently’.52 In the end, Walker hesitates whether constitutional pluralism provides a useful framework to grasp relationships existing in the global world: Constitutional pluralism, conceived of as idea of a constitutionally relevant connection between self-authorizing constitutional sites, silently assumes something like the statist template of constituent power as the legitimate basis for the self-authorization of the post-national constitutional sites. If self-authorization increasingly lacks that legit­ imation, however, the focus of our concern shifts to the broader question of what form   N Walker’s Chapter in this volume, 18.   Ibid 19. 51   Ibid 20. 52   Ibid 25. 49 50

10  Matej Avbelj and Jan Komárek of legitimation is possible in place of or in supplementation of site-specific self-­ authorization.53

At the same time, however, Walker admits that if we think of constitutional pluralism . . . as referring to the continuing relevance of constitutionalism in addressing the mix of empirical and normative factors which contribute to the deep pluralism of the emerging global order, then it certainly remains a relevant conceptual point of departure.54

Mattias Kumm examines various conceptual frameworks within which one can think of the question of ‘[u]nder what circumstances [it is] appropriate to conceive of the relationship between different legal orders in pluralist terms, rather than thinking about them in terms of hierarchical integration?’55 Kumm suggests that there are three such frameworks: ‘legal (or legalist) monism’, ‘democratic statism’ and finally, ‘cosmopolitan constitutionalism’. The first ‘is sceptical about any kind of legal pluralism and can analyse it only as a case of law in crisis’.56 According to Kumm, judgments of the Court of Justice in Costa and Kadi can be read in this way.57 The second is epitomised by the German Constitutional Court’s judgments concerning European integration, but as Kumm shows, its traces can also be found in the Court of Justice’s decision in Costa. However, Kumm argues that all these decisions can be read through the lenses of cos­mopolitan constitutionalism too. Within this framework, the refusal of a legal order to recognize itself as hierarchically integrated into a more comprehensive legal order is justified, if that more comprehensive order suffers from structural legitimacy deficits – relating to human rights protection of democratic legit­ imacy – that the less comprehensive legal order does not suffer from. The concrete norms governing the management of the interface between legal orders are justified, if they are designed to ensure that the legitimacy conditions for liberal-democratic governance are secured.58

According to Kumm, cosmopolitan pluralism thus provides normative criteria to decide the question he sets at the beginning of his Chapter. Miguel Poiares Maduro is also strongly related to the development of the idea of constitutional pluralism. His Chapter is a continuation of his works: constitutional pluralism can be conceived ‘not only as remedy for constitutional conflicts of authority, but as the theory that can best embrace and regulate the nature of the European Union polity’.59 In that relation Maduro presents three claims that constitutional pluralism makes and defends them against recent criticisms. The first,   Ibid 37.   Ibid 54. 55   M Kumm’s Chapter in this volume, 41. 56   Ibid 42. 57   Case 6/64 Costa [1964] ECR 585 and Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat International Foundation v Council and Commission [2008] ECR I-6351. 58   M Kumm’s Chapter in this volume, 43. 59   M Poiares Maduro’s Chapter in this volume, 68. 53 54

Introduction  11 empirical claim, asserts that ‘constitutional pluralism is what best describes the current legal reality’.60 Maduro directly responds to Alexander Somek’s doubts raised as regards this claim in his contribution to this very volume61 and also to arguments based on either explicit national constitutional amendments, which would provide evidence ‘of national constitutional supremacy that would, otherwise, prevent the entry into force of the new Treaty’62 or on those who claim that in fact supremacy of EU law is now the rule. The second and third claims are normative, either in a weaker form, related closely to the EU, or a thick one, according to which constitutional pluralism is ‘the best representation of the ideals of constitutionalism for the current context of increased pluralism and deterritorialization of power’.63 According to Maduro: To understand this we need to start by recognising that pluralism is inherent in constitutionalism. In fact, constitutionalism aims to simultaneously guarantee and regulate such pluralism: a pluralism of interests, ideas and visions of the common good that is reflected in the paradoxes of constitutionalism and the balance between democratic deliberation and constitutional rights that its modern liberal form embraced.64

Daniel Halberstam responds in the following Chapter to some of the challenges identified by Walker by reinterpreting ‘the traditional view of constitutional law as the complete consolidation and hierarchical ordering of legal authority’ in pluralist terms. This ‘may produce a constitutional practice that is more true to the ideals of constitutionalism than the traditional model of consolidation and hierarchy itself’.65 Halberstam distinguishes between systems pluralism and institutional pluralism. The former entails a simultaneous conflict of final legal authority between two overlapping legal systems, which however ‘contains an inherent openness to the claim of authority of the other’.66 Contrary to what some people think, including probably Neil Walker, Halberstam suggests that this ‘mutual embedded openness’ exists not only in the European Union, but also on the international level, in the form of some universalisable principles. According to Halberstam: Claims of authority from beyond the system must give voice to a legitimate political will, respect or advance the rights of individuals and groups, and promote the effectiveness of governance. Unless the claim from the outside can be made in these terms, it will fall on deaf ears.67

Institutional pluralism, contrary to the Montesquieuian conception of separation of powers, whereby legislative, executive and adjudicative functions are separated, protects liberty through granting the same actors the same kind of legal authority   Ibid 70.   See 73. 62   M Poiares Maduro’s Chapter in this volume, 73. 63   Ibid 78. 64   Ibid (references omitted). 65   D Halberstam’s Chapter in this volume, 86. 66   Ibid 95. 67   Ibid 109. 60 61

12  Matej Avbelj and Jan Komárek to interpret the foundational framework – the Constitution. This kind of pluralism is best illustrated by the example of the United States Constitution, but Halberstam shows that it can be found in other systems too. According to Halberstam, the openness and lack of settlement brought about by both forms of pluralism promotes and does not contradict the constitutional idea of limited collective self-governance through law – be it on the national, supranational or global level. Franz Mayer and Mattias Wendel present in Chapter 6 the concept of multilevel constitutionalism developed by Ingolf Pernice in response to the German Federal Constitutional Court’s Maastricht decision.68 They suggest that multilevel constitutionalism is largely compatible with constitutional pluralism, since according to them, ‘the core assumption of [both] is the existence of multiple, independent and incommensurable claims of constitutional autonomy’.69 They argue that quarrels raised in Germany against the concept are not limited to that country – they are not just querelles allemandes. At the same time, they are still meaningful – they are not quarrels about nothing, or, as the French expression goes, they are not querelles d’Allemand. René Barents, however, argues in the following Chapter that the theory of multilevel constitutionalism is a fallacy. In his opinion, it is based on dubious assumptions and suffers from serious contradictions. Although it was developed in response to theories that tie the Constitution together with the state, which informed the Maastricht decision, they are both based on democratic constitutional ideology. Where the latter is ‘based on the will of the democratic states as the supreme source of politics and law, in the [multilevel constitutionalism] theory this foundation is exchanged for the hypothetical common will of the citizens’.70 However, in Barents’ opinion, ‘[n]either the will of the democratic Member States nor the democratic will of the citizens can provide for a sound theoretical explanation of what the Union is’.71 Robert Schütze moves the debate on constitutional pluralism from the European context to the other side of the Atlantic. In Chapter 8 Schütze goes into the history of the United States Constitution in order to show ‘that constitutional pluralism is an inherent characteristic of American federalism’72 based, contrary to Europe, on an idea of divided sovereignty. Schütze discusses four episodes concerning identity of the final arbiter in conflicts concerning the scope and interpretation of the federal Constitution and claims that ‘[t]he theory of constitutional pluralism speaks federal prose, without – as Molière’s Monsieur Jourdain – being aware of it’.73 68   German Federal Constitutional Court, Case 2 BvR 2134, 2159/92 Maastricht [1993] BVerfGE 89, 155, published in English as [1994] 1 CMLR 57. 69   F Mayer and M Wendel’s Chapter in this volume, 135. 70   Barents’ Chapter in this volume, 182. 71   Ibid. 72   R Schütze’s Chapter in this volume, 187. 73   Ibid, footnote omitted.

Introduction  13 In Chapter 9 Ola Zetterquist argues that the foundations of the Union, as expressed in the case law of the Court of Justice, have moved from a pacta sunt servanda principle, centred on the state, to a res publica approach, centred on the individual and her rights, echoing ‘Cicero’s definition of the republic as a community based on legal agreement and community of interest’.74 Understanding the Union in republican terms has several advantages according to Zetterquist: Firstly, it places the emphasis on individuals rather than on states . . . Secondly, it focuses on substance rather than form and leads to value hierarchies rather than normhierarchies . . . Thirdly, it broadens the concept of voice to include the legal domain in addition to the political . . . Finally, a republican lens provides a way out of the sovereignty trap where sovereignty is essentially a zero-sum game: either the Member States or the EU has it. These three elements combined form building stones for a pluralistic European republican constitution.75

Jan Komárek’s Chapter discusses a particular feature of constitutional pluralism – its capability of making institutional choice inherent in the constitutional form of government, visible. Constitutional pluralism, through its contestation of finality and conclusiveness, highlights the role of particular institutions which take decisions of constitutional significance. Komárek argues that this institutional dimension is one of constitutional pluralism’s principal virtues. Most theori­es of European constitutionalism do not recognise this and leave important questions unexplored. Those that do so focus too much on conflict and choice. Komárek proposes the reorientation of institutional analysis towards commun­ ication and involvement among the institutions. Julio Baquero Cruz’s Chapter shares with the previous one the emphasis on institutions. Baquero Cruz, who is marked out by many contributors to this volume as a representative of the ‘hard-line approach’ to the European Union’s constitutional authority, seeks to interpret occasional instances of national constitutional courts’ reservations to European Union law as ‘as exceptional instances of institutional disobedience rather than as examples or a more or less virtuous general practice of legal or constitutional pluralism’.76 Baquero Cruz first explores various conceptions of the relationship between Union and national law: ‘state-centred’, ‘constitutionalism beyond the state’, and legal or constitutional pluralism. As regards the last of them, Baquero Cruz observes that the clashes concerning the final authority, which are supposedly at the heart of pluralist approaches, are quite rare. Thus, to Baquero Cruz, they are expressions of disobedience, not from the citizens, but from institutions, which under certain – exceptional – circumstances can be legitimate. Gareth Davies in Chapter 12 also proposes a framework through which constitutional conflicts in the European Union could be resolved in a pluralist fashion, through balancing and the proportionality principle:

  O Zetterquist’s Chapter in this volume, 227.   Ibid 229. 76   J Baquero Cruz’s Chapter in this volume, 249. 74 75

14  Matej Avbelj and Jan Komárek States often have to adapt their policies and institutions to comply with EU law. That is an unavoidable consequence of membership. However, if the application of EU rules makes achievement of important and legitimate national policy preferences effectively impossible, or unreasonably difficult, then, depending upon the degree of EU or other interest in full application of that rule, it may be disproportionate to apply the rule in the particular context in question. States must provide evidence that amending their systems or policies to achieve their goals in a way compatible with EU law would either be unreasonably difficult or disproportionately harmful to other interests.77

Daniel Sarmiento in Chapter 13 offers a fresh look at one of the bridging mechan­ isms78 in the pluralist settings of the European Union – the preliminary ruling procedure and a particular strategy which European courts employ to preserve pluralism: ‘silent judgments’ by the Court of Justice and responses by national courts that can be ‘deaf’ to these silent judgments to various degrees. Sarmiento assesses the virtues but also the vices of such strategies and proposes to interpret Union law ‘silently, but not in silence’. It means adopting the famous CILFIT criteria on last instance courts’ duty to refer to the Court of Justice in order to give more freedom to national courts while at the same time taking this duty more seriously. Xavier Groussot’s Chapter 14 also explores constitutional dialogues in the European Union – the communication between national constitutional courts (or national highest courts with jurisdiction over constitutional matters) and the Court of Justice. It identifies several ways of communication beyond the preliminary ruling procedure and also its substantive boundaries lying primarily in national constitutional identities. Groussot submits that a ‘doctrine of deference’ relating to the deep national interests has emerged in the Court’s jurisprudence, which should provide the Court with legal tools to respect these identities. At the same time Groussot identifies several ways in which the Court’s functioning could be improved to enhance constitutional dialogue, especially through its greater openness. In some Chapters of this volume monism is seen as an outdated concept, while constitutional pluralism should provide a more accurate theoretical position, both descriptively and normatively. Alexander Somek does not think so and in Chapter 15 explains why. Somek argues that pluralism and cosmopolitanism, which in his opinion accompanies it, do not offer new alternatives to presumably outdated concepts of dualism and monism. According to Somek: If defensible, pluralism, either in simple or cosmopolitan form, formulates a new version of monism; if not, it amounts to a travesty of constitutional ideas, which assimilates legality to the mindset of administrative action . . . [C]hoosing one construction over the other is a question of political philosophy. Monism commends itself not least owing to its superb constitutionalist sensibilities.79   G Davies’ Chapter in this volume, 281.   N Walker’s Chapter in this volume, 22. 79   A Somek’s Chapter in this volume, 343–44. 77 78

Introduction  15 Chapter 16 by Matej Avbelj closes the circle opened by Neil Walker’s contribution. Instead of reinterpreting the notion of constitutionalism in pluralist terms, as Daniel Halberstam suggests, Avbelj concludes that constitutionalism and pluralism are incompatible to such an extent that using the term ‘constitutional pluralism’ is not very helpful for understanding European integration. After defining the two concepts Avbelj critically assesses whether constitutionalism can be ‘translated’ into the context of the pluralist European integration. Despite a number of arguments raised in favour of such translation, Avbelj suggest ‘innovation instead of translation: thou shalt not be called constitutionalism’. However, we are quite sure that while Avbelj’s Chapter concludes this volume, it does not conclude the debate on constitutional pluralism – either in a European or in a global context. We hope that this volume will be an invitation to further exploration of questions which constitutional pluralism seeks to address.

2 Constitutionalism and Pluralism in Global Context NEIL WALKER

I.  Three Forms of Scepticism about Constitutional Pluralism Constitutional pluralism divides opinion. Those features that make it attractive to some in a globally connected world also account for the scepticism it provokes in others. The allure of constitutional pluralism lies in its ambition to square two ideas – ‘constitutionalism’ and ‘pluralism’ – that are typically understood as quite distinct and presumptively incompatible, or at least as of limited compatibility. On the one hand, the idea of constitutionalism – of a legal code that supplies a legitimate foundation and framework for our common forms of political life – has been traditionally understood in unitary and hierarchical terms. That is to say, it is taken to refer to a single, bounded, and ultimately indivisible ‘unit’ – paradigmatically the state – and to do so in terms of an unbroken chain of authority and an encompassing legal ordering.1 On the other hand, when we speak of pluralism, whether we are concerned with a ‘first order’ pluralism of social constituencies,2 or of institutions,3 or of values,4 or of value sets and world-views,5 or – of most direct immediate relevance – with a ‘second order’ pluralism of legal and political systems as a whole, the emphasis is always upon multiplicity and diversity and upon the non-hierarchical terms of the recognition and accommodation of that multiplicity and diversity. In crude terms, the constitutional pluralist seeks to 1   See, eg, D Grimm, ‘The Constitution in the Process of Denationalization’ (2005) 12 Constellations 447; M Kumm, ‘The Cosmopolitan Turn in Constitutionalism: On the Relationship between Constitutionalism in and beyond the State’ in JL Dunoff and JP Trachtman (eds), Ruling the World? Constitutionalism, International Law, and Global Governance (Cambridge, Cambridge University Press, 2009). 2   See, eg, RA Dahl, ‘The Concept of Power’ (1957) 2 Behavioural Science 201, 201–15. 3   See, eg, V Bader, ‘Religious Diversity and Democratic Institutional Pluralism’ (2003) 31 Political Theory 265. 4   See, eg, I Berlin, Four Essays on Liberalism (Oxford, Oxford University Press, 1969). 5   See, eg, J Rawls, Political Liberalism (New York, Columbia University Press, 1996); R Bellamy, Liberalism and Pluralism: Towards a Politics of Compromise (London, Routledge, 1999).

18  Neil Walker retain from constitutionalism the idea of a single authorising register for the political domain as a whole while at the same time retaining from pluralism a sense of the rich and irreducible diversity of that political domain. For the advocate of constitutional pluralism, moreover, the attraction is a matter both of fact and of value – of the force of circumstance as well as of preference. The fact that the constitutional landscape today – in our post-Westphalian age where globalising economic, cultural, communicative, political and legal influences have both spread and diluted public power – is no longer organised into mutually exclusive nation-state domains but instead occupies much overlapping transnational space, cannot help but alter our understanding of constitutional ordering. It means that, at least as the constitutional pluralist views the world, it becomes increasingly difficult if not impossible not to conceive of the environment of constitutionalism in non-unitary terms – as a place of heterarchically interlocking legal and political systems.6 The dimension of value lies in viewing this changing landscape not as a threat to the maintenance of the traditional template of constitutionalism but as a welcome opportunity to integrate what in conventional constitutional wisdom tend to be treated as contrasting and even opposing modalities of normative thought. The constitutional pluralist, in short, seeks to make a virtue out of necessity. For the sceptic, on the other hand, any such sense of opportunity can only be the product of wishful thinking. Rather than achieving the reconciliation of opposites, constitutional pluralism is always poised to collapse under the weight of its internal contradictions. And if it does so, this will not signal a new constitutional dawn. Rather, it will imply, at best, a retreat to a state-centred constitutional orthodoxy, and, at worst, the degrading or even the exhaustion of the constitutional paradigm as a whole in the late modern age. More specifically, for the sceptic there are three potential structural weak-points, and so three points of possible implosion, within constitutional pluralism. A consideration of each allows us to introduce three key challenges. In the first place, constitutional pluralism may, on closer inspection, simply mutate and settle into a new form of constitutional monism or singularity. That is to say, the tendency towards unity and hierarchy in constitutional logic and in the constitutional mindset may be strong or even incorrigible, and if this is so then new constitutional initiatives, practices or world-views that reach into the transnational sphere will tend to adopt the form of the statist original. Whether we are talking about the constitution of the European Union, or the United Nations’ ‘world order’ constitution,7 or even the informal ‘higher order’ constitution suggested by the elevated status of certain contemporary international law norms,8 6   On the descriptive dimension of constitutional pluralism, see N Walker, ‘The Idea of Constitutional Pluralism’ (2002) 65 Modern Law Review 317. 7  See in particular B Fassbender, ‘The United Nations Charter as the Constitution of the International Community’ (1998) 36 Columbia Journal of International Law 529. 8   See, eg, E de Wet, ‘The International Constitutional Order’ (2006) International and Comparative Law Quarterly 55, 51–76.

Global Context  19 what we see wherever and whenever constitutionalism is invoked beyond the state, and whatever its ostensible commitment to openness and sustainable diversity, is a tendency towards a new manifestation of closure and a new reduction to unity; towards the old familiar of everything deemed constitutional being contained – ‘constituted’ indeed – within the one hierarchically layered legal and political system.9 There is no room in that perspective for the unresolved heterarchical configuration or the open-ended jurisdictional extension of a constitutionalism decoupled from a singular legal and political order. In the second place, and conversely, traces of constitutionalism beyond the state may be viewed not as an extension and mutation that will ultimately take the form of a new and encompassing unity, but, just as in the classic age of the Westphalian state system, as a series of separate reductions. On this view, constitutional pluralism turns out to be nothing more than constitutional plurality. That is to say, the flip-side of the structural tendency of constitutional framing to provide the bounded and hierarchically ordered legal space of the state may be that if anything is to escape such a space but still be considered as properly ‘constitutional’ in character, it can only do so on the basis of its belonging to a quite distinct and unconnected bounded and hierarchically ordered constitutional entity. For if constitutional norms operate according to a singular and hierarchical regulatory logic, then there is simply no conceptual scope for any heterarchical legal relations that operate between distinct constitutional singularities to possess their own properly and distinctly constitutional character, or at least not from the perspective of these constitutional singularities themselves. In other words, if we seek to distinguish the overlapping and interlocking of constitutional orders from mere constitutional plurality or diversity on the basis that it involves a commitment to the common recognition and accommodation – and to that extent the integrity – of the diverse parts notwithstanding their diversity, then the exhaustiveness of each of the different constitutional orders in their own terms means that we lack a constitutional code that operates independently of the overlapping and interlocking constitutional orders in which any such transversal integrity can be registered.10 Whether we are dealing with the new type of relations between the 9   See, eg, D Kennedy, ‘The Mystery of Global Governance’ in JL Dunoff and JP Trachtman (eds), Ruling the World? Constitutionalism, International Law, and Global Governance (Cambridge, Cambridge University Press, 2009); M Koskenniemi, ‘The Fate of Public International Law: Between Technique and Politics’ (2007) 70 Modern Law Review 1; E Christodoulidis, ‘Constitutional Irresolution: Law and the Framing of Civil Society’ (2003) 9 European Law Journal 401. 10   There are in fact two closely related if apparently quite distinct versions of this concern or criticism. One – closely associated with a certain type of approach which remains presumptively sympathetic to constitutional pluralism – raises the prospect that there is simply nothing left to say in constitutional, or indeed in any kind of legal terms, about the relations between constitutional orders which are each already conceived of in a bounded manner. Here, the danger is that constitutional pluralism is left conceptually barren. This so-called radical pluralist approach is further considered in Section II of the text below. A second criticism, presumptively unsympathetic to constitutional pluralism, holds that an acceptance of the pluralist scenario is likely to lead not to a conceptual void in the law, and so to a domain of non-law, but to a situation of overabundance. For if constitutional pluralism simply alerts us to a plurality of legal order unities, then rather than an absence of legal answers to difficult questions in areas of overlapping jurisdiction what we have, strictly speaking, are too many

20  Neil Walker constitutional orders of states and that of the supranational EU, or between NAFTA and the states of North America, or the UN and the states of the world, or amongst the various emergent non-state polities, or whether we revert our gaze to the ‘old-fashioned’ terms of exchange between different states themselves, therefore, on this view the idea of constitutional relations between distinct constitutional orders is simply incoherent. In the third place, if and to the extent that it is nevertheless possible to think of relations between different legal entities as pluralist in quality, and not simply collapsing into either the monolithic discipline of constitutional singularity or the mutual indifference of constitutional plurality, then this may be precisely because the entities in question do not possess or claim just such a constitutional character. If we want to conceive of different legal entities within the increasingly fragmented global archipelago as connected in ways which remain legally meaningful without these legal relations resulting in such entities being ultimately subsumed within a single legal order, the development of the requisite legal imaginary may only be possible if we dispense with the constraining and increasingly anachronistic language of constitutionalism as an appropriate characterization of such entities.11 To recap, then, constitutional pluralism may be rejected either on the basis that its pluralist credentials do not add up – that it is ultimately either monism with new horizons or mere plurality – or on the basis that if it is genuinely pluralistic then this is at the expense of its specifically constitutional quality. Taken together, these three challenges introduce a formidable range of arguments against constitutional pluralism in the new global context. In what follows, I will examine how different theories of the global regulatory configuration stand in relation to constitutional pluralism and its critique – whether as explicit advocates of one or more of the three key challenges to constitutional pluralism, or at least as assuming a position consistent with such challenges; or as taking a position that invites one or more of such challenges; or as actively addressing and responding to such challenges. Before doing so, however, I want to say something about the implications of the fact that constitutional pluralism was first developed in the European supranational theatre rather than in the wider global arena. On the one hand, the particular terms of the European debate accounted for much of the early buoyancy of constitutional pluralist thinking and for its readiness to rise to the sceptical challenge. On the other hand, by developing the theoretical perspective of constitutional pluralism in conditions that were unusually favourable, this regional concentration has skewed the terms of debate. And in so doing it has answers, each valid from its own systemic perspective. Which law happens to prevail in practice becomes a matter of circumstance rather than principle, and the law as a whole in the area of contested overlap may thus come to lack predictability or a coherent framework of justification. See, eg, J Baquero Cruz, ‘The Legacy of the Maastricht-Urteil and the Pluralist Movement’ (2008) 14 European Law Journal 389; P Eleftheriadis, ‘Pluralism and Integrity’ (2010) 23 Ratio Juris 365. 11  See in particular the work of Nico Krisch: N Krisch, ‘Global Administrative Law and the Constitutional Ambition’ in P Dobner and M Loughlin (eds), The Twilight of Constitutionalism? (Oxford, Oxford University Press, 2010); N Krisch, ‘The Case for Pluralism in Postnational Law’ LSE Legal Studies Working Papers No 12/2009, www.lse.ac.uk/collections/law/wps/wps1.htm.

Global Context  21 retarded – or at least left untested – the capacity of constitutional pluralist thinking to confront the full weight of the sceptical challenge in the wider global context. Nevertheless, I will argue in the concluding sections that, for all its overreliance on the European context, and for all the difficulties posed by the broader transnational regulatory environment, there remain today good arguments for pursuing the project of adapting the language and mindset of constitutionalism to meet the pluralist imperatives of broader global conditions.

II.  Constitutional Pluralism in Europe The idea of constitutional pluralism derived a lot of its initial focus and momentum from the circumstances of high-profile constitutional clashes over the implications of Europe’s supranational arrangements. The key sites of these clashes were the supreme or constitutional courts of the Member States. Faced with issues such as the compatibility of new instruments of supranational authority with national standards of human rights,12 the reconciliation of a Treaty-by-Treaty expansion of overall supranational jurisdiction into areas of public policy traditionally associated with the nation state with the basic idea of national democratic control,13 the tension between accession to a mature transnational polity and a minimum sense of sovereign self-determination,14 or the extent to which transnational security concerns may encroach on core national responsibilities in criminal justice,15 national courts have in a prolonged series of high profile cases been required to adjudicate on the basic source and conditions of final constitutional authority in contexts where the states and the EU palpably possessed overlapping competence. And in so doing, these national courts have tended to affirm or to develop conceptions of constitutionalism which, in stressing or assuming the autochthonous quality of state constitutional authority and the national distinctiveness of its content, have been prepared to countenance the claims to authority emanating from the judicial organs of the EU only on their own nationally conditional terms and not on the absolute terms set or assumed by the EU itself. 12   See, eg, German Federal Constitutional Court, Case BvL 52/71 Solange I [1974] BVerfGE 37, 271, published in English as [1974] 2 CMLR 540. 13  See, eg, German Federal Constitutional Court, Case 2 BvR 2134, 2159/92 Maastricht [1993] BVerfGE 89, 155, published in English as [1994] 1 CMLR 57. This landmark case concerned the constitutionality of the Maastricht Treaty, but every subsequent European Treaty, including the abortive Constitutional Treaty and the Lisbon Treaty which succeeded it, has likewise given rise to litigation in national constitutional or supreme courts. For reflection on the decisions of the German and other top courts prior to ratification of the Lisbon Treaty see, eg, the special issue of the (2009) 10 German Law Journal. 14   See, eg, Polish Constitutional Tribunal, Case K 18/04, OTK Z.U. 2005/5A/49, English summary http://www.trybunal.gov.pl/eng/summaries/documents/K_18_04_GB.pdf. See, more generally, W Sadurski, ‘Solange Chapter 3; Constitutional Courts in Central Europe – Democracy – European Union’ (2008) 14 European Law Journal 1. 15   See, eg, the various decisions on the legality of the European Arrest Warrant, discussed in Baquero Cruz, n 10.

22  Neil Walker As an account of these cases and of their context of emergence and reception, constitutional pluralism has an immediate plausibility. If we take the three core challenges in turn, to each the European case has offered a strong prima facie answer. In the first place, the European example is one where, whatever fears may be expressed in different quarters about the overweening ‘constitutional’ ambitions either of the Member States or of the EU itself, the diversely sourced and wide-ranging invocation of the language and logic of constitutionalism in the face of legal and political contestation shows no realistic prospect of being resolved in terms of a newly minted, widely accepted and broadly effective constitutional unity. The relevant organs of the EU remain implacable in their own claims to self-standing authority, but equally, the relevant constitutional organs of the 27 Member States continue to make plausible and robust claims to their own original and final constitutional authority for all matters within their national purview, including the jointly designed supranational edifice.16 In the second place, however, this does not mean that the European supranational domain is easily categorised merely as a plurality of constitutional unities without a plausibly constitutional connection. Institutionally, we can point to a number of bridging mechanisms which in the round provide more intimate terms of communication and exchange between the relevant state and the non-state legal entities than is the case in any other postnational setting. If we consider the provisions for the direct domestic applicability (in the case of regulations) or compulsory transposition (in the case of directives) of supranational legislation as well as for its judicial enforcement, for the unmediated implementation of much supranational administration on the part of the Commission and various European agencies, and for the obligatory reference of questions of the authoritative interpretation of supranational law from national to supranational courts, it is clear that both within and across the three key constitutional departments – legislature, executive and judiciary – there is close structural linkage between national and supranational sites of authority. Culturally, too, there is a thick familiarity of national constitutional heritages, one nurtured and reinforced by the gradual development first by judicial and then by statutory means of the idea of the ‘common constitutional traditions’ of the Member States as an active agent of convergence.17 Of course, these concurrent structural and cultural forces do not automatically transmute into constitutional matter. Indeed, as we shall see,18 much of the debate within constitutional pluralism has concerned what, if anything, is possible, and if anything is possible, what is necessary or desirable to complete the process of constitutional alchemy. What is clear, nonetheless, is that the background conditions 16   In an earlier article I coined the term ‘epistemic pluralism’ to emphasise the fact that ‘descriptive pluralism’ in the European context had a deep, hermeneutic quality. That is to say, pluralism is appropriate here not just as an external description of the constitutional landscape, but is corroborated and reinforced by the deepest role self-understanding of the key actors themselves; see Walker, n 6. 17  See, eg, F Balvesi, ‘The “Common Constitutional Traditions” and the Integration of the EU’ (2006) 6 Diritto & questioni pubbliche 21. 18   See Section III below.

Global Context  23 for communication between different constitutional orders are comparably favourable in supranational Europe. In the third place, the argument that it is possible to conceive of constitutional relations between the two levels of constitutional order – state and supranational – cannot easily be defeated by the objection that the European level does not bring ‘true’ constitutional credentials to the table. For sure, the precise constitutional status of the EU is heavily contested, in particular the qualities in which and the degree to which the constitutionalism of the EU resembles that of the state. Indeed, much of the political debate surrounding the eventual failure in 2007 of the EU’s first explicit experiment in documentary constitutionalism concerned this very question.19 Alongside deep disputation of the detailed constitutional credentials of the EU, however, there has in recent years grown up a consensus that the EU does nonetheless possess a constitutional character of sorts.20 In legal terms, with its doctrines of primacy and direct effect and its overall development of an autonomous legal order, and in institutional terms, with its dense and complex governance architecture of Commission, Council, European Council, Parliament and Court, the EU appears to have a material constitution that is closely analogous to and often draws heavily from the state tradition. It may lack many of the background factors normally associated with a ‘thicker’ ‘foundational’ constitutionalism and with a self-conscious political baptism21 but few today would deny it certain ‘thinner’ but still highly familiar constitutional credentials.22 Importantly, then, the sheer constitutional familiarity of the European set-up has diverted attention from what might be regarded as a key question. The emphases have very much been on what kind of constitution Europe can have – and in particular how close to the state template – rather than whether it can have a constitution at all. In other words, for the most part the focus has been on which of various diverse or graduated conceptions of constitutionalism is appropriate rather than on the threshold applicability of very constitutional concept.23 These various factors come together to provide a kind of regional comfort zone for the ideas of constitutional pluralism. The co-existence of a number of sites of undeniably significant legal authority making overlapping and inconsistent claims over the nature, scope and implications of their various jurisdictions, and the fact that these different sites are broadly understood by actors and observers alike as ‘constitutional’ in quality, provides a ready set of answers to the first and third challenges. The second challenge – concerning the prospect of properly constitutional 19   See, eg, N Walker, ‘Reframing EU Constitutionalism’ in JL Dunoff and JP Trachtman (eds), Ruling the World? Constitutionalism, International Law, and Global Governance (Cambridge, Cambridge University Press, 2009). 20   On some of the reasons for this, see Walker, n 19, 149–50. 21   Which, of course, a successful documentary constitutional process would have sought to provide. 22  Although some who would not deny these credentials would still argue that the best way to understand and augment the relations between the different levels within the EU is by reference to a pluralist perspective which excludes the language of constitutionalism. See, eg, M Avbelj’s Chapter in this volume. 23  On the distinction between a concept and its various conceptions, see, eg, R Dworkin, ‘The Jurisprudence of Richard Nixon’ (1972) 18(8) The New York Review of Books 27.

24  Neil Walker relations between and across constitutional units – is the most acute one. And, as it raises the question of the normative dividend of constitutional pluralism, it is also, as already noted, the one that has excited most discussion within the field. On the one hand, there are those, often labelled radical pluralists, for whom nothing strictly constitutional can be said about the relations between different constitutional entities, although the fact that they are constitutional entities suggests that these relations may be conducted in terms which trade on common sensibilities or a shared understanding of the strategic context of interaction.24 On the other hand, there are those who try to complete the process of constitutional alchemy, whether by reference to universal constitutional principles and values of a substantive and structural nature,25 or by reference to jurisgenerative features of the particular dialogue between the different constitutional actors,26 or indeed some combination of the two.27 Yet the practical importance of this area of difference and disagreement in the European context of debate should not be overstated. The underlying descriptive and explanatory diagnosis is largely shared across the various pluralist perspectives, and given the close cultural and legal–structural ties between the states and the EU, those normative problems of reconciliation of the different orders that remain unanswerable or disputed are treated as of ‘manageable’ dimensions – centred upon disagreements between ‘top courts’ – rather than as fault lines affecting the overall configuration of authority in the European legal space. This is not to say that constitutional pluralists analysing the European field have been entirely blind to the fact that, just as there is more to constitutions than constitutional courts, so too there must be more to relations between constitutions than merely judicial difference and dialogue. For all their awareness in principle of the involvement of other institutions, however, the majority of commentators have in fact homed in on the courts as the most visible arena and the clearest manifestation of the problem – an exotic but essentially treatable symptom which tended to dominate consideration of the ailment as a whole.28 24   See, in particular, N MacCormick, ‘Beyond the Sovereign State’ (1993) 56 Modern Law Review 1; though he later modified his view, reintroducing public international law as the tertium quid to regulate relations between the national and the supranational levels. See his Questioning Sovereignty: Law, State and Nation in the European Commonwealth (Oxford, Oxford University Press, 1999) 97–121 (Chapter 7 entitled ‘Juridical Pluralism and the Risk of Constitutional Conflict’). See also Walker, n 6. 25   See, eg, M Kumm, ‘The Jurisprudence of Constitutional Conflict: Constitutional Supremacy in Europe Before and After the Constitutional Treaty’ (2005) 11 European Law Journal 262; also , n 1. 26   See, eg, M Poiares Maduro, ‘Contrapunctual Law: Europe’s Constitutional Pluralism in Action’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart Publishing, 2003). 27   Arguably, Poiares Maduro, n 26, combines the two approaches. For his more recent views, now extended to the wider global context, see his ‘Courts and Pluralism: Essay on a Theory of Judicial Adjudication in the Context of Legal and Constitutional Pluralism’ in JL Dunoff and JP Trachtman (eds), Ruling the World? Constitutionalism, International Law, and Global Governance (Cambridge, Cambridge University Press, 2009). 28   This tendency has probably been accentuated by the fact that one of the more influential pluralist thinkers, Miguel Poiares Maduro, has served as an Advocate General at the European Court of Justice, and has delivered opinions which seem to reflect some of his academic thinking. See, in particular, his Opinion, delivered on 16 January 2008, in Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat v Council and Commission [2008] ECR I-6351. To be fair, however, Maduro himself has stressed more

Global Context  25

III.  Constitutional Pluralism Beyond Europe If we look at the prospect for the constitutionalisation of transnational sites and relations beyond the EU, the challenges set out above are posed much more sharply and insistently. Faced with the proliferation of global institutions around the permanent framework of the United Nations, of global and regional human rights charters and standard-setting bodies, of new forms of regional economic organisation beyond Europe, of functionally specialist regimes of global public authority in matters such as crime, labour relations and environmental protection, and of private and hybrid public–private forms of self-regulation and administrative capacity in other areas of specialist practical and epistemic authority from global cyberspace to international sport, constitutional pluralism finds itself in a less obviously receptive environment.29 So much so, indeed, that much of the broader literature on the global legal configuration implicitly or explicitly rejects the ideas of constitutional pluralism, while those approaches which seek to keep faith with constitutional pluralism and adapt it to the global scene struggle to justify their approach and occupy a less confident and secure position within the debate than they do in the European context. Let us again look at each of the three sceptical challenges in turn in order to illustrate these points. If we begin with the question of the tendency of constitutionalism to embrace all normative phenomena within a singular logic and encompassing framework, this might seem the least likely ground of challenge to the appropriateness of constitutional pluralism within the wider transnational context. After all, are the most obvious features of the global legal landscape not precisely those that are ‘disorderly’?30 Rather than as a coherent whole, do we not think of the global legal configuration as fragmentary,31 as ‘polycontextual’,32 as embracing a ‘strange multiplicity’,33 as part of the diverse and sometimes impenetrable ‘mystery of global governance’?34 And should we not, therefore, expect constitutionalism conceived of in a global key to match and reflect this underlying deep diversity, thereby adopting a sensibility that is pluralism-friendly? In some influential quarters of transnational constitutional thinking, however, just the opposite is the case. For some who want to take constitutionalism to the than most the need to look beyond the courts to broader institutional structures in order to understand pluralism in the round. See, in particular, Poiares Maduro, n 27. 29   See, eg, N Walker, ‘Beyond Boundary Disputes and Basic Grids; Mapping the Global Disorder of Normative Orders’ (2008) 6 International Journal of Constitutional Law 373; M Rosenfeld, ‘Rethinking Constitutional Ordering in an Era of Legal and Ideological Pluralism’ (2008) 6 International Journal of Constitutional Law 415. 30   Walker, n 29. 31   Koskenniemi, n 9. 32   G Teubner, ‘Constitutionalising Polycontextuality’ (unpublished paper). 33   J Tully, Strange Multiplicity: Constitutionalism in an Age of Diversity (Cambridge, Cambridge University Press, 2005). 34   Kennedy, n 9.

26  Neil Walker global level, it is precisely as a reaction against and in response to these underlying tendencies toward fragmentation. Constitutionalism is embraced just because it is believed to have the capacity to re-impose order, to re-establish hierarchy, to articulate and apply a comprehensible redesign. This steering ambition comes in different variants. In one version, the singular model of transnational constitutionalism is institutionally located in the United Nations, its Charter functioning as an ersatz written Constitution for the post-war world order.35 In other versions, the basis of constitutional order is lexical rather than institutional. In particular, there are a number of strains of the so-called constitutionalisation of international law, in which ‘international law’ itself is protected and projected as a single juristic category.36 Typically under this approach some types of international rules such as customary international law, ius cogens, human rights law, ‘world order’ treaties and obligations erga omnes are deemed to have a special facility to organise the international order in a ‘Constitution-like’ way. Whether due to their generative capacity, or their trumping quality, or their comprehensive reach, they stand apart from and above other international rules and lend some measure of coherence and integrity to the whole. We should be careful not to overstate the unifying ambition of any of these brands of global constitutionalism. They are far from suggesting a world state to subsume and replace the category of nation states, and, indeed, rarely propose any kind of top-loaded federal design.37 As noted, their impulse tends to be reactive rather than proactive, a limited ‘re-ordering’ response to the deepening anarchy of global legal relations in a world of ever more divergent and complexly overlapping jurisdictions rather than a new and constitutive set of markings on a legal tabula rasa. But these efforts do, nonetheless, continue to display distinct traces of a certain kind of singular and hierarchical strain of juristic thought that is closely associated with the tradition of state constitutionalism. The performative meaning of making a claim about the global regulatory sphere in ‘constitutional’ terms is one of authorisation – indeed self-authorisation. The language of constitutionalism is resorted to not just as a familiar trope of the legal imagination but as a way to outrank other rules and outflank other ways of conceiving of the global legal order.38 Yet a self-defeating irony surely lurks within such a bold discursive move. On the one hand, it is precisely the lack of any agreed and settled overall framework of legal authority for the proliferation of new sites of transnational legal authority in the dense mosaic of global regulation that tempts a certain type of singular constitutional discourse to fill the vacuum. On the other hand, if constitutionalism’s   See, eg, Fassbender, n 7.   See, eg, de Wet, n 8. 37   See, eg, J Habermas, ‘Does the Constitutionalization of International Law Still have a chance?’ in J Habermas, The Divided West (Cambridge, Polity, 2008) 115–210. Even though Habermas is unusual in explicitly proposing a multi-level institutional structure, of the three levels he proposes – global, regional and national – he allows the global by far the most limited jurisdiction, restricted to questions of peace and human rights. 38   See, eg, Koskenniemi, n 9; Kennedy, n 9. 35 36

Global Context  27 ambition is to put its own claim to final authority beyond question, then the inherent disputability of any and all ‘global metaprinciples of legal authority’39 which underscores the unsettled quality of the transnational legal sphere means that constitutionalism in this singular mode cannot achieve its own ambition. What is more, just because of the underlying lack of settlement or of agreed general grounds for the justification of postnational constellation, any such singular constitutional discourse deserves to fail in its presumption of unassailable authority. In summary, there is a monistic strain in transnational constitutionalism which, for all the comparative (to the state tradition) modesty of its remit, is fated to fall short in its bid to place its own authority beyond question, and justifiably so. Yet it is an active, and indeed growing, dimension of the discourse on transnational constitutionalism, one which implicitly or explicitly sets itself at odds with the various strains of constitutional pluralism, and one, therefore, which con­ tributes to the overall hostility of the regulatory environment to the very idea of constitutional pluralism. This monistic strain, it follows, should be carefully distinguished precisely from those other explicitly constitutional conceptions of the global transnational order that seek to emphasise the diversity of transnational sites of authority. In these cases, the second and opposite challenge – namely the reconciliation of plurality in terms which remain at all constitutionally meaningful – comes into play, although, as we shall see, the first challenge continues to lurk in the near background. Those who stress the variety of the constitutional register at the global level, in turn, can be further divided into different sub-categories. On the one side, there are those for whom pluralism, including a pluralism of constitutional sites and relations, is an unavoidable and irreversible consequence of the functional differentiation of world society. In a perspective closely associated with contemporary systems theory, the ever increasing autonomy of the globally ramified spheres of economy, ecology, science, education, health, sport, media, virtual communications etc, is postulated as both consequence and reinforcing cause of the decline of the role of the traditional politico-legal constitutionalism of the state as the effective container of the various specialist sub-systems within a particular territorial demarcation.40 Yet the demise of a comprehensive mode of politico-legal constitutionalism – of a constitutionalism built around an idea of a self-contained community in which all matters of ‘public’ interest are contested and resolved in common, need not mean the end of constitutionalism tout court. Instead, in the systemic pluralist vision we are witnessing the development of new transnational forms of ‘societal constitutionalism’.41 According to this new global dynamic the

  Walker, n 29 at 386.   See, in particular, the work of Gunther Teubner, n 32; see also his ‘Societal Constitutionalism: Alternatives to State-Centred Constitutional Theory?’ in C Joerges, I-J Sand and G Teubner (eds), Transnational Governance and Constitutionalism (Oxford, Hart Publishing, 2004). 41   Teubner, n 40. 39 40

28  Neil Walker ‘self-constitutionalisation’42 of the various specialist functional sectors is no longer grounded in and reducible to the articulations either of state law or the orthodox treaty regimes of international law, or indeed of any other canonical legal form.43 The new societal constitutions will continue to draw on these familiar juridical sources in their continuous processes of reflexive self-organisation, but the basic impulse towards self-constitutionalisation and its governing logic is provided by the very character and domain concerns of the functional specialism itself; by the methods available within its special medium of practice – and to those actors implicated in that medium of practice – of communicating and realising the forms of social power or influence distinctive to that medium of practice. A more modest and familiar version of this kind of functionally driven global constitutional pluralism can be found in the idea of ‘sectoral constitutionalisation’.44 Here the focus is upon the institutional centres and their conventional legal foundations rather than the functionally coded sites of practices as a whole. The accent is on the hybrid ‘treaty-constitutions’45 of special international organisations or regimes, such as the International Labour Organization or the World Trade Organization. These are constitutive instruments for the legal domains in question, not just in terms of providing an institutional and norm-generating frame and claiming an original juridical authority, but also, and increasingly, in endorsing or encouraging a broader form of erga omnes constitutional sensibility in terms of rights protection for the individuals affected by the regimes.46 To these positions the second challenge is a clear and pressing one. What makes the basic plurality of constitutional orders they describe pluralistic in nature? In what does the constitutional coherence between the parts consist? If, as Gunther Teubner, the leading exponent of modern systems theory, declares, ‘in the sea of globality there are only islands of constitutionality’,47 where are the constitutional causeways that connect these islands? The answer is not clear. If the emphasis is on the specificity of the newly emergent societal or sectoral constitutions in the absence of any corresponding newly emergent legal-political totality, then what, if anything, links these constitutionally justified specificities in constitutional terms is problematic. One part of the answer may depend on structural analogy. Arguably, a key ‘constitutive’ puzzle faced by the stakeholders of relatively autonomous global subsectors and by those who occupy their various external environments, namely how to balance the freedom of those most centrally concerned with and affected by a practice to govern that practice against the need to limit its expansion into other spheres and to curb its tendency to encroach on the autonomy of other sectors of   Teubner, n 32.   Such as the common-law based lex mercatoria. See Teubner, n 32. 44   See, eg, A Peters, ‘Membership in the Global Constitutional Community’ in J Klabbers, A Peters and G Ulfstein (eds), The Constitutionalization of International Law (Oxford, Oxford University Press, 2009) 201–3. 45   Ibid 203. 46   Ibid 212–15. 47   Teubner, n 32. 42 43

Global Context  29 social practice and their key stakeholders, is the functional equivalent under a globally differentiated order of the traditional state constitutionalist concern to safeguard the ‘internal sovereignty’ of ‘the people’ while ensuring that their ‘external sovereignty’ did not compromise the internal sovereignty of others.48 A second part of the answer may, more straightforwardly, concern common transversal norms. In particular, proponents of a differentiated form of global constitutionalism may argue that basic human rights standards should prevail across different societal or institutional sectors regardless of these cleavages. Indeed, on this view, the very proliferation of such cleavages and the problems of achieving ‘thicker’ forms of democratic constitutionalism in consequence serve to underline the importance of the alternative protection provided by globally guaranteed human rights standards.49 A third and final part of the answer might concern the relational dynamics themselves. If the global Constitution is one of multiple and variable sectors, one in which the marginal connections and relations between sites of governance become central rather than peripheral, then perhaps there is some kind of underlying relational logic or, less passively, perhaps there can be developed terms and patterns of constitutional exchange between these various sectors which can be accounted for or justified in terms of some kind of defensible constitutional reason. At a minimum, does the fragmentation of the transnational constitutional order into a heterarchy of sites not permit and even encourage the development of some kind of framework of mutual recognition and contestation and of checks and balances between sites and their different claims to authority? And does the complex cross-polity institutionalisation of a system of countervailing power not provide the basis from which pluralism can be transformed into a recognisable set of constitutional virtues?50 Certainly, there is in the approach of the systemic constitutional pluralist some recognition of all such solutions. The claim to move beyond plurality to pluralism remains a precarious one, however. It stands in sustained tension with the sheer number, diversity, unpredictable emergence and uncontainable evolution of the islands of self-norming and institutional capacity in the new global constitutional archipelago. And it is in response to this and in an attempt to fashion a more systematic and encompassing set of constitutional steering mechanisms that we find another more universalist strain within global constitutional pluralism. This thread of constitutional pluralist thought, closely associated with Mattias Kumm51 and others,52 adopts a different and more resolute approach to the tension between the two constitutional imperatives of the postnational constellation – the auto­ nomy of the particular parts and the coherence of the whole.   Ibid. See also Krisch (2009), n 11.   See, eg, Teubner, n 32; Peters, n 44. 50   See Krisch (2009), n 11; see also Rosenfeld, n 29. 51   Kumm, n 1. 52   See, in particular, D Halberstam, ‘Constitutional Heterarchy: the Centrality of Conflict in the European Union and the United States’ in JL Dunoff and JP Trachtman (eds), Ruling the World? Constitutionalism, International Law, and Global Governance (Cambridge, Cambridge University Press, 2009). 48 49

30  Neil Walker For Kumm, the modernist past remains the key to the future. The philosophical core of constitutionalism has not changed since the advent of modern constitutionalism through the medium of the maturing state system of late-eighteenth-century Europe and America. Crucially, what is constitutionally basic for him is not a matter of institutional design but of underlying normative principles. These normative principles flow from the basic modernist ambition of persons self-conceived as free and equal individuals to act collectively to deliberate, develop and implement their own conception of the common interest or public good. Such meta-political foundations distinguish the modern age from the traditional hierarchies and the sense of human society as in thrall to a prior order of things which characterise earlier forms of social organisation and their associated social imaginaries.53 And from these foundations, according to Kumm, we can derive a set of universal constitutional commitments to principles of legality, subsidiarity, adequate participation and accountability, public reason and rights protection.54 Against this larger canvas the traditional state-centred constitutional system assumes a more modest significance than is often appreciated within constitutional thought. It is exposed as but one architectural representation of the underlying principles, rather than an exclusive or dominant or even optimal template for constitutional government. Instead, under conditions of intensifying globalisation the basically cosmopolitan texture of a constitutionalism committed to universal principles becomes more apparent, and the state is now but one constitutional player on a wider stage. As free and equal persons operating under certain constraints of interest, information, geography and affinity, we continue to respect particular contexts of decision making and public interest formation, and the principles of subsidiarity, participation and accountability recognise this. However, as free and equal persons we are also categorically committed to acknowledgment of the freedom and equality of all others, and so to an accept­ ance of the universalisability of our political condition. In this way, we can reconcile our commitment to particular polities and sites of authority with a belief in an overarching normative framework which informs the terms of our various particular manifestations of public authority. In the final analysis, the global division of the world into particular polities remains inevitable but the particular form that such a division takes is not so; rather, it is contingent upon shifts in the underlying circuits of social and economic power. By replacing institutional or lexical hierarchy with normative universalism, Kumm, and those with similar visions, find a more robust answer to the second challenge than is available to the systemic pluralists while avoiding the more obvious dangers of constitutional monism. Inevitably, however, the idea of constitutionalism as a single cloth, however divorced from traditional conceptions of hierarchy, brings the first challenge very much back into the frame. Is such a confident claim on behalf of constitutionalism – even if its focus is on general principles rather than a particular vertical design of rules or institutions, not just   See, eg, C Taylor, Modern Social Imaginaries (Durham, NC, Duke University Press, 2004).   Kumm, n 1.

53

54

Global Context  31 one more hegemonic move on behalf of a singular constitutional vision? And how genuinely pluralist can such a vision be if its basic normative contours are settled in advance, even if only at the high level of abstraction proposed by Kumm? One author who has posed these questions more keenly and insistently than most is Nico Krisch.55 For him, it seems that constitutionalism in a global age is caught in a Procrustean dilemma. On the one hand, the kind of ‘foundational constitutionalism’56 well known from the state tradition – the ‘thick’ variant based upon the constituent power of the collective people living in a distinct all-­ embracing political society – simply does not suit the more fragmented circumstances of the global age. On the other hand, if we try to stretch and adapt constitutionalism to fit these new conditions we are faced with a series of unsatisfactory alternatives. Either, in a first case, we retain something of the monistic legacy of constitutionalism – a holistic architectural or (at least) intellectual vision which, in its excessive ambition and self-assertion, lacks both legitimacy and plausibility in an age of global diversity. Or, in a second case, we are guilty of a kind of constitutional dilution or corruption, retaining the term ‘constitutional’ as an overstated or inappropriate label for an entirely new type of institutional and normative complex. In particular, if, as is the case with the more systemic forms of pluralism, all we retain from the tradition of state constitutionalism is a commitment to various of its ‘thin’ properties – juridical autonomy, an institutional framework of checks and balances, and fundamental rights protection – but without any plausible sense of an authoritative frame for locating these within a single constitutional universe, then perhaps the constitutional label becomes a mere placebo or distorting diversion. That is to say, constitutionalism may become a source of complacency – a false promise and false comfort in a world that no longer bends to its design, or a source of confusion – a category mistake in a world which needs new categories. In either event constitutionalism threatens to become an impediment rather than a guide in the search for optimal solutions to the question of governing new configurations of social power. This takes us directly to the third challenge and the alternative solutions suggested by that third challenge. For pluralism to make sense as a normative register for the contemporary global order – and bearing in mind the extent to which empirical conditions of global regulation militate against anything other than a pluralist understanding – then perhaps the ‘constitutional’ descriptor just has to be dropped. As Krisch himself suggests, in the last analysis constitutionalism and the scale and quality of the pluralist understanding adequate to the global age may simply be irreconcilable.57 Perhaps, the best way of ensuring the pluralist virtues of mutual contestation, recognition and adaptation and a complex framework of checks and balances, conceived of as a modest framework of co-ordination between relatively autonomous polities,58 is to detach them from a constitutional   See references, n 11.   Krisch (2010), n 11. 57   Ibid. 58   Krisch (2009), n 11. 55 56

32  Neil Walker discourse which is unsympathetic on either side of this delicate ambition; either in the strength of its traditional championing of the autonomy of the parts or in its effort to conceive of the new in terms of an idea of totality and integrity which also borrows from the old.

IV.  Pluralism and the Constitutional Legacy So, what, if anything, does constitutional pluralism under conditions of globalisation have left to offer in the light of these challenges? This question is most profitably addressed by adjusting our lens slightly and by approaching the constitutional predicament from a somewhat different angle than above, and by taking note of a clear bifurcation that has emerged in the use and treatment of constitutional ideas in the global age. On the one hand, as a source of doctrine the accumulated arsenal of constitutional thought is treated in an ever more eclectic manner in the global age. Constitutional doctrine is drawn upon for both epistemic and symbolic reasons – as a rich resource of resilient ideas of good governance couched in a language which also happens to carry a distinguished and potentially authority-inspiring legacy. The spread and adaptation well beyond the traditional container of the nation state of tried and tested aspects of constitutional doctrine such as fundamental rights protection, separation of powers and institutional balance, federalism and subsidiarity, due process and natural justice, proportionality and balancing, or ‘hard look’ doctrines and requirements to give good reasons, speaks to a process of widespread ‘low intensity’59 dissemination. Constitutionalism becomes a mobile resource, a ‘thin’ and footloose structure and stylisation of norms used to qualify and dignify the emergent sites of a new global regulatory structure of authority without being constitutive of these sites in the ‘thick’ manner redolent of the nation state. Constitutionalism on this view is a matter of detail, adding an older texture to new governance forms rather than providing a formative inspiration. On the other hand, we also find constitutionalism used as reference point for the most encompassing (re)imagination of the global body politic. Whether in the work of Habermas, or Teubner, or Kumm, or – even if he ultimately rejects the constitutional label – of Krisch, constitutionalism provides a point of departure for the broadest consideration of the nature and resilience of the modernist settlement in legal and political thought. Again, as with constitutionalism as doctrine, constitutionalism as imagination sends a reasonably coherent message – certainly at the highest levels of abstraction reached by this broader mode of thought. Recall that, for Kumm, constitutionalism is about the political promise of an 59   M Poiares Maduro, ‘The Importance of Being Called a Constitution: Constitutional Authority and the Authority of Constitutionalism’ (2005) 3 International Journal of Constitutional Law 332, 340.

Global Context  33 unprecedented epoch in which free and equal individuals make over society in their own terms;60 or, as Habermas or Krisch would have it, constitutionalism is about the development of the very idea of public autonomy – about how individuals constitute themselves in public as a public and with due regard to and in symbiotic relationship with their equal freedom in the sphere of private autonomy;61 or as Teubner would argue, constitutionalism is about the balance between the autonomy and self-limitation of different functional sectors inter se in a differentiated order – with autonomy retained as a deep freedom and equality-respecting ideal even as its emergent forms escape our received modern distinction between a generically public and a generically private sphere.62 In all cases, constitutionalism serves as a reminder of modernity’s resilient ambition for the collective self-constitution of the social and political world in a moral universe in which the individual is the basic unit. Where constitutionalism as doctrine is about detail, constitutionalism as imagination, by contrast, sets the broadest of horizons. Crucially, however, for all their contrasting features, the two levels of constitutional discourse for a global age share a common absence. Where constitutionalism as doctrine is too specific in its various remits and too past-derivative to provide a key formative influence for the new post-Westphalian sites of authority, constitutionalism as imagination for its part is both too general in scope and substantive ambition and too much in the shadow of the dominant procedural heritage of state Constitution-making to provide a formative influence for these new constitutional sites. In other words, we are faced in a post-Westphalian world with a situation in which constitutionalism arguably flies too low or too high, either too dependent on other forms or too independent of any particular forms. Why this is so is both consequence and reinforcing cause of the changing structure of constitutional authority in a post-state world. In the state tradition, the imaginative and the doctrinal dimensions of constitutionalism tended to be closely aligned through the dimension of constituent power. For constitutionalism in this mode was concerned as much with formative influence – with the particular pouvoir constituant and the ideas of guiding purpose and ultimate justification associated with the making of political community – as with the tool-kit of mechanisms through which the duly formed and constituted authority – the pouvoir constitué – seeks through doctrine to express and represent its constitutive source. State constitutionalism, in other words, was concerned both with the framing of the particular sites of authority and with the detail of what was framed. Constitutionalism in the state tradition, therefore, was always about treating the ‘spirit’ and the ‘letter’ of 60   Kumm, n 1. See also the exchange between Kumm and Krisch on EJIL: Talk! in 2009, following the publication of JL Dunoff and JP Trachtman (eds), Ruling the World? Constitutionalism, International Law, and Global Governance (Cambridge, Cambridge University Press, 2009), www.ejiltalk.org/author/ mkumm/. 61  See, eg, J Habermas, ‘Constitutional Democracy: A Paradoxical Union of Contradictory Principles’ (2001) 28 Political Theory 766. Krisch (2009), n 11 and his EJIL: Talk! exchange with Kumm, n 60. 62   Teubner, n 32.

34  Neil Walker the law within a single frame of reference, about background culture as well as foreground text, about the regulative ideal as well as the regulated practice, about deep ‘second-order’ justification as well as immediate ‘first-order’ validity. In short, it was about both imagination and doctrine, and about how the imaginative and the doctrinal were closely joined and mutually nourished through the container of the self-constituting and self-constituted polity. Certainly, there was also a dimension to the constitutional imagination which was prepared to reach beyond the state, which treated the Constitutions of different and other free and equal peoples as morally comparable and ethically associated units. But this global dimension remained parasitic upon the more basic connection between the imaginative and the doctrinal dimensions in the context of the state.63 Crucially, the post-Westphalian world of constitutionalism severs this basic connection between the doctrinal and the imaginative while often remaining in retrospective thrall to the significance of such a connection in the high modern era of state-centred constitutionalism. An appreciation of this point allows us to reconceptualise and restate the various dilemmas of constitutional pluralism in global context as flowing from the expectation to do too much with too few resources. The low-flying constitutionalism as doctrine seems to claim too much, at least by implication, in using the historically formative register of constitutionalism to account for a regulatory context in which such constitutionalism as is available is no longer doing and can no longer do that formative work, but is instead merely supplying the regulatory technology for an already and otherwise formed site of authority. Hence the criticism that constitutionalism is tendentially but inappropriately inclined to monism, and the related claim that the language of global regulatory pluralism finds a more becoming modesty if the descriptor ‘constitutional’ is removed from the units we seek to conceive of within the pluralist structure. Equally, however, the high-flying constitutionalism as imagination presumes too much if it treats itself as an encompassing meta-authoritative normative frame for the plurality of sites of global constitutional authority. Rather, its claim and message is prior to the particular forms of constitutionalism and the particular norms associated with these forms. What it can offer is precisely not a higher-order or framing legal normativity – a kind of constitutional super-doctrine – for that would presuppose a formative and framing role which it does not possess and which it could not posses without claiming new con­ stitutional unity, but a deeper and normatively unrealised form of constitutional pre-orientation. If we return briefly to the special case of the European Union and its inappropriateness as a paradigm for postnational constitutionalism more generally, we may observe how the severing of the two registers of constitutionalism is here less evident, and less evidently problematic. Constitutionalism as doctrine in the European Union bears such a close resemblance to many state forms and remains 63   See N Walker, ‘Out of Time and Out of Place: Law’s Fading Co-ordinates’ (2010) 14 Edinburgh Law Review 13.

Global Context  35 so closely connected to its statist roots that, as we have seen, its ‘thin’ credentials are widely respected, and also treated by many as a sound basis on which a pluralist connection between the national and the supranational spheres of influence might be forged. And while there has been much controversy over just how much legitimacy this ‘thin’ constitutionalism supplies, and also about whether it can or should be supplemented by a ‘thick’ foundational constitutionalism, at least the EU has developed in a sufficiently state-like direction that the linking of the supranational constitutional imagination to a recognisable politically constituent process has remained a viable ambition for many – or at least did so until the demise of the Constitutional Treaty in 2007. In short, neither constitutionalism as doctrine nor constitutionalism as imagination seem to be as disconnected from their traditional basis of support as they are in the wider global sphere. Constitutional pluralism appears more plausible, as too in some measure does the alternative of a new constitutional unity. In the global context, in the absence of the lock of constituent power the two levels of constitutional discourse are more clearly stratified and more palpably incomplete in the absence of the other. Yet just as postnational constitutionalism in general is not best understood in the paradigm of the European Union, post­national constitutionalism beyond the European Union should not be discounted just because in some respects its development compares unfavourably with that of the European Union. For it does not follow from the misalignment of the two constitutional discourses – constitutionalism as doctrine and constitutionalism as imagination – that there is no value in seeking to preserve and develop the modern constitutional legacy at either or both levels under conditions of contemporary global pluralism. Rather, it seems that the continuing value of constitutionalism, and the basis for believing that any such value outweighs its disadvantages, lies precisely in the combination of those answers it does still provide and those questions it raises in lieu of the answers it can no longer provide. If we first consider constitutionalism as doctrine, as already noted we cannot deny the value of the constitutional normative resource-set accumulated over the period of political modernity, or its continuing applicability to non-state sites, however partial, fragmented and ‘non-holistic’64 many of these sites are. The various functionally specific and/or institutionally clustered points of non-state authority may have come to resemble nodes in a global network, each made up of a complex mix of internal self-regulation and diverse external regulation, rather than each providing a self-contained regulatory universe of its own, as in the state tradition.65 Yet many of the same basic puzzles of governance are being addressed, and so much of the same technology of governance remains appropriate. This point is placed in sharp and reinforcing perspective when we turn to reconsider constitutionalism as imagination. In one respect, this serves as an orientating reminder of what should underscore and inform our puzzles of governance in state 64  See N Walker, ‘Beyond the Holistic Constitution?’ in P Dobner and M Loughlin (eds), The Twilight of Constitutionalism? (Oxford, Oxford University Press, 2010). 65   Ibid 297–303.

36  Neil Walker or state-like holistic settings and non-holistic settings alike. The constitutionalist vision recalls the abiding importance of the meta-political question of how to generate, adjudicate and apply our common interest in accordance with our common standing as free and equal persons, even in a post-state world in which the subject, mechanisms and object of common interest are out of kilter, and where, accordingly, ‘we’ increasingly do not get to address the common interest question in its entirety in common. However, this paradoxical feature of the common interest should not defeat but should guide our efforts to interpret the ‘letter’ of constitutional doctrine in light of the ‘spirit’ of constitutionalism. The contextual appropriateness and refinement of all our particular inherited constitutional techniques, from rights protection through doctrines of consultation and due process to our manifold methods for the devolution of legal power, should be informed by our adjusted sense of the elusive but still vital centrality of the idea of common interests amongst equals in a multi-centred world of overlapping and partial authorities. Yet constitutionalism as imagination, as well as showing us how to keep the cup of self-government half full, is also salutary in underlining our sense that it is half empty. As well as serving as an important reminder of the deeper purpose of particular constitutional doctrines and the flexibility of their application, it also highlights what we no longer have or can guarantee to preserve. Constitutionalism as imagination recalls to us that in a context of constitutional foundationalism our sense of the political realm, of constituent power and of constituted power, was linked together in a continuous framing logic, but that the sorts of constitutional questions we once posed and addressed within a joined-up political container now increasingly arise in a manner so fragmented and loosely coupled that they threaten the very promise of the political as embodying our capacity to make over the world in our own terms. Constitutionalism as imagination thus also functions as a kind of ‘placeholder’66 for what is in danger of being lost if we abandon our commitment to think and act as authors of the constitutive conditions of political society – however diverse and complexly intermingled the transnational societal reference of that political society might be – and acts as a continual prompt for us to seek to retain that aspiration, however formidable, and fashion its pursuit to our new circumstances.

V.  Constitutional Pluralism? But even if in these ways constitutionalism in general does remain relevant to the global conditions of late modernity, one last important question of language remains. Does the kind of loosely aligned dual-pronged approach to the sustenance of a constitutional discourse suggested here fit well with the particular perspective of constitutional pluralism which provided the starting point for our analysis?   Koskenniemi, n 9.

66

Global Context  37 The answer is a mixed one. In one sense constitutional pluralism is a product of the very structure of state-centred political modernity we are trying to look and think beyond. It is an attempt to solve a problem that is becoming outmoded. Constitutional pluralism, conceived of as idea of a constitutionally relevant connection between self-authorising constitutional sites, silently assumes something like the statist template of constituent power as the legitimate basis for the selfauthorisation of the post-national constitutional sites. If self-authorisation increasingly lacks that legitimation, however, the focus of our concern shifts to the broader question of what form of legitimation is possible in place of or in supplementation of site-specific self-authorisation.67 At the same time, with the weakening of the sources of internal, site-specific legitimation, our sense of the constitutional ‘closure’ of the various sites is reduced, and so in consequence is the puzzle of how such increasingly ‘open’ sites can relate constitutionally. In other words, the less site-specific we understand constitutional authority to be, the less problematic we conceive constitutional movement across boundaries, and the less sharply framed the original definitive questions of constitutional pluralism appear. On the other hand, if we think of constitutional pluralism not as a series of doctrinal or otherwise constitutionally relevant answers to the puzzle of how different Constitutions connect, but simply as referring to the continuing relevance of constitutionalism in addressing the mix of empirical and normative factors which contribute to the deep pluralism of the emerging global order, then it certainly remains a relevant conceptual point of departure. Our understanding of constitutionalism may have been unbundled to a degree that makes the original Europe-centred debate about the constitutional ‘plurality of unities’68 less paradigmatic. Yet that very process of unbundling and the new horizons of meta-political debate it opens up are strikingly indicative of the ways in which the constitutional legacy remains relevant both as a path-dependent influence upon and as a means of collective sensitization towards our complexly differentiated and interconnected global order.

67   See, eg, S Besson, ‘Whose Constitution(s)? International Law, Constitutionalism, and Democracy’ in JL Dunoff and JP Trachtman (eds), Ruling the World? Constitutionalism, International Law, and Global Governance (Cambridge, Cambridge University Press, 2009). 68   H Lindahl, ‘Sovereignty and Representation in the European Union’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart Publishing, 2003).

3 Rethinking Constitutional Authority: On the Structure and Limits of Constitutional Pluralism MATTIAS KUMM

I.  Constitutional Pluralism: Between Hierarchical Integration and Deep Pluralism It is widely recognized that European constitutional practice has a pluralist structure. The legal orders of Member States are not hierarchically integrated into the European legal orders. Instead, from the point of view of Member States’ highest courts, the status of European Union law is a matter to be determined with reference to national constitutional norms. The relationship between the European legal order and the legal orders of Member States has been the focus of much writing on pluralism in the European Union. What the recent Kadi decision1 has clarified is that – according to the European Court of Justice (ECJ) and in contrast with the position of the European Court of First Instance (ECFI) – the relationship between the European legal order and UN law should also be conceived in pluralist terms. The ECJ has insisted that the EU legal order is not hierarchically integrated into an international legal order, in which UN law supersedes all other Treaty law as established by Article 103 UN Charter. Instead, EU law determines on its own terms the conditions under which UN resolutions may be enforced by European law. The legal orders established by Member States, the European legal order and the international legal order are thus, from the perspective of the highest courts of the respective legal orders, to be conceived as distinct and separate legal orders that are not hierarchically integrated in the more encompassing legal order. The Kadi decision is an invitation to deepen questions relating to legal pluralism, by focusing not just on familiar terrain of the relationship between EU law and the law of Member States, but also the relationship between EU law and UN law. 1   Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat International Foundation v Council and Commission [2008] ECR I-6351.

40  Mattias Kumm Generally, there are two jurisprudential questions that a pluralist legal structure raises. The first concerns the management of the interface between pluralist (nonhierarchically integrated, heterarchical) legal orders and has been the focus of a great deal of scholarly attention in the context of the relationship between EU law and national law. If different legal orders find themselves in a non-­ hierarchical relationship to one another, how should they manage their relationship to one another? In recent years commentators have pointed out that, notwithstanding the pluralist nature of legal practice, the relevant actors – and courts in particular – have established mechanisms and designed doctrines that allow for constructive mutual engagement between different legal orders. Joseph Weiler, for example, has pointed to the constitutionally tolerant nature of Member States’ engagement with European law and has identified in that tolerance the normative core of the European integration project.2 On a doctrinal level Miguel Maduro has described the ‘contrapuntal’ nature of the interaction between the ECJ and Member States’ courts.3 And this author has argued that the doctrines typically used by national courts can be reconstructed as the result of applying constitutional principles that are shared across legal orders in Europe to questions of conflicts between different legal norms.4 Legal pluralism in Europe is thus guided, constrained and structured in a way that might justify describing that practice in constitutional terms, even in the absence of hierarchical ordering conventionally associated with constitutionalism.5 These authors – and others6 – have provided an account of the relationship between different legal orders that clarify how legal coherence is possible even absent hierarchical integration. The idea of constitutional pluralism carves out a third way of conceiving of the legal world between hierarchical integration within one legal order on the one hand and a deep pluralism on the other, where actors of each legal order proceed without systemic regard for the coherence of the whole.7 Authors addressing the Kadi decision from this angle, as

2   JHH Weiler, ‘In Defence of the Status Quo: Europe’s Constitutional Sonderweg’ in JHH Weiler and M Wind (eds), European Constitutionalism Beyond the State (Cambridge, Cambridge University Press, 2003). 3  M Poiares Maduro, ‘Contrapunctual Law: Europe’s Constitutional Pluralism in Action’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003), see also ibid, ‘Europe and the Constitution: What if this is As Good As It Gets?’ in JHH Weiler and M Wind (eds), European Constitutionalism Beyond the State (Cambridge, Cambridge University Press, 2003). 4   M Kumm, ‘The Jurisprudence of Constitutional Conflict: Constitutional Supremacy Before and After the Constitutional Treaty’ (2005) 11 European Law Journal 262. 5   See N Walker, ‘The Idea of Constitutional Pluralism’ (2002) 65 Modern Law Review 317. For an overview of the state of the debate see also M Avbelj and J Komárek (eds), ‘Four Visions of Constitutional Pluralism – Symposium Transcript’ (2008) 2 European Journal of Legal Studies 325. 6   A von Bogdandy, ‘The Past and Promise of Doctrinal Constructivism: A Strategy for Responding to the Challenges Facing Constitutional Scholarship in Europe’ (2009) 7 International Journal of Constitutional Law 364; S Besson, ‘From European Integration to European Integrity: Should European Law Speak with Just one Voice?’ (2004) 10 European Law Journal 257. 7   For a scepticism about using a constitutionalist framework for the construction of law beyond the state, see N Krisch, Beyond Constitutionalism: The Pluralist Structure of Postnational Law (Oxford, Oxford University Press, 2010).

Rethinking Constitutional Authority  41 Grainne de Búrca8 and Daniel Halberstam together with Eric Stein9 to some extent do, analyse whether the ECJ or at least the Advocate General’s opinion has appropriately engaged and taken into account the prerogatives of the international order when articulating its position and what exactly that might mean in practical terms under the circumstances. The following will leave these questions aside. The focus here will be a second question that is in some ways more basic, even though those writing about constitutional pluralism have generally not focused on it.10 There is what some have regarded as a fashionable disposition to embrace legal pluralism11 without providing an account of the conditions under which it is desirable and the conditions under which it is not. It is not very plausible to believe that legal pluralism is always attractive. To illustrate the point, think of the context of a classical federal state and its relationship to its constituent units. If the state court of New York were to start questioning the supremacy of the US Constitution as a matter of New York State law, most would probably hesitate to conclude that state courts have at last embraced the virtues of legal pluralism and instead insist that something has gone awry. The same is true if Bavaria were to start subjecting the law of the Federal Republic of Germany to state constitutional requirements. Of course the context is different when the question turns to the relationship between UN law and EU law or EU law and the law of Member States. But which features of the classical federal context that are different from the contexts of the relationship between EU law and Member States law or EU law and UN law are relevant for determining whether to describe the relationship in pluralist rather than hierarchically integrated monist terms? And even if we were to simply accept the proposition that constituent units are hierarchically integrated within a federal structure exactly because it is the structure of a state, nothing at all is gained for thinking about the relationship between the EU and the UN, both of which are not states. The general question, then, is the following: Under what circumstances is it appropriate to conceive of the relationship between different legal orders in pluralist terms, rather than thinking about them in terms of hierarchical integration? Under what circumstances should, for example, the legal order of the European Union and the legal order of Member States be regarded as a hierarchically integrated whole? Under what circumstances should the European legal order be regarded as an integrative element of an international legal order? What are the conditions under which legal pluralism is a virtue and what are the conditions under which it is a vice? 8   G de Búrca, ‘The European Court of Justice and the International Legal Order after Kadi’ (2009) 51 Harvard International Law Journal 1. 9   D Halberstam and E Stein, ‘The United Nations, the European Union, and the King of Sweden: Economic Sanctions and Individual Rights in a Plural World Order’ (2009) 46 Common Market Law Review 13. 10   But see Kumm, n 4. 11   See JHH Weiler, ‘Prologue: Global and Pluralist Constitutionalism – Some Doubts’ in G de Burca and JHH Weiler (eds) The Worlds of European Constitutionalism (Cambridge, Cambridge University Press, 2011) 8; J Baquero Cruz, ‘The Legacy of the Maastricht Judgment and the Pluralist Movement’ (2008) 14 European Law Journal 389.

42  Mattias Kumm From an external point of view – the point of view of a legal sociologist – the answer to the question whether a pluralist or hierarchically integrated relationship between legal orders exists is relatively easy to give. If the highest court of a legal order insists on applying the law of the more encompassing legal order only under conditions defined by its legal order and the decisions of that court are generally taken as authoritative by other officials of that legal order, then the relationship between the legal orders is pluralist as a matter of fact. But the perspective to be taken here is not the external perspective of the sociologist, but the internal point of view taken by a participant in legal practice. Here the question changes: Under what circumstances should a judge on the highest court of a legal order – the ECJ when adjudicating questions involving UN law as in the Kadi case, or a national constitutional court adjudicating questions involving EU law – accept or refuse hierarchical integration in the more encompassing legal order and insist on the application of its own internal standards? Or to put it another way: Under what circumstances should a court accept the hierarchical integration in the more encompassing legal order? If the ECJ was right to insist on the autonomy of the European legal order in Kadi, what would need to happen for the ECJ to accept that primacy of UN law as indicated by Article 103? If Member States’ courts are right not to accept the ECJ’s claim that EU law has primacy over national law at the current state of integration, what would need to happen for that to change? What is the right conceptual framework for thinking about these issues? The following will explore these questions from the internal perspective. I will argue that the proper place and limits of constitutional pluralism depends on competing theories of public law that are embedded in European constitutional practice. As it turns out, there are three different frameworks used by different courts at different times, each giving a different answer to the question. I will refer to them as Legal Monism (I), Democratic Statism (II) and Cosmopolitan Constitutionalism (III) respectively. Whereas Legal Monism is sceptical about any kind of legal pluralism and can analyse it only it as a case of law in crisis, Democratic Statism insists on a dualist construction of the legal world – the law within states and the law through which states relate to one another. Pluralism here takes the form of a dualist account of the legal world: the state orders are internally unified, as is the world of international law. But both of them constitute independent legal orders, for so long as the international legal order is not replaced by a democratic world state. The relationship between Member States and the European Union will remain pluralist if and for so long as the European Union is not a state. And until that time the European Union remains a Treatybased organization that is part and parcel of the international legal order. In fact, strictly speaking, there is no space for ‘constitutional’ pluralism, because constitutionalism is associated with the establishment of the supreme law of the state. The international legal order can’t, therefore, qualify as a constitutional order properly so called, because it is not a state. In practice there is only the national Constitution as the supreme law of the land that determines the conditions under which international law will be applied domestically according to whatever

Rethinking Constitutional Authority  43 standards that Constitution provides. Finally, Cosmopolitan Constitutionalism is the name of a conceptual framework that can help make sense of the idea of constitutional pluralism both analytically and normatively, while also establishing the conditions under which hierarchical integration is preferable to pluralism. The refusal of a legal order to recognize itself as hierarchically integrated into a more comprehensive legal order is justified, if that more comprehensive order suffers from structural legitimacy deficits that the less comprehensive legal order does not suffer from. The concrete norms governing the management of the interface between legal orders are justified, if they are designed to ensure that the legitimacy conditions for liberaldemocratic governance are secured. In practice that means that there are functional considerations that generally establish a presumption in favour of applying the law of the more extensive legal order over the law of the more parochial one, unless there are countervailing concerns of sufficient weight that suggest otherwise.

II.  General Scepticism about Legal Pluralism: Legalist Monism Legalist Monism12 is the position that underlies the ECJ’s jurisprudence in Costa v ENEL, justifying the primacy of EU law over all national law, including national constitutional law. The following will briefly analyse the justification of that position in Costa v ENEL (1) and then assess what the implication of such an account of the world of public law would be for assessing the relationship between EU and UN law (2). As will become apparent, there is a tension between Costa and Kadi. The Legal Monist position in Costa should have led the ECJ to confirm the ECFI’s decision that it should not review acts implementing UN law on European fundamental rights grounds. The ECFI was correct under this conception to recognize that UN Law and EU Law should be conceived as hierarchically integrated, just as EU Law and the law of Member States are conceived as hierarchically integrated.

A.  The Primacy of EU Law As every student of EU law knows, since the 1960s the ECJ has consistently held that, in case of a conflict between European and national law, Member States’ courts are under an obligation to set aside all national law, even national constitutional law.13 For all practical purposes European law as interpreted by the ECJ is claimed to be the supreme law of the land. 12   The classic literature on Monism includes Hans Kelsen, General Theory of Law and State (Cambridge, Harvard University Press, 1945) 363–80; Alfred Verdross, Die Einheit des rechtlichen Weltbildes auf Grundlage der Völkerrechtsverfassung (Tübingen, JCB Mohr, 1923); Sir Hersch Lauterpacht, International Law and Human Rights (New York, FA Praeger, 1950). 13   Case 6/64 Costa [1964] ECR 585; Case 45/76 Comet [1976] ECR 2043; Case 106/77 Simmenthal [1978] ECR 629.

44  Mattias Kumm The ECJ has supported that claim with the proposition that the EC Treaties established a new legal order and later referred to the Treaties as Europe’s ‘constitutional charter’.14 To substantiate this claim, the ECJ employed three arguments. First, the Court makes a conceptual argument. If the Treaty is to establish legal obligation properly so called, it can’t be permissible for a Member State to unilaterally set it aside unless authorized to do so by EU law. Such unilateral unauthorized action would undermine the status of EU law as law properly so called binding on all Member States. This argument is of Kelsenian heritage. According to Kelsen, the world of law has to be conceived in monist terms. Taking a legal point of view is incompatible with the claim that from the point of view of one legal order (say, the European legal order) x is the case, but from the point of view of another legal order (the legal order of Member States), y is the case. There can be no coexistence of different legal systems constituted by different ultimate legal rules. The world of law is unified not as an empirically contingent matter, but as a conceptual matter.15 This is not the place to provide a comprehensive discussion and critique of the argument. Here it must suffice to point out that the argument is not at all obvious. It is not clear why it would undermine the status of EU law as law that there is another legal system that incorporates EU law on its own terms. It is unclear why it should be conceptually impossible, as opposed to, say, undesirable on pragmatic grounds, to imagine the legal world in pluralist terms. What is conceptually wrong with acknowledging the possibility of the existence of different legal orders, each of which recognizes the authority of the law of the other on its own terms? There does not have to be only one legal point of view, even though it might be desirable that there be only one on other normative grounds. Member States may or may not be doing the right thing if they insist on determining the status of EU law in light of their own national constitutional requirements. But they are not thereby undermining the status of EU law as law properly so called. The second argument put forward by the ECJ is, at first sight, no less mysterious. The Court lists a number of features of the Treaties that distinguish it from ordinary Treaties of international law. Within the considerable competencies defined in the Treaty, EC institutions can enact legislation directly binding citizens in Member   Case 294/83 Les Verts v Parliament [1986] ECR 1339.   According to Kelsen, the demand that the world of law be monist does not require a Grundnorm according to which transnational law trumps national law. It is also possible to posit a Grundnorm according to which there is no law except as established by national law. But the choice of the latter Grundnorm would imply a kind of national solipsism, comparable to a Cartesian subject who denies the existence of other such subjects and claims that all other persons are nothing but emanations of his consciousness. Adopting such a position is logically possible. Notwithstanding significant efforts by generations of philosophers, solipsism is a position remarkably difficult to refute conclusively. On the other hand, there are just no plausible arguments in its favour, so only the most anxious foundationalists have been troubled by the problem. Others have been happy to assume the reality of other subjects as the more plausible starting point. Similarly lawyers, it might be implied from Kelsen, would do well to posit that there is a European legal order and that the legal orders of Member States are equally subject to it. But of course, the whole argument depends on the non-obvious claim that there can only be one legal order and that the idea of legal pluralism is untenable for conceptual reasons. That is an issue not to be deepened here. 14 15

Rethinking Constitutional Authority  45 States. Furthermore, since the enactment of the Single European Act, most decisions concerning the Common Market and increasingly in other domains as well are made following a non-consensual procedure that allows valid legislation to be passed by qualified majority vote and the participation of a European Parliament, even in the face of resistance by powerful Member States. European law, then, has emancipated itself in its day-to-day workings from its international law foundations and the idea of state consent. It has established its own autonomous legal order. It is not easy to make sense of this argument. It might be understood in one of two ways. First, it could be understood as a weak claim that the EU is sufficiently different from ordinary Treaties that it should not be assessed within the conventional statist paradigm, which constructs international law in contractualist terms based on state consent. But for that argument to be persuasive, it would have been necessary to point out how exactly the highlighted features and the relative autonomy of EU law are relevant to the claim of primacy. The Treaties of Rome might be different from most run-of-the-mill Treaties, but they are still Treaties and not a genuine Constitution of a new state. At least that would be the claim made by a democratic statist. What exactly is wrong with that claim? The ECJ does not say. But the listing of ways in which the EU is different from other Treaties might just have the function to open the door to the third argument, which lies at the heart of the case for primacy. It turns out that the particular features of the EU and the autonomy of the legal order matter for functional reasons. A Treaty regime that has these features and has been designed to fulfil ambitious purposes – including the establishment of a common market – can only function in a coherent way if the laws it establishes have primacy over national law. Without primacy, the effective and uniform enforcement of Europe’s laws would be endangered. The purpose of the European Union to fulfil its various objectives, including the establishment of a common market, could not be achieved if Member States did not accept the primacy of EU law. An aggressively purposive interpretation of the Treaties leads the ECJ to conclude that EU law has to be accepted as the supreme law of the land in order to fulfil its purpose. This is an empirically contestable claim about the consequences of not accepting the unconditional primacy of EU law connected to a normatively contestable functional account of the basis of public authority. The claim to primacy is further strengthened by two more general functional arguments, not explicitly made in Costa, but standard fare in the literature supporting that decision. First, there is the idea that legally constraining the relationship between Member States is an effective remedy against the great evils that have haunted the continent throughout much of the nineteenth and the first half of the twentieth century: clashes of interest between nation states have a dangerous propensity to degenerate into bloody wars. Within the framework of a coherent legal order, the definition, articulation and negotiation of national interests occurs in such a manner as to make such a development highly unlikely. Second, legal integration can be seen as a mechanism which tends to immunize nationally organized peoples from the kind of passionate political eruptions that have led to totalitarian

46  Mattias Kumm or authoritarian governments and/or discrimination of minorities that have characterized European history in the nineteenth and twentieth centuries. This could not be achieved to the same degree, so the argument goes, if the final decision concerning what is to be applied as law in a Member State rests on a decision ultimately made by Member States themselves. There is no space for Legal Pluralism.

B.  EU Law and International Law: Kadi Whatever the merits of these arguments might be, what are the implications of such a position for deciding whether the ECJ should subject a regulation implementing a UN Security Council obligation to EU fundamental rights law? It turns out that the arguments for the primacy of EU law in Costa should also push the ECJ towards recognizing the primacy of UN law over EU law. First, if the conceptual argument is plausible in the context of the relationship between EU law and the law of Member States, there is no reason why it should not also be determinative in the relationship between UN law and EU law. If the fact that Member States subject EU law to national constitutional standards subverts the very idea that EU law is law properly so called, does not the fact that the ECJ effectively subjects UN resolutions to review based on EU standards undermine the very idea that UN law is law properly so called? Second, there are many features of the UN that distinguish it from an ordinary Treaty. The UN Charter establishes complex institutions with their own competencies, it allows the UN Security Council to create new legal obligations following a non-consensual procedure. Is the UN Charter not also a constitutional charter of an autonomous legal order? Of course there are differences between the UN and the EU: the ECJ has compulsory jurisdiction over issues of EU law, for example, whereas the ICJ does not have compulsory jurisdiction over issues arising under the UN Charter. There is no comparable doctrine of direct effect with regard to UN law. But not all differences suggest that the EU is more of an ‘autonomous legal order’ than the UN. Some differences point in the opposing direction. Take the name: the UN is called a Charter, not a Treaty. Think of the explicit claim to primacy in Article 103 UN Charter, a claim still not explicitly made by the EU Treaties even after the latest rounds of reform.16 At any rate, whatever differences may exist between the UN and the EU, it remains unclear why these differences should be decisive for the purpose of establishing that the EU is an autonomous legal order that makes a claim to primacy, whereas the UN is not. Third, the functional reasons supporting primacy of EU law over Member States law also support the claim to primacy of UN law over EU law. If the UN is to effectively succeed in ‘maintaining peace and security’ and if the UN Security Council is to effectively take up its ‘primary responsibility for the maintenance of international peace and security’, does that not require that other 16   Of course, the failed ‘Treaty Establishing a Constitution for Europe’ had a qualified supremacy clause. The Lisbon Treaty on the other hand merely confirms the ECJ’s jurisprudence in Declaration No 115 attached to it.

Rethinking Constitutional Authority  47 actors accept the primacy of UN Security Council Resolutions? Would anything else not undermine the effective and uniform enforcement of UN law? The judgment of the ECFI, with its emphasis on functional arguments, to a large extent reflects the Legal Monist positions embraced by the ECJ in Costa. But the position of the ECJ does not. Is the only way to make sense of the ECJ’s Kadi decision to understand it as the ECJ having taken a statist turn, albeit a statist turn without the resources of democratic constitutional theory to back it up? Part IV will argue that there is a more generous interpretation of Kadi that connects it to a Cosmopolitan Constitutionalist conception of Public Law.

III.  The Classical World of Public Law: Democratic Statism and Deep Dualism A.  Statism in Three Historical Versions: Conceptual, Realist and Democratic The core feature of statist accounts of the world of law is a deep divide: there is national or state law on the one hand and there is international or interstate law on the other. State law is the paradigmatic case of law. International law is the impoverished stepchild of state law. The former is in some sense derived from the latter and yet it seems lacking in some basic way. Historically, three different accounts of that divide were offered, distinguishable by their account of what exactly international law lacked. During much of the nineteenth century, legal theorists spent a great deal of time grappling with the conceptual question whether international law properly so called could exist, or whether it was really just a kind of positive morality, given the absence of an international sovereign. If law was the command of a sovereign17 or sovereignty was a predicate that precluded being subject to legal obligations,18 how could international law exist as law properly so called? Even though conceptual arguments lost their sway in the twentieth century, after the Second World War so-called ‘realism’19 provided what was believed to be an empirically grounded account of why the structure of the international system made the establishment of the rule of law beyond the state a utopian exercise. In an international system where states simply followed their national interests,   J Austin, The Province of Jurisprudence Determined (London, John Murray, 1832).   See eg A Lasson, Prinzip und Zukunft des Völkerrechts (Berlin, Hertz, 1871). The argument was actually pleaded and dismissed by the Permanent Court of Justice, the predecessor of the ICJ; see Permanent Court of International Justice, Judgment of 17 August 1923, SS Wimbledon (UK v Japan), PICJ Series A, No 1 at 25. 19   For a classical statement see HJ Morgenthau, Politics Among Nations: The Struggle for Power and Peace 5th edn (New York, Alfred A Knopf, 1978) 4–15. 17 18

48  Mattias Kumm international law could not function as an independent guide or constraint for state action. But by the end of the Cold War, with the spread of failed states on the one hand and the rise of relatively effective Treaty regimes and practices of global governance on the other, that argument too fell out of favour. The terms of the debate shifted again. General scepticism was in retreat as a rich literature burgeoned that tried to explain the widespread phenomenon of compliance with international law.20 As liberal constitutional democracies were increasingly constrained perhaps not by a Weberian iron cage but at least a strong web of transnational legal norms, statism took a democratic turn: now the deep divide between national law and international law was justified with reference to democratic constitutional theory. State law ultimately derives its authority from ‘We the People’ imagined as having acted as a pouvoir constituant to establish a national Constitution as a supreme legal framework for democratic self-government.21 International law, on the other hand, derives its authority from the consent of states. Of course states may decide to establish all kinds of international institutions to address specific issues. They can also establish courts and tribunals and even establish Treaties that create rights and obligations for individuals. But no matter how important these Treaties might be to address basic collective action or coordination problems, no matter the internal complexity of the institutions they set up: none of this takes away from the fact that these international institutions are ultimately based on Treaties requiring the ratification by states following national constitutional requirements. Member States remain the masters of international Treaties, so long as they are not replaced by genuine constitutions attributable to a constitutive act of ‘We the People’. There are two important consequences connected to a construction of the legal world informed by Democratic Statism. First, Democratic Statism insists that national constitutional law, as the supreme law of the land, determines if and under what conditions international law is to be enforced domestically. The authority of international law, from the perspective of national law, remains a matter to be determined by national constitutional law. Of course a violation of international law may trigger the responsibility of the violating state on the international level, but as a matter of domestic law such a consequence might well be legally irrelevant. The legal world thus has a dualist structure. Second, given that state law is connected to the idea of ‘We the People’ governing themselves democratically and the institutional and social infrastructure for collective selfgovernment is absent beyond the state, international law is inherently infected by a democratic deficit. That deficit is less when there is a close and concrete link between a specific international legal obligation and state consent. But even then there is a residual problem, because many international obligations can’t be unilaterally revoked by the state as a matter of international law, even when the 20   An overview of the debate as it stood in the 1990s and further references can be found in HH Koh, ‘Why Do Nations Obey International Law?’ (1997) 106 Yale Law Journal 2634. 21  See E-J Sieyès, ‘Qu’est-ce que le tiers-état?’ in Roberto Zapperi (ed), Ecrits Politiques (Paris, Archives Contemporaines, 1985).

Rethinking Constitutional Authority  49 majority of citizens using democratic procedures want to do so. Problems of democratic legitimacy become even more serious when Treaties authorize international institutions to make important social and political choices. Even if ultimately problem-solving or co-operation-enhancing benefits associated with international law may legitimate international law, there remains an aura of illegitimacy that hangs over international law.

B.  Democratic Statism and the ECJ’s Claim to Primacy over Domestic Constitutional Law When the ECJ made the claim in Costa v ENEL that EU law has primacy over national law, even national constitutional law, the court implicitly rejected Democratic Statism and embraced an alternative account of public law as was described above. Here the focus is on how to make sense of such a claim from the point of view of Democratic Statism, a framework adopted by some Member States’ highest courts, perhaps most prominently the German Federal Constitutional Court in its Maastricht22 and Lisbon23 decisions. Within the framework of Democratic Statism a claim to primacy is plausible only if the European Union has in fact become a federal state with a Constitution properly so called. The central question becomes: is the EU an international organization, ultimately based on Treaties deriving their authority from the ratification of Member States according to their national constitutional requirements? Or has the EU become a federal state, based on an act by ‘We the People’ acting as a pouvoir constituant? If it is the former, EU law cannot, at least as a matter of domestic law, claim primacy over national constitutional law. If it is the latter, Member States have lost their ultimate authority as their national constitutional order has been hierarchically integrated into the legal order of the new federal superstate. Note that within the democratic statist model there is no third alternative. The EU qualifies either as a state or as an international institution. Calling the EU an institution sui generis, as has become customary in EU law circles to avoid asking that question, just clouds the issue. The EU may be sui generis in all kinds of ways, but it will be either a sui generis state or a sui generis international institution, tertium non datur. The question is whether it is one or the other.

1.  The Test for Establishing Statehood: Why the EU is no State So how do you know whether it is one or the other? Courts and scholars using a Democratic Statist framework generally focus on some combination of three factors.24 The first is institutional and focuses on the structure of the EU Treaties.   Case 2 BvR 2134, 2159/92 Maastricht [1993] BVerfGE 89, 155.   Case 2 BvE 2/08, judgment of 30 June 2009, Lisbon, English translation http://www.bverfg.de/ entscheidungen/es20090630_2bve000208en.html. 24   See Kumm, n 4. For an account of a similarly structured discussion in the US pre-Civil War context see D Golove, ‘New Confederalism: Treaty Delegations, Executive and Judicial Authority’ (2003) 55 Stanford Law Review 1697. 22 23

50  Mattias Kumm Here the general focus tends to be whether Member States can still plausibly be described as the ‘Masters of the Treaty’ or whether EU institutions have emancipated themselves from the control of Member States to a sufficient degree. The focus is on a variety of factors that include but are not limited to the amendment procedure (is the unanimity required to amend the Treaties or is a qualified majority enough?), the ordinary legislative procedure (is the Council in charge or does Parliament dominate the legislative process and even if it does, is it constituted in a way that reflects the idea of equality of citizens?), competencies (how far does EU law authorize legislation in core traditional areas of sovereignty, such as taxes, defence, social security and criminal law?). The second factor is procedural. Were the Treaties the result of an ordinary treaty ratification procedure or was there a constitutional convention or some other mechanism which allowed for the kind of high-level participation and deliberation associated with actions appropriately attributable to ‘We the People’? Third, there are sociological factors: do EU citizens have the kind of cohesion, do they share the kind of bond char­acteristic of ‘the people’? Do they have what it takes to be a ‘demos’? To determine whether that is the case, some authors focus on shared history, culture, religion, language, etc. Others focus on the structure of the public sphere and the institutions of civil society relating to media, political parties, interest group organizations, etc. Here these differences can’t be addressed and it must suffice to point to the structure of the argument. Even though an application of one or another of these factors might give rise to debate, there is not a single court or author I am aware of that has embraced the Democratic Statist framework and then concluded that, on application, the EU qualifies as a state. Among Democratic Statists there seems to be a consensus that the EU is based on Treaties, not a Constitution properly so called, and that it qualifies not as a state, but as an international institution. Democratic Statists thus conclude that the primacy claim made by the ECJ is mistaken, at least if it is understood as a claim that national courts should apply EU law even in the face of opposing national constitutional norms. Since there is no European Constitution plausibly grounded in an act of a European pouvoir constituant and ensuring the democratic self-government of a European people, European law can not plausibly be conceived as the supreme law of the land. Recognizing the position of the ECJ would undermine commitments to democratic self-government central to Democratic Statism. Instead, EU law ultimately derives its authority from Member States who have ratified the Treaties according to their national constitutional requirements. The status of EU law as a matter of domestic law depends ultimately on what the national Constitution determines. EU law trumps national law only to the extent prescribed by the national Constitution. National constitutional law remains the supreme law of the land and the European legal space remains pluralist.

2.  Consequences for the Domestic Application of EU Law Where does that leave the application of EU law by national courts? Those who adopt a Democratic Statist framework insist that such a commitment is distinct

Rethinking Constitutional Authority  51 from national constitutional parochialism. It does not necessarily entail a commitment to national constitutional doctrines that are inimical to the application of EU law and the functional imperatives of European integration. States might well adopt an ‘open Constitution’, allowing for far-reaching openness to and engagement with the international legal order. Fears invoked by those advocating European law’s primacy in the name of ensuring the effective and uniform enforcement are overblown. It is a mistake to believe that only the recognition that EU law is the supreme law of the land ensures the effective and uniform functioning of EU law. National constitutions may well contain norms that specifically authorize the enforcement of EU law in most cases. There is no problem inherent to the idea of constitutional self-government as it is conceived by Democratic Statists. The real question is merely how to conceive of the self that governs itself constitutionally25 and how that identity translates into national constitutional conflict norms. The core point is this: following the disasters of the First and Second World Wars, the national selves that govern themselves constitutionally have committed themselves to constitutionally tolerate the European laws enacted collectively by Member States and their peoples. On this, authors ranging from conservatives such as Paul Kirchhof,26 Judge Rapporteur of the Maastricht judgment,27 to liberal constitutionalists, such as Dieter Grimm28 or Joseph Weiler,29 notwithstanding important differences among them, would agree. The idea of constitutional tolerance lies at the heart of what makes European integration possible and describes its normative core.30 Most national constitutions have been amended in the process of European integration. All of them are interpreted by national courts to require the enforcement of EU law, even when it conflicts with national statutory law. Constitutional tolerance of EU law – the openness of national legal orders to EU law – is hard-wired into national constitutional law as it is interpreted by Member States’ highest courts. But tolerance remains very much a feature of the national Constitution and those who interpret it. Conceptually, Democratic Statism and its connection between the supremacy of the Constitution and the idea of constitutional self-government remain untouched. Focusing on the Schmittian question – who has the final say? – misses the point. It obscures the remarkable fact that in Europe the everyday enforcement of European law is guaranteed by national constitutional provisions and their interpretation by national courts. The true innovation in European integration is the not the establishment of a new ultimate constitutional authority on the European level. And it is not the 25   The idea and its deeper normative significance is developed in JHH Weiler, ‘To Be a European Citizen: Eros and Civilization’ in Weiler, The Constitution of Europe (Cambridge, Cambridge University Press, 1999). Practical doctrinal implications for thinking about constitutional conflict within this paradigm are developed in M Kumm, ‘Who is the Final Arbiter of Constitutionality in Europe?’ (1999) 36 Common Market Law Review 356. 26  P Kirchhof, ‘The Balance of Powers Between National and European Institutions’ (1999) 5 European Law Journal 225. 27  See Maastricht, n 22. 28   D Grimm, ‘Does Europe Need a Constitution?’ (1995) 1 European Law Journal 282. 29   Weiler, n 2. 30   Ibid.

52  Mattias Kumm abdication of national constitutional authority. Europe’s genius and the key to understanding its sui generis character lies in the reinterpretation of national constitutional traditions to reflect a commitment to constitutional tolerance. National constitutional authority, structured and exercised to reflect a commitment to constitutional tolerance, lies at the heart of the European integration process. The challenge is to amend and interpret national constitutions in such a way that it reflects appropriate respect for national democratic commitments, while enabling appropriate engagement with European law. European integration is inherently beset by a tension between genuine democratic self-government that takes place on the national level, and functional considerations that justify some delegation of powers to the European Union and some degree of opening up of the national legal orders to EU law. That tension has to be carefully calibrated and reflected in the constitutional doctrines that national courts as ultimate guardians of constitutional legality enforce. EU law, no doubt in many ways sui generis, ultimately remains treaty-based international law and not constitutional law properly so called.

C.  Democratic Statism and the Relationship between UN Law and European Law Democratic Statists would have an easy time addressing the issue whether the ECJ had the authority to effectively review the UN Security Council Resolution on European fundamental rights grounds. The EU is, like the UN, a treaty-based organization. It came about by states negotiating, signing and ratifying a treaty according to their national constitutional requirements. Given that both treaties derive their authority from the same source and are both equally international law, the relationship between the two cannot be resolved with reference to sourcebased conflict rules. Instead, the issue is one of conflicting treaty provisions. Conflicts between treaties are resolved first of all with reference to stipulations made by the treaties themselves about their status in case of conflict.31 Luckily in this case, both treaties contain concurring propositions that indicate what should be done in case of conflict. Both the UN Treaties and the EU Treaties require or permit that EU primary law is not applied to prevent the effective implementation of UN law. On the one hand, Article 103 UN Charter in conjunction with Article 25 UN Charter clearly stipulates that, in case of a conflict between the obligations of the Members of the United Nations under the Charter and their obligations under any other international agreement, the Charter shall prevail. On the other EU side, Article 307 EC Treaty specifically provides that nothing in the Treaty is to be considered as incompatible with previously assumed international obligations 31   For the interpretation of treaties, see Arts 31–33 of the Vienna Convention on the Law of Treaties (VCLT). Note that Art 30 VCLT contains a specific set of conflict rules governing treaties relating to the same subject matter.

Rethinking Constitutional Authority  53 by Member States.32 Since UN obligations were assumed before Member States joined the EU, the EU Treaties should not preclude the application of UN law. Other provisions specifically accommodate or create a space for co-operation with the UN. Nothing suggests that the Treaties should have primacy over the UN. Clearly, therefore, the ultimate conclusion reached by the ECFI that UN law is not to be effectively subjected to EU law fundamental rights review is correct and the position of the ECJ is wrong. The position of the ECJ in Kadi does become more plausible, however, if one starts off with the assumption that the EU is a state and that EU primary law is, like the Constitution of a state, the supreme law of the land. Is that how the ECJ justifies its claim that it can effectively subject UN law to EU fundamental rights review? At the crucial juncture of the decision the court is remarkably obscure in its reasoning and apodictic in its formulation: the EU is committed to the rule of law and cannot avoid reviewing acts it undertakes on the basis of its own constitutional charter, which establishes an autonomous legal system.33 Obligations imposed by international agreements cannot change the allocation of powers under that constitutional charter and cannot have the effect of prejudicing constitutional principles that form part of the foundations of the Community.34 Ultimately the ECJ is only pronouncing itself on an EU measure, not a UN measure, and it is doing so applying EU fundamental principles. One obvious criticism to this type of rhetoric is that it is statist in the worst sense: it is not even democratic statist. It may sound like some of the more recalcitrant Member States’ courts asserting the supremacy of their national constitutions, when confronted with the ECJ’s claim that EU law takes primacy over national law, including national constitutional law. The only difference is that the rhetoric of ‘the EU as an autonomous legal order’ substitutes the invocation of state sovereignty. But notice how this kind of argument resembles a pre-­democratic conceptual statism: a conceptual claim – this time not ‘statehood’ or sovereignty’, but the idea of an ‘autonomous legal order’ – substitutes for an argument relating to ‘We the People’ and democracy. The idea of an autonomous legal order is not enriched by arguments about what it is about an order that has certain features that justifies according primacy to it. Read in this way, the ECJ’s Kadi decision is even worse than national constitutional decisions that make claim to supreme authority. If you take the more articulate cases of national recalcitrance, perhaps best exemplified by the German Federal Constitutional Court in its Maastricht 35 and more recent Lisbon decision36 these decisions at least provide a theoretical basis for their recalcitrant approach: they provide for a Democratic Statist constitutional

32   According to Art 30(2) VCLT, when a treaty specifies that it is not to be incompatible with an earlier treaty, the provision of the earlier treaty shall prevail. 33   Kadi, n 1, paras 281 and 282. 34   Ibid, paras 282 and 285. 35   Maastricht, n 22. 36   Lisbon, n 23.

54  Mattias Kumm theory. Democratic Statism may be ultimately unconvincing, but it is made explicit and allows for serious engagement and criticism.37 Of course it is not surprising that the ECJ does not draw on Democratic Statism to justify the EU’s primacy over UN law. Arguments about democracy and the role of a pouvoir constituant would not resonate when applied to the EU. As discussed above, the perceived absence of a demos and the perception that the EU is based on treaties rather than an act of a European pouvoir constituant has made it plausible for many national courts to insist on limiting the extent to which the primacy claim of the ECJ is enforced over specific national constitutional commitments. Not surprisingly, the ECJ in Costa did not rely on democracy-based arguments to justify the primacy of EU law over Member States’ law either. So if the ECJ did not embrace Democratic Statism to justify its claim to primacy in Costa v ENEL, what did it rely on? And how might that justification fit in with the Court’s position with regard to UN law in Kadi?

IV.  Cosmopolitan Constitutionalism: The Conditional Embrace of Legal Pluralism There seems to be an irresolvable tension between Kadi and Costa, just as there seems to be an irresolvable tension between Costa and most of the decisions of national constitutional courts, which insist on drawing constitutional red lines in the sand. But there is a way to re-read Kadi, Costa and national constitutional court decisions that suggests that all three in fact embrace a very similar con­ ception of public law. But that conception is neither Democratic Statist, nor is it Legal Monist. Instead, it is what I call Cosmopolitan Constitutionalist. Cosmopolitan Constitutionalism refers to a position according to which a set of universal principles central to liberal democratic constitutionalism undergird the authority of public law and determine which norms take precedence over others in particular circumstances. Contrary to Democratic Statism it is not the case that the idea of ‘democracy’ and ‘national self-government’ connected to statehood and sovereignty provides the decisive principle to help determine where ultimate authority lies. But nor is it the case that the idea of legality and the functional reasons in support of it are sufficient to justify a Monist account of the legal world. Instead, constitutionalism provides the following framework for the analysis of constitutional conflicts. First, Legal Monism is right that the idea of legality – respect for the rule of law and the co-ordinating and co-operative practices it makes possible – is central also beyond the state and plausibly provides for a presumption of some weight: that the law of the more expansive community should be respected by public   Eg Kumm, n 4.

37

Rethinking Constitutional Authority  55 authorities, regional or national law to the contrary notwithstanding. There is a presumption that UN law trumps EU law and that EU law trumps Member States’ law ultimately grounded in functional considerations. Given the collective action and co-ordination problems that these regimes are trying to solve, which states individually are unable to address effectively, the resolutions provided for by the regimes ought to carry with them a presumption of authoritativeness. But in liberal democracies, legitimate authority is not tied to formal and functional considerations alone. It is also tied to procedural and substantive requirements that are reflected in constitutional commitments to democracy and the protection of rights. That does not mean that the authority of UN law, from the perspective of EU law, should be determined exclusively by EU primary law or that the authority of EU law should be determined exclusively by Member States’ constitutions. Both Legal Monism and the dualist conception of the legal world provided by the statist version of national constitutionalism ultimately provide one-sided and thus unpersuasive accounts of the principle of legality. Instead, the refusal of a legal order to recognize itself as hierarchically integrated into a more comprehensive legal order is justified, if that more comprehensive order suffers from structural legitimacy deficits – relating to human rights protection of democratic legitimacy – that the less comprehensive legal order does not suffer from. The concrete norms governing the management of the interface between legal orders are justified, if they are designed to ensure that the legitimacy conditions for liberal-democratic governance are secured. In practice that means that there are functional considerations that generally establish a presumption in favour of applying the law of the more extensive legal order over the law of the more parochial one, unless there are countervailing concerns of sufficient weight that suggest otherwise. To illustrate the idea of Cosmopolitan Constitutionalism as a distinct model of the world of public law, I will first provide an alternative account of the approach taken by some Member States’ courts in their engagement with the ECJ’s primacy claim (1), then revisit Costa (2) and finally provide another reading of Kadi (3). The point is to illustrate how central elements of both decisions can be understood as reflecting a commitment to Constitutionalism, rather than Statism or Legalist Monism. Neither of the latter two can describe very well the contemporary place of European law in the world of public law. The constitutionalist model of the world of public law, on the other hand, is descriptively more powerful.

A.  Constitutionalism and the German Federal Constitutional Court’s Response to Costa One important consequence of conceiving of legal authority as a function of the realization of a set of principles is that whether or not EU law should be recognized as having primacy over national constitutional law is a question that allows for qualified answers. It admits to more answers than just yes or no. Even if European Union law does not, without some qualification, establish the supreme

56  Mattias Kumm law of the land, it could still effectively reconstitute legal and political authority in Europe. The authority of EU law is possibly a question of degree. It may depend on the degree to which constitutional principles are realized by EU institutions. It admits to the possibility that neither EU law nor national constitutional law effectively establishes the supreme law of the land. In this way constitutionalism can help shed light on a fascinating aspect of national courts’ reception of EU law. For the most part national courts have not accepted that EU law is the supreme law of the land. But nor have they simply assumed that national constitutional law is the supreme law of the land. There is something deeply misleading in claiming that, to the extent that national courts have not accepted EU law as the supreme law of the land, they are merely interpreting their national constitutions to establish what the status of EU law should be as a matter of domestic law. National courts have generally adopted an intermediate position. They generally accept neither EU law nor the national Constitution as the supreme law of the land. Instead they look to both EU law and the national Constitution and try to make sense of what the best understanding of the competing principles in play requires them to do.38 Theirs is a constitutionalist conception of public law. To flesh out what this means practically and provide an example, a brief and somewhat schematic analysis of the German Federal Constitutional Court’s approach to the authority of EU law will follow. The jurisprudence of the Court may be widely known by European Union lawyers, but its reconstruction in terms of a commitment to constitutionalism may prove illuminating, even if it can only be brief and schematic. The German Constitution, until the early 1990s,39 contained no specific provisions addressing European integration, though the Preamble mentions Germany’s commitment to strive for peace in a united Europe. The Constitution did authorize Germany to enter into treaties establishing international institutions.40 And it contained general provisions giving international treaties the same status as domestic statutes.41 Yet the ECJ had claimed that EU law was to be regarded as the supreme law of the land requiring Member States’ courts to set aside any national law, even national constitutional law, if it was in conflict. How was the court to respond? Was the ECJ’s claim really plausible that Member States had established a new supreme law of the land by signing and ratifying a set of treaties the core objective of which was to establish a common market? On the other hand, was it plausible to claim that the EU Treaties, which established institutions that had been endowed with significant legislative authority, and played a significant role to secure peace and prosperity in war-ravaged Europe, should be treated like any other treaty? Was it really adequate to apply the general rule applicable to treaties according to which an ordinary statute enacted after the treaty was ratified would trump it? If the court simply accepted the basic ideas underlying Democratic   For a fully developed argument to this effect, see Kumm, n 4.   In the context of the ratification of the Maastricht Treaty, Art 23 of the Basic Law was amended to address questions of European integration. 40   See Art 24 of the Basic Law. 41   This is the dominant interpretation of Art 59 II of the Basic Law. 38 39

Rethinking Constitutional Authority  57 Statism and its idea of constitutional self-government, that is probably the conclusion the court would have reached. If, on the other hand, the court accepted EU law as legitimate constitutional authority on legalist grounds, it would follow the ECJ. But the court chose neither of these options. It embraced an intermediate solution. That intermediate solution illustrates the connection between constitutionalism and the complex set of doctrines that national courts have in fact developed for assessing the ECJ’s claims concerning the supremacy of EU law. First the German Federal Constitutional Court (FCC) accepted without much ado that EU law trumps ordinary statutes, even statutes enacted later in time, because of the importance of securing an effective and uniformly enforced European legal order.42 The principle of ensuring the effective and uniform enforcement of EU law – expanding the rule of law beyond the nation state – was a central reason for the court to recognize the authority of EU law over national statutes. This meant that in Germany EU treaties were effectively granted a more elevated status than ordinary treaties, to which a ‘last in time’ conflict rule generally applies. Yet, contrary to the position of the ECJ, the court recognized that that principle was insufficient to justify the supremacy of EU law over all national law. The principle of legality matters, but it is not all that matters. The second issue before the Court was whether it should subject EU law to national constitutional rights scrutiny. Could a resident in Germany rely on German constitutional rights against EU law? Could the protection of national residents against rights violations guaranteed in the national Constitution be sacrificed on the altar of European integration? Like other questions concerning the relationship between EU law and national law, the German Constitution provided no specific guidance on that question. In Solange I 43 the FCC balanced the need to secure the fundamental rights of residents against the needs of effective and uniform enforcement of EU law and established a flexible approach: for so long as the EU did not provide for a protection of fundamental rights that is the equivalent to the protection provided on the national level, the court would subject EU law to national con­ stitutional scrutiny. At a later point44 the court determined that the ECJ had significantly developed its review of EU legislation and held that the standard applied by the ECJ was essentially equivalent to the protection provided by the FCC’s interpretation of the German Constitution. For so long as that remained the case, the FCC would not exercise its jurisdiction to review EU law on national constitutional grounds. Because the ECJ, through its own jurisprudence, provided the structural guarantees that fundamental rights violations by EU institutions would generally be prevented, it conditionally accepted the authority of EU law. To put it another way: structural deficits in the protection of fundamental rights on the European level were the reason for the FCC to originally insist that it should not 42   Case 1 BvR 248/63, 216/67, EWG-Verordnungen [1967] BVerfGE 22, 293 and Case 2 BvR 255/69 Milchpulver [1971] BVerfGE 31, 145. 43   Case BvL 52/71 Solange I [1974] BVerfGE 37, 271. 44   Case 2 BvR 197/83 Solange II [1986] BVerfGE 73, 339.

58  Mattias Kumm accept the authority of EU law, insofar as constitutional rights claims were in play. When those specific concerns were effectively addressed by the ECJ, the authority of EU law extended also over national constitutional rights guarantees and the FCC as their interpreter. The authority of EU law, then, was in part a function of the substantive and procedural fundamental rights protections available to citizens as a matter of EU law against acts of the European Union. But this is not yet the whole story. There are two residual lines of resistance drawn by national courts to the wholesale acceptance of the authority of EU law. The drawing of these lines is justified by reference to the principle of democracy and the absence of meaningful democratic politics and a meaningful European identity on the European level. In its Maastricht decision45 and later in its Lisbon decision46 the FCC determined that it had jurisdiction to review whether or not legislative acts by the European Union were enacted ultra vires or not. If such legislation were enacted ultra vires, it would not be applicable in Germany. As a matter of EU law it is up to the ECJ, of course, to determine as the ultimate arbiter of EU law whether or not acts of the European Union are within the competencies established by the Treaties.47 But the ECJ had adopted an extremely expansive approach to the interpretation of the EU’s competencies, raising the charge that it allowed for Treaty amendments under the auspices of Treaty interpretation. Under these circumstances the FCC believed it appropriate for it to play a subsidiary role as the enforcer of limitations on EU competencies of last resort. In the decision, arguments from democracy played a central role. Democracy in Europe remains underdeveloped, with electoral politics playing a marginal role. The national domain remained the primary locus of democratic politics. Under those circumstances, ensuring that EU institutions remain within in the competencies established in the Treaties and respecting the basic jurisdictional principle of subsidiarity is of paramount importance. Yet even while insisting that it had the competency to review EU law on jurisdictional grounds, that review would be highly deferential.48 Only if the transgression of the EU competencies is sufficiently serious and the act in question leads to a structurally significant shift in the allocation of competencies would the acts be deemed ultra vires and inapplicable in Germany.49 This deferential standard reflects concerns for the rule of law and effective functioning of a transnationally integrated community, where reasonable interpretative disagreement about the jurisdictional limits of the EU is possible. This points to a final line of resistance. When a national Constitution contains a specific rule containing a concrete national commitment – say a commitment to   Maastricht, n 22.   Lisbon, n 23. 47   See Art 230 ECT. 48   Case 2 BvR 2661/06, judgment of 6 July 2010, Honeywell, http://www.bundesverfassungsgericht. de/en/decisions/rs20100706_2bvr266106en.html. 49  See M Payandeh, ‘Constitutional review of EU Law after Honeywell: Contextualizing the Relationship between the German Constitutional Court and the EU Court of Justice’ (2011) 48 Common Market Law Review 9–38. 45 46

Rethinking Constitutional Authority  59 free secondary education,50 or a restriction to national citizens of the right to vote in municipal elections,51 a categorical prohibition of extradition of citizens to another country52 or otherwise reflects the constitutional identity of the political community53 – these commitments will not generally be set aside by national courts. Instead national courts will insist that the Constitution is amended to ensure compliance with EU law. This line of cases, too, reflects an understanding that the realm of the national remains the primary locus of democratic politics. For so long as that remains the case, a commitment to democracy is interpreted by some Member States’ courts to preclude setting aside national constitutional commitments as they are reflected in these concrete and specific rules. It is then up to the constitutional legislature to initiate the necessary constitutional amendments. This highly stylized and schematic account illustrates the operation of a conception of legitimate constitutional authority that puts the principles of constitutionalism front and centre.54 The principle of legality and its extension beyond the nation state has an important role to play to support the authority of EU law, but concerns relating to democracy and human rights may provide countervailing reasons for limiting the authority of EU law in certain circumstances. Furthermore the republican principles that govern the relationship between national and EU law do not themselves derive their authority from either the national Constitution or EU law. The relative authority of EU and national constitutions is a question to be determined by striking the appropriate balance between competing principles of constitutionalism in a concrete context.

B.  Constitutionalist Elements in Costa and Beyond As was argued above, Costa v ENEL more than any other decision reflects a commitment to Legal Monism, not Cosmopolitan Constitutionalism. But there are elements in Costa that make better sense when interpreted in a constitutionalist, rather than Legal Monist prism (a). Furthermore, EU law has evolved in a way that suggests that the courts’ continued insistence on primacy may today not only be justifiable in constitutionalist terms, but also be generally compatible with the position taken by Member States’ courts (b).

50   Belgian Constitutional Court, Case 12/94, judgment of 3 February 1994, European Schools [1994] BS 6137. 51   Spanish Constitutional Court, Case 1236/92, judgment of 1 July 1992, Maastricht [1992] 177 BOE 5. 52   Polish Constitutional Tribunal, Case P 1/05, judgment of 27 April 2005, European Arrest Warrant [2005] OKT ZU 2005/4A/42. 53   See most recently Case 1 BvR 256/08, 1 BvR 263/08, 1 BvR 568/08, judgment of 2 March 2010, Data Retention, http://www.bundesverfassungsgericht.de/entscheidungen/rs20080311_1bvr025608. html. 54   For a more fully developed account, see Kumm, n 4.

60  Mattias Kumm

1.  Costa and Constitutionalism: Making Sense of the Idea of an Autonomous Legal Order So how does constitutionalism make sense of the claim that the authority of EU law is not simply derived from the fact that Member States have signed and ratified it following their respective constitutional requirements? What, if anything, justified the claim that EU law establishes an autonomous legal order and what follows from it? Whether a treaty qualifies as a Constitution of an autonomous legal order or merely as an ordinary treaty under international law depends on how its claim to legal authority is best understood. If its claim to legal authority is best understood to rest exclusively on the fact that Member States have signed and ratified it, then it is an ordinary treaty of international law. The decisive feature of constitutional treaties establishing an autonomous legal order in some sense is that its claim to authority is in part directly grounded in constitutional principles.55 Its claim to authority is not grounded exclusively in the fact that Member States have signed and ratified it, even if it had not come into existence without Member States’ ratification. The idea of a Constitutional Treaty, then, is contrary to claims by Democratic Statists, not a contradiction in terms. Nor does it matter that the Treaties of the European Union have gradually evolved in a piecemeal fashion, rather than having been created in a legal revolutionary moment, a kind of constitutional ‘big bang’ or a creatio ex nihilo. Whether or not something that takes the form of a treaty is in fact merely a treaty of international law or a form of transnational constitutional law that has greater authority is a question of interpretation. The European Union explicitly claims to be founded on constitutional principles. Article 6 EU Treaty states 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’. The EU, in its self-presentation, is neither founded on the will of a European ‘We the People’, nor is it founded on the ‘will of Member States’. It is founded on the constitutional principles that are a common heritage of the European constitutional tradition as it has emerged in the second half of the twentieth century. And, as the ECJ found more than 40 years ago and is now confirmed in a declaration attached to the Treaty of Lisbon,56 EU law makes a claim to primacy. Whether and to what that claim to authority deserves to be recognized by Member States’ courts is not an easy question and gives rise to the kind of concerns that were described above. But an answer to that question will not make use of unhelpful dichotomies between treaties and constitutions or the ‘will of Member States’ or ‘We the People’.

55   Compare also A von Bogdandy, ‘The Prospect of a European Republic: What European citizens are voting on’ (2005) 42 Common Market Law Review 913. 56   Declaration No 17 concerning primacy [2007] C 306/256.

Rethinking Constitutional Authority  61

2.  Costa Today When the ECJ made the claim that EU law has primacy over all national law, including national constitutional law, it was making a claim that national courts were right not to accept it in an unqualified way. At the time EU law did not provide for adequate constitutional rights protection, it did not provide for an adequate democratic legislative procedure and there was no indication that the Court took seriously the limits of competencies in the Treaty. These concerns became more serious as the decision making moved from unanimity to qualified majority vote in more and more areas since the mid 1980s. To a significant extent the response of Member States’ courts can be understood as a general acceptance of the ECJ’s claim to primacy, but with the proviso that EU law did not violate fundamental rights, remained within its competencies and did not encroach on fundamental constitutional commitments that defined the democratic identity of the Member State. To the extent that Member States’ responses fit this description, they generally comply with constitutionalist requirements. What is remarkable, however, is that all these concerns are now addressed to a large extent, even if not always effectively, by EU law itself. The story about the evolution of the EU’s fundamental rights guarantees is well known and has finally led to the entry into force of the European Charter of Fundamental Rights in December 2009. The concerns relating to competencies has arguably led the ECJ to pay greater attention to delimitation of competencies, even though here there are still good grounds for scepticism.57 Furthermore the Treaty of Lisbon contains interesting procedural innovations involving national Parliaments that might make some contribution to help establish a culture of subsidiarity in Europe. Finally the structural problems relating to democracy have not really been addressed so far by EU actors, even though the legal framework established by the Treaty of Lisbon might allow for the evolution of greater electoral accountability of the Commission in the future thereby making the elections of the European parliament more meaningful.58 But more importantly EU law now specifically requires that Member States’ constitutional identity be respected.59 A plausible interpretation of that provision suggests that it might not violate EU law if a Member State refused to apply EU law in a situation where a fundamental national constitutional commitment is in play.60 If that is correct, it is not implausible that a claim to primacy made by EU law that shares these features may no longer be implausible from a constitutionalist point of view. The justification for the primacy of EU law at the time Costa was decided might not have been plausible. And the limited acceptance 57   See M Kumm, ‘Constitutionalizing Subsidiarity in Integrated Markets’ (2006) 12 European Law Journal 503. 58   See M Kumm, ‘Why Europeans Are Not Constitutional Patriots’ (2008) 6 International Journal of Constitutional Law 117. 59   See Art 4(2) EU Treaty: ‘The Union shall respect the equality of Member States before the Treaties as well as their national identities, inherent in their fundamental structures, political and constitutional . . .’ 60   See M Kumm and V Ferreres Comella, ‘The Primacy Clause of the Constitutional Treaty and the Future of Constitutional Conflict in the European Union’ (2005) 3 International Journal of Constitutional Law 473, 491 and 492.

62  Mattias Kumm by Member States might in most instances have been justified. But to a significant extent EU law has absorbed the concerns that fostered legitimate resistance by Member States, just as Member States had, over time, opened up their legal orders to accept the application of EU over national law in most instances. Shared constitutional principles seem to have provided the focal point of complementary evolutions of both EU law and national constitutional law. The tensions created by a conflict between Legalist Monism and Democratic Statism have to a large extent been replaced by a common commitment to Constitutionalism. That still leaves open the possibility of conflict on application, but it ensures a common framework within which concrete disagreements are addressed.

C.  A Constitutionalist Reading of Kadi But is Kadi compatible with an account that emphasizes the spread of constitutionalism? On the surface the ECJ may seem to have adopted a relatively conventional dualist, statist approach. It insisted on the primacy of EU constitutional principles and explicitly rejected applying these principles deferentially, even though the EU Regulation implemented a UN Security Council Resolution. But on close examination it becomes apparent that important elements of that decision reflect constitutionalist analysis. First, the court specifically acknowledges the function of the UN Security Council as the body with the primary responsibility to make determinations regarding the maintenance of international peace and security.61 Second, the court examines the argument whether it should grant deference to the UN decisions and rejects such an approach only because at the time the complaint was filed there were no meaningful review procedures on the UN level and even those that had been established since then62 still provide no judicial protection.63 Only after an assessment of the UN review procedures does the Court follow that full review is the appropriate standard. This suggests that, echoing the European Court of Human Rights’ approach in Bosphorus,64 more adequate procedures on the UN level might have justified more deferential review. This is further supported by the ECJ’s conclusion that under the circumstances, the plaintiff’s right to be heard and the right to effective judicial review were patently not respected. This language suggests that even under a more deferential form of judicial review the court would have had to come to the same conclusion. This section of the opinion suggests that the Court was fully attuned to constitutionalist sensibilities. It just turns out that the procedures used by the Sanctions

  Kadi, n 1, para 297.  See resolutions of the United Nations Security Council S/RES/1730 of 19 December 2006; S/RES/1735 of 22 December 2006, S/RES/1822 of 30 June 30, 2008. 63   Kadi, n 1, paragraphs 321 and 322. See most recently S/RES/1904 of 17 December 2009, which provides at least certain minimal guarantees. 64   European Court of Human Rights, Case 45036/98, judgment of 30 June 2005, Bosphorus v Ireland [2006] 42 EHRR 1. 61 62

Rethinking Constitutional Authority  63 Committee were so manifestly inappropriate – given what was at stake for the blacklisted individuals – that any jurisdictional considerations in favour of deference were trumped by these procedural deficiencies, thus undermining the case not just for abstaining from judicial review altogether, but also for engaging in a more deferential form of review. Third, the court shows itself attuned to the functional division of labour between the UN Security Council and itself when discussing judicial remedies: the court does not determine that the sanctions must be lifted immediately, but instead permits them to be maintained for three months, allowing the Council to find a way to bring about a review procedure that meets fundamental rights requirements. Finally, during all of this the court is careful to emphasize that nothing it does violates the UN resolution, given that international law leaves it to states to determine by which procedures obligations are enforced. Notwithstanding serious problems that remain, it seems that forceful judicial intervention has had a salutary effect, with serious reform proposals discussed and in part enacted at the UN level.65 An ECJ committed to constitutionalism takes international law seriously. But taking international law seriously does not require unqualified deference to a seriously flawed global security regime. On the contrary, the threat of subjecting these decisions to meaningful review might help bring about reforms on the UN level. Only once these efforts bear more significant fruit will the ECJ have reasons not to insist on meaningful independent review of individual cases in the future. Even if the above suggests that it is a mistake to read Kadi merely as a case of entrepreneurial but jurisprudentially dubious state-building by the ECJ,66 there are still plausible grounds to criticize Kadi on constitutionalist grounds: why did the court not emphasize the universal nature of the human rights it was applying? Why did the court not follow the Attorney General’s lead and was more explicit about the conditional nature of its lack of deference? And was it justifiable for the court to preclude Member States from finding their own ways to address the tensions between compliance with UN sanctions and the relevant human rights concerns? But notwithstanding scope for legitimate criticism, constitutionalist sensibilities were not lacking in Kadi.

V. Conclusions: The Limits and Structure of Constitutional Pluralism Much of the disagreement about the structure of the relationship between UN law, EU law and national constitutional law can be traced back to competing 65   See S/RES/1904, n 62, which now provides at least minimal, even if still inadequate, procedural guarantees. 66   See J Goldsmith and E Posner, ‘Does Europe Believe in International Law?’ The Wall Street Journal, 25 November 2008, A15.

64  Mattias Kumm conceptual frameworks used to construct the world of public law. I have called these conceptual frameworks Legalist Monism, Democratic Statism and Cosmopolitan Constitutionalism respectively. Of course practice is complex and varied. Not everything fits and much of what was presented here was necessarily schematic and in parts underdeveloped. But it makes explicit the structure of Constitutionalism as a distinct cosmopolitan framework for the construction of coherently principled, yet pluralist world of public law. Helping to make explicit how the principles of constitutionalism work allows legal actors to more conscientiously embrace them as a focal point of an overlapping consensus that spans national, European and international law. So what are the core characteristics of a world of public law described in constitutionalist terms? First, unlike the world imagined within the framework of Democratic Statism, the world of public law is imagined as constituted and held together by a shared commitment to constitutional principles. There is no fundamental distinction between state law and international law. State law and law beyond the state have more in common than statists suggest. Constitutional authority and constitutional principles are constitutive not only of national law and politics, but of law and politics tout court. In that sense constitutionalism is reconceived in a cosmopolitan and not statist framework. Second, nor is the legal world imagined as a monist whole structured by clear hierarchies. Instead constitutionalism helps give an account of the deeply pluralist and fragmented nature of the world of public law. There are two ways in which constitutionalism and pluralism are connected. First, constitutionalism explains how there can be a plurality of legal regimes that make claims to authority which go beyond their origin with the consent of states. These regimes may be based on a treaty, but these treaties are, just like domestic constitutions in traditional constitutional theory, genuinely constitutive: with them a new legal authority comes into the world. Instead of deriving their authority from the legal acts that made them possible, their claims to authority derive at least in part directly from the constitutional principles they embody and help realize. Second, constitutional principles provide the mediating principles for a deeply pluralist structure of public law. In practice regime pluralism sometimes leads to the enactment of rules that conflict with one another. When they do there is no guarantee that conflicts between them will be resolved in the same way by each regime. So there is a distinct possibility of contradictory claims that are part of the legal world without law having the resource to resolve them conclusively. But notwithstanding the possibility of irresolvable legal conflict, this kind of pluralism is not deep and hard, but shallow and soft. It is shallow and not deep because constitutionalist principles are the shared foundation of public law practices. And it is not hard but soft, because constitutionalist principles serve as a common framework to mediate potential disputes and give rise to principled practices of engagement and deference that reduce the occasions and limit the stakes of conflict. The world of public law, constructed through the prism of the best understanding of European legal practices, is both constitu-

Rethinking Constitutional Authority  65 tionalist and pluralist. To put it another way: it is the world of constitutional pluralism.67 But constitutional pluralism is no panacea and is not always attractive. First, any pluralism described in constitutionalist terms presupposes a shared commitment to constitutional principles. Without such a shared background, neither the foundations of claims for legitimate authority, nor their mediation could be understood as requiring engagement with and interpretation of shared constitutional principles. Second, constitutional pluralism is not inherently superior to hierarchical constitutionalism. Whether it is or not itself depends on how potentially competing constitutional principles play out in particular contexts. If the account provided above is plausible, Cosmopolitan Constitutionalism is not only descriptively the most persuasive account of the world of public law as it relates to issues of pluralism. It also provides normative criteria that help assess whether a pluralist rather than hierarchical conception of the relationship between different legal practices is justified and, to the extent a pluralist conception is attractive, what its internal structure should be. The refusal of a legal order to recognize itself as hierarchically integrated into a more comprehensive legal order is justified, if that more comprehensive order suffers from structural legitimacy deficits that the less comprehensive legal order does not suffer from. The concrete norms governing the management of the interface between legal orders are justified if they are designed to ensure that the legitimacy conditions for liberal-democratic governance are secured. In practice that means that there are functional considerations that generally establish a presumption in favour of applying the law of the more extensive legal order over the law of the more parochial one, unless there are countervailing concerns of sufficient weight that suggest otherwise. It is unlikely that the legal world will develop in such a way that the UN Charter as interpreted by the UN organs would deserve the unqualified respect as the supreme law of the globe. From the make-up of the UN Security Council to the procedures used for decision making and the lack of meaningful judicial review, structural deficiencies are too great and too deeply institutionalized, even if piecemeal reforms will always be discussed and might one day be implemented. The relationship between EU law and the law of Member States, too, will probably remain pluralist at least in the weak sense that certain national constitutional commitments will be recognized by national courts as a reason not to apply EU law in case of conflict. Even presuming that all it took to address the democratic deficit was to ensure the genuine electoral accountability of the Commission, there is no discernible will in Europe currently to help bring about such a shift. For so long as that remains the case, constitutional pluralism remains preferable to hierarchical integration.

67   The constitutionalist account of constitutional pluralism is only one among others. For further references and basic positions see the debate between M Poiares Maduro, N Walker, J Baquero Cruz and M Kumm in Avbelj and Komárek, n 5.

4 Three Claims of Constitutional Pluralism MIGUEL POIARES MADURO* It has become consensual to recognize that the European Union is governed by a form of constitutional law.1 But, to a large extent, the Constitution of Europe has developed without a constitutional theory.2 There has been an important doctri­ nal effort in the dogmatics of European constitutional law, highlighting how con­ stitutional concepts have been integrated into the law of the European Union and dominate its organization of power. At the same time, however, the constitutional theorization of the European Union has been limited, to a large extent, to a dis­ cussion on whether or not it deserves the constitutional label. More recently there has been an increased discussion on the nature of European constitutionalism.3 But even this discussion appears more often as an upgrade on two old questions: Does Europe have a Constitution? Does Europe need a Constitution? Rarely do we see any work on the constitutional consequences of different constitutional conceptions of the European Union: what theories of fundamental rights, separ­ ation of powers or judicial review ought to dominate EU constitutional law? If

*  I would like to thank the participants at the Oxford workshop that constitutes the basis for this book for comments on an earlier version. Particular thanks go to the editors of the book, Jan Komárek and Matej Avbelj. Thanks are also due to Giuseppe Martinico for comments and suggestions and Boris Rigod for his research assistance. 1   Even sceptics of the constitutional authority of EU law such as Udo de Fabio (the well-known judge of the German Constitutional Court) recognize the constitutional character of EU law. Cited in F Mayer and M Wendel, ‘Multilevel Constitutionalism and Constitutional Pluralism: Querelle Allemande or Querelle d’Allemand?’ Chapter 6 in this volume. 2   M Poiares Maduro, ‘Europe and the Constitution: What If This Is As Good As It Gets?’ in M Wind and J Weiler (eds), European Constitutionalism Beyond the State (Cambridge, Cambridge University Press, 2003) at 74, 76; M Poiares Maduro, We the Court: The European Court of Justice and the European Economic Constitution (Oxford, Hart Publishing, 1998) 7 ff. 3   JHH Weiler, The Constitution of Europe. “Do the New Clothes Have An Emperor?” and Other Essays on European Integration (Cambridge, Cambridge University Press, 1999); N Walker, ‘EU Constitutionalism and New Governance’ in G de Búrca and J Scott (eds), Constitutionalism and New Governance in Europe and the United States (Oxford, Hart Publishing, 2006); N Walker, ‘Reframing EU Constitutionalism’ in J Dunoff and J Trachtman (eds), Ruling the World? Constitutionalism, International Law and Global Governance (Cambridge, Cambridge University Press, 2009); I Pernice, ‘Multilevel Constitutionalism and the Treaty of Amsterdam: European Constitution-Making Revisited’ (1999) 36 Common Market Law Review 703; I Pernice, ‘Multilevel Constitutionalism in the European Union’ (2002) European Law Review 511; A von Bogdandy and J Bast (eds), Principles of European Constitutional Law (Oxford, Hart Publishing, 2010); K Tuori and S Sankari (eds), The Many Constitutions of Europe (Farnham, Ashgate, 2010); M Poiares Maduro (1998) and (2003), n 2.

68  Miguel Poiares Maduro European constitutionalism changes constitutionalism itself, how does that impact on those aspects of a theory of constitutional law? Constitutional pluralism4 has been, perhaps, the most successful attempt at theorizing the nature of European constitutionalism. However, it has not yet pro­ vided a constitutional theory of EU law. Some understand it, in fact, simply as theory regulating conflicts of constitutional authority. In other words, constitu­ tional pluralism would not define the identity of European constitutionalism itself but the nature of its relationship with other constitutional orders (national and, possibly, international). In this piece, I want to discuss the real potential of constitutional pluralism as a constitutional theory. I conceive constitutional pluralism not only as remedy for constitutional conflicts of authority but as the theory that can best embrace and regulate the nature of the European Union polity. I will put forward three different claims that can be made by constitutional pluralism. I will start by highlighting how those claims entail a more ambitious conception of constitutional pluralism when compared with the more modest claim usually attributed to constitutional pluralism. As a consequence, I will dis­ cuss the relationship between constitutional pluralism and EU constitutional law and constitutionalism in general. The broadest claim will present constitutional pluralism as the basis for an upgrading of the theory of constitutionalism in general but will also highlight the challenges still to be faced by constitutional pluralism to perform that role.

I.  The Claims of Constitutional Pluralism It has been stated that constitutional pluralism has emerged as a response to the Maastricht judgment of the German Constitutional Court.5 This judgment brought to the fore the risks of constitutional conflicts between EU law and national Constitutions emerging from the claims of final authority embodied in the case law of the European Court of Justice and national constitutional courts. Constitutional pluralism is often presented as a reaction to this judgment, attempting simultaneously to describe that reality and accommodate those competing constitutional claims. 4   Including its German variant of multi-level constitutionalism. See Pernice (1999) and (2002), n 3 and Pernice, ‘The Treaty of Lisbon. Multilevel Constitutionalism in Action’ (2009) 15 Columbia Journal of European Law 349–407; F Scharpf, ‘Legitimacy in the Multi-level European Polity’ in P Dobner and M Loughlin (eds), The Twilight of Constitutionalism? (Oxford, Oxford University Press, 2010); N Walker, ‘Multilevel Constitutionalism: Looking Beyond the German Debate’ in Tuori and Sankari, n 3. 5  J Baquero Cruz, ‘The Legacy of the Maastricht-Urteil and the Pluralist Movement’ (2008) European Law Journal 389–422 at 412 ff, Mayer and Wendel, n 1 and A Somek, ‘Monism: A Tale of the Undead’, Chapter 15 in this volume.

Three Claims of Constitutional Pluralism  69 But it is also possible to construct a narrative that sees the driving force behind the constitutional pluralism movement in the seminal article by Neil MacCormick ‘Beyond the Sovereign State’6 (which preceded the judgment). The latter was certainly more important in my own path towards constitutional pluralism. My initial interest in the ideas that led to constitutional pluralism was not so much determined by the risks of constitutional conflicts but more by an inquiry into the character of the European Constitution itself.7 In this respect, the analysis of European constitutionalism undertaken in the works of Joseph Weiler also already hinted at other aspects of constitutional pluralism, particularly when he later articulated the principle of constitutional tolerance.8 In this second narra­ tive, constitutional pluralism emerges as a theory of European constitutionalism and not simply as a theory of constitutional conflicts. The former focuses on the legitimacy of European constitutionalism and its model of organizing power while the latter focuses in its relationship with other constitutional orders. It must be recognized, however, that it has been the risks of constitutional con­ flicts highlighted by the Maastricht judgment that have fed the interest in consti­ tutional pluralism. This shaped its agenda. Most works on constitutional pluralism have focused on courts and the risks of constitutional conflicts of authority embedded in their respective case laws. But one does not need to limit constitu­ tional pluralism to this. In my view, constitutional pluralism is both better under­ stood and more useful if not limited to that. Furthermore any debate on how to solve or regulate constitutional conflicts of authority inherently involves a debate on the nature and legitimacy of the competing constitutional claims of final authority. As such, it always requires a broader understanding of the nature of the European and national Constitutions and their relationship with constitutional­ ism in general. It is with this in mind that I will present three different claims of constitutional pluralism in order to clarify what constitutional pluralism is and the extent to which it can help us develop a constitutional theory of EU law and of constitu­ tionalism in general.

II.  The Empirical Claim The starting point of constitutional pluralism is empirical. Constitutional plural­ ism identifies the phenomenon of a plurality of constitutional sources and claims of final authority which create a context for potential constitutional conflicts that   N MacCormick, ‘Beyond the Sovereign State’ (1993) 56 Modern Law Review 1.   This is visible in Poiares Maduro (1998) and (2003), n 2.   See JHH Weiler, ‘In Defence of the Status Quo: Europe’s Constitutional Sonderweg’, in Weiler and Wind, n 2 and Weiler, ‘Why Should Europe Be a Democracy: The Corruption of Political Culture and the Principle of Toleration’ in F Snyder (ed), The Europeanisation of Law: The Legal Effects of European Integration (Oxford, Hart Publishing, 2000) 213, 217. 6 7 8

70  Miguel Poiares Maduro are not hierarchically regulated. More broadly, it refers to the expansion of rele­ vant legal sources, the multiplication of competing legal sites and jurisdictional orders, and the existence of competing claims of final authority. In EU law, where the current movement started, constitutional pluralism also mapped what is usually described as a discursive practice between the European Court of Justice and national constitutional courts, aimed at reducing the risks of constitutional conflicts and accommodating their respective claims of final authority. If I had to summarize the core empirical claim of constitutional pluralism, it would be the following: constitutional pluralism is what best describes the current legal reality of competing constitutional claims of final authority among different legal orders (belonging to the same legal system) and the judicial attempts at accommodating them. This leaves open the question of whether it is more appro­ priate to conceive of constitutional pluralism in the EU as a pluralism of legal orders (EU and national) or as a pluralism of constitutional claims of authority within the same legal order (with national legal orders being part of the broader European legal order in its respective field of application). In my previous work I have conceived of a European legal order composed of national and EU legal orders. However, the best way to present the current legal reality in the practice of courts may be by making use of Tuori’s distinction between legal order and legal system.9 While the legal order refers to law as a symbolic-normative phenome­ non, the legal system refers to the legal practices where the legal order is produced and reproduced (lawmaking, adjudication and legal scholarship).10 Making use of this distinction, we can conceive of the EU and national legal orders as auto­ nomous but part of the same European legal system. For those practising law in Europe, this European legal system implies a commitment to both legal orders and imposes an obligation to accommodate and integrate their respective claims. The importance of this resides, among other things, on the hermeneutic require­ ments imposed on national and European courts when acting within the EU legal system.11 The empirical thesis assumes that both the European Court of Justice and national constitutional courts are aware of their competing constitutional claims and act accordingly, by accommodating their respective claims so as to minimize the risks of constitutional conflicts.12 The most well known example of this regards the fundamental rights jurisprudence of the national constitutional courts and the European Court of Justice.13   See K Tuori, ‘The Many Constitutions of Europe’, in Tuori and Sankari, n 3.   K Tuori, Towards a Theory of Transnational Law, manuscript, not yet published. 11   In this respect, see the principles of contrapunctual law that I develop in ‘Contrapunctual Law: Europe’s Constitutional Pluralism in Action’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart Publishing, 2003). 12   Note, however, that this does not imply that courts are aware of constitutional pluralism and see themselves as practising it. Constitutional pluralism only requires for courts to act so as to accommo­ date their conflicting constitutional claims. It does not require them to actually acknowledge those competing constitutional claims. 13   For a classical account B de Witte, ‘The Past and Future Role of the European Court of Justice in the Protection of Human Rights’ in P Alston (ed), The EU and Human Rights (Oxford, Oxford 9

10

Three Claims of Constitutional Pluralism  71 This empirical claim is challenged by some. Alexander Somek makes the most developed challenge to the empirical thesis of constitutional pluralism to date. He claims that pluralism does not fit the existing legal practice.14, 15 For him, at best, there is nothing really new in constitutional pluralism. Constitutional pluralism would be a form of monism under national law. In this light, the question of final authority is not open. There is a legal answer. Instead, he argues that EU law itself does not prevent national judges from adopting decisions disrespecting EU law (what he calls ‘false decisions’) since ulti­ mately their decisions will not be void and the only consequence may be tort lia­ bility. In his own words: ‘It makes sense to say that Member States retain the power to have their judges adopt false decisions, at any rate, as long as States are willing to pay for it.’16 To this, he adds the lack of effective EU powers to protect EU law against recalcitrant Member States.17 The conclusion is that ‘the overarch­ ing legal system vests the power to adjudicate supremacy conflicts in the national system’.18 What constitutional pluralism presents as conflicting claims of final authority is reconstructed by Somek as an actual recognition of the supremacy of national constitutionalism. I disagree with his empirical assessment. First, the existence of false decisions is, as Somek himself recognizes, a usual feature of any legal sys­ tem.19 Unfortunately, it is not uncommon for courts to wrongly interpret and apply the law. Under EU law, as under national law, those ‘false decisions’ are not void, because the national courts that take them have been empowered to inter­ pret and apply EU rules by EU law itself. It is because national courts are part of the judicial system set forth by EU law that their decisions (even when ‘false’ in the sense mentioned by Somek) are valid and become final when all appeals have been exhausted. This is, therefore, a consequence of EU law and not national law. Second, the argument that the existence of liability for judicial acts violating EU law amounts to permit States to violate EU law so long as they are willing to pay for it is equally unconvincing. A similar argument can be made in respect of national law itself. Consider, for example, national laws imposing the liability of University Press, 1999); A von Bogdandy, ‘The European Union as a Human Rights Organization? Human Rights and the Core of the European Union’ (2000) 37 Common Market Law Review 1307; J Lockhart and JHH Weiler, ‘“Taking Rights Seriously” Seriously: The European Court and its Fundamental Rights Jurisprudence’ (1995) 32 Common Market Law Review 51 and 579. For a reassess­ ment of this account see BO Bryde, ‘The ECJ’s Fundamental Rights Jurisprudence – A Milestone in Transnational Constitutionalism’ in M Poiares Maduro and L Azoulai (eds), The Past and Future of EU Law (Oxford, Hart Publishing, 2010). 14   Somek, n 5. 15   The following section is based on an earlier version of ch 15 by Alexander Somek. As a result, not all references fully coincide with ch 15 in its final form. Since the substance of the ideas remains the same as the earlier version, the decision was made to publish this chapter with most of the original references intact. 16   Somek, n 5 (original version). 17   Ibid. 18   Ibid. 19   Ibid.

72  Miguel Poiares Maduro the state (including, in some cases, by virtue of judicial acts) for violations of the law, in particular the Constitution. Do we put into question the supremacy of the law and the Constitution in the national legal order whenever there is liability for state acts violating it? Liability is an additional instrument of enforcement and supremacy, not a price to be paid in order to be exempted from enforcement and supremacy. For Somek, the decisive element appears to be the lack of an effective enforce­ ment instrument against recalcitrant Member States. But he measures the enforce­ ment power of EU law in light of an ‘ideal’ that does not correspond to the reality existing in many States. Usually courts do not have any material force to impose on the political process compliance with their decisions.20 That does not lead us to put into question their legal authority. Even if it can be stated that the European Union lacks the army at the disposal of federal governments, it has other alterna­ tive and perhaps more effective instruments in the day-to-day effectiveness of the law. It now has the power to impose fines on States not complying with EU law. This does not exist, for example, in the United States. If there may be greater com­ pliance with the supremacy clause in the United States21 I don’t think it is due to the enforcement mechanisms at the disposal of the federal government.22 The core of Somek’s criticism must and does lie somewhere else. First, as it is obvious from the above citation, he does not consider it possible for national courts to accommodate the claims of EU law without severing their ties with the national legal order. But, in fact, national courts are now tied to both the national and EU legal orders. They are part of a European legal system composed of both the EU and national legal orders.23 What we see in the practice of courts is an attempt to accom­ modate the claims of those legal orders without severing ties with neither of them. When conflicting claims may exist they make use both of principles of EU law, such as supremacy and direct effect, and of principles developed under national law, such as the so lang doctrine of the German Constitutional Court or the counterlimits of the Italian Constitutional Court, to reconcile those claims. More importantly, however, Somek argues that constitutional pluralists give up precisely where an answer is most needed: what happens when the constitutional conflict cannot be prevented or solved?24 There are different possible replies to this

20   There are even examples, at national level, of constitutional decisions that have not been com­ plied with by the political process or only complied after extensive delays. Italian Corte Costituzionale, Decisions No 826/1988 of 13 July 1988 and No 420/1994 of 5 December 1994 and No 466/2002 of 20 November 2002 (all on pluralism of the media). To some extent also German Constitutional Court, Case 1 BvR 1087/91, [1995] BVerfGE 93, 1 (Crucifixes in Bavarian Classrooms). 21   Which has not always been the case. See R Schütze, From Dual to Cooperative Federalism: The Changing Structure of European Law (Oxford, Oxford University Press, 2009). 22   See below the discussion on the degrees of constitutional pluralism. 23   See above. 24   Somek, n 4. Nico Krisch, a pluralist himself, puts forward a different, but related, open question. For him, the parallelism often made between pluralism and checks and balances ‘leaves open a crucial question: who should be entitled to check whom and why?’. See Beyond Constitutionalism: The Pluralist Structure of Postnational Constitutional Law (Oxford, Oxford University Press, 2010) 89.

Three Claims of Constitutional Pluralism  73 critique. They depend on different normative standpoints on constitutional plural­ ism to be discussed below. At this point, it is enough to state that one of the pur­ poses of constitutional pluralism is precisely to legitimate leaving that question open and that, at an empirical level, the fact that the question remains open is a simple description of the constitutional status quo in Europe and only serves to reinforce the value of constitutional pluralism since its articulation of techniques of accommodation of the competing constitutional claims helps decreasing the num­ ber of times that the question will pose itself. Another usual empirical challenge departs from the example of the amend­ ments to national Constitutions whenever a new Treaty of the EU (or any EU act for that purpose25) collides with a national Constitution. Those amendments of national Constitutions are presented as evidence of national constitutional supremacy that would, otherwise, prevent the entry into force of the new Treaty.26 This would be particularly the case when such collision has been ascertained by a national constitutional court and the latter imposes amending the Constitution in order to eliminate such incompatibility.27 In national systems that allow ex-ante constitutional review of a Treaty subject to national ratification (including EU Treaty amendments), this requirement of constitutional conformity before the Treaty can be ratified would be evidence that the final authority ultimately rests with national Constitutions. However, such constitutional control takes place as part of the ratification process of the Treaty and, as such, is better conceived as part of the Treaty amendment process itself. In other words, it is more appropri­ ate to talk about such national constitutional ex-ante control as part of the EU law or Constitution-making process and not as a limit to the supremacy of an EU law that is not yet binding. Moreover, one can question if constitutional amendments generated by EU Treaty amendments are, in fact, evidence of EU law subordina­ tion to national Constitutions. Instead, can they not be presented as a require­ ment imposed by EU law on national Constitutions?28 In fact, whenever a possible conflict was detected between a national Constitution and an EU Treaty it was the national Constitution that was amended and not the EU Treaty.29 The same hap­ pened when the Polish Constitutional Court declared, in ex-post review, that the EU arrest warrant was contrary to its national Constitution. The Polish Court did 25   Polish Constitutional Tribunal, judgment of 27 April 2005, P 1/05, English translation www. trybunal.gov.pl/eng/summaries/summaries_assets/documents/P_1_05_full_GB.pdf. 26   French Constitutional Council, Decision of 9 April 1992 No. 92-308 DC (Maastricht I); German Constitutional Court, Case 2 BvE 2/08, judgment of 30 June 2009, Lisbon, English translation http:// www.bverfg.de/entscheidungen/es20090630_2bve000208en.html, paragraph 113; Spanish Constitutional Court, Spanish Constitutional Court, Case 1236/92, judgment of 1 July 1992, Maastricht [1992] 177 BOE 5. 27   Polish Constitutional Court, n 25. 28   In this sense, FL Pires, ‘Competência das Competências: Competente mas sem competências?’ (1998) Revista de Legislação e Jurisprudência No 3885, 356. 29  Spanish Constitutional Court, n 26; French Constitutional Council, n 26 and judgment of 2 September 1992, Case No 92-312 DC (Maastricht II); Portuguese Constitutional Court, Decision of 1 February 1989, Case No 184/89 (ERDF); Cyprus Supreme Court, Civil Appeal No 294/2005, judg­ ment of 7 November 2005, European Arrest Warrant. See also references in J Baquero Cruz, n 5, 397 ff.

74  Miguel Poiares Maduro not declare the EU act inapplicable. Instead, it preserved its application while granting the Polish political process some time to amend the Constitution in order to eliminate such incompatibility.30 It is for these reasons that some challenge the empirical claim of constitutional pluralism from the opposite direction: the supremacy of EU law would now be the rule and an established fact and that would be so even with respect to national Constitutions.31 This would amount to a monism of EU law, with national courts having changed their primary allegiance towards EU law. The latter would pro­ vide their rules of recognition or the grundnorm. However, both the statements of national courts (including the usual reconstruction of EU law supremacy as derived from national constitutional law) and their practice defy this narrative.32 Moreover, if it is true that the EU claim of supremacy forces national courts to reconstruct even their national constitutional law, it is equally true that national constitutional claims also shape how EU law is developed and sensitive to national constitutional traditions. The current reality is better understood as one where EU and national legal orders can be construed as normatively autonomous but also institutionally bonded by the adherence of their respective actors to both legal orders. The latter bond is institutionally operated but founded on a normative commitment to European constitutionalism that has important consequences. In particular, it requires a coherent and integrated construction of the European legal system by all those different actors. Empirically, the open question remains open. The examples of a discursive prac­ tice among courts acknowledging this situation abound.33 This does not involve 30  Polish Constitutional Tribunal, n 25. See J Komárek, ‘European Constitutionalism and the European Arrest Warrant: In Search of Limits of “Contrapunctual Principles”’ (2007) 44 Common Market Law Review 9, 19; and also Z Kühn, ‘The European Arrest Warrant, Third Pillar Law and National Constitutional Resistance/Acceptance: The EAW Saga as narrated by the Constitutional Judiciary in Poland, Germany and the Czech Republic’ (2007) 3 Croatian Yearbook of European Law & Policy 99. 31   See R Barents, ‘The Fallacy of European Multilevel Constitutionalism’, Chapter 7 in this volume. 32   For a general overview and more detailed presentation of this argument see Poiares Maduro, n 11. See also D Chalmers, ‘Judicial Preferences and the Community Legal Order’ (1997) 60 Modern Law Review 164 and ‘The Autonomy of EU Law and the supply of Legal Goods’ (on file with the author); M Kumm, ‘The Jurisprudence of Constitutional Conflict: Constitutional Supremacy in Europe before and after the Constitutional Treaty’ (2005) 11 European Law Journal 262 and the contributions in AM Slaughter, A Stone Sweet and JHH Weiler (eds), The European Courts and National Courts: Doctrine and Jurisprudence (Oxford, Hart Publishing, 1998). For recent decisions of national constitu­ tional courts which only accept a conditional form of supremacy see: the German Constitutional Court, n 26 (Lisbon) and Case 2 BvR 2661/06, judgment of 6 July 2010, Honeywell, English translation http://www.bundesverfassungsgericht.de/en/decisions/rs20100706_2bvr266106en.html; the Czech Constitutional Court, judgment of 28 November 2008, Case Pl. US. 19/08, Lisbon I, and judgment of 3 November 2009, Case Pl. US. 29/09 Lisbon II; French Conseil d’Etat, Decision of 7 February 2007, Case No 287110, Arcelor. Approaches of national courts to such conditionality vary, however. For a case study of patterns of national judicial decisions openly disregarding EU law see F Pereira Coutinho, ‘Os Tribunais Nacionais na Ordem Jurídica Comunitária: O Caso Português’ PhD thesis, Universidade Nova de Lisboa (2004). 33   See cases cited in the previous note and in Poiares Maduro, n 11. See also several examples in D Sarmiento, ‘The Silent Lamb and the Deaf Wolves: Constitutional Pluralism, Preliminary References

Three Claims of Constitutional Pluralism  75 courts using the language of constitutional pluralism. Constitutional pluralism does not require courts to talk about constitutional pluralism in their decisions. It does not even require for courts to engage expressly with other courts. Those that say that courts do not endorse constitutional pluralism, because they neither talk about con­ stitutional pluralism nor cite decisions of other courts, miss the point. The fact that courts continue to narrate the law according to the internal viewpoint of their legal order does not mean that such viewpoint has not been altered by reason of constitu­ tional pluralism. The primary example is how many national courts have inter­ preted their Constitutions so as to incorporate the demands arising from the supremacy claim of EU law without formally accepting, in most cases, such suprem­ acy. The narrator is still the national Constitution but the script has changed. What constitutional pluralism claims, in this respect, is that judicial actors have changed the internal perspective of their legal order in order to accommodate the claims of the other legal order. As such, the new internal perspective is informed by constitu­ tional pluralism. Courts are not institutionally blind. Courts are aware of the con­ text of constitutional pluralism and react to it. A different issue is if they react in the best way. The latter is a normative question, not an empirical one.

III.  The Normative Claim While the empirical thesis of constitutional pluralism limits itself to state that the question of final authority remains open, the normative claim is that the question of final authority ought to be left open. Heterarchy34 is superior to hierarchy as a normative ideal in circumstances of competing constitutional claims of final authority. This normative thesis implies, in practice, another: that those compet­ ing constitutional claims are of equal legitimacy or, at least, cannot be balanced against each other in general terms. Therein lays another of the usual challenges to constitutional pluralism: that it recognizes a constitutional order where there is none. Such an unjustified extension of constitutionalism might even end up undermining constitutionalism itself. I will address this particular risk below when discussing the relationship between constitutionalism and constitutional pluralism. For now, I will limit myself to the other issues. As stated, constitutional pluralism recognizes that there is a constitutional claim on the part of the European Union. As a normative thesis it assumes, furthermore, that such claim is legitimate. That the claim exists is now largely

and the Role of Silent Judgments in EU Law’, Chapter 13 in this volume and J Komárek, ‘Institutional Dimension of Constitutional Pluralism’, Chapter 10 in this volume. Some, however, deny that this discursive practice actually exists. See, for example, Somek, n 5, (original version). 34   To use the expression of D Halberstam, ‘Constitutional Heterarchy: The Centrality of Conflict in the European Union and the United States’, in Dunoff and Trachtman, n 3.

76  Miguel Poiares Maduro undisputed.35 But is it legitimate? Questioning this legitimacy would solve the risk of constitutional conflicts by recognizing the supremacy of national constitution­ alism (considered the only or higher legitimate form of constitutionalism). Constitutional pluralism implies, therefore, recognizing the legitimacy of the EU constitutional claim. This is the real starting point of constitutional pluralism in the European Union.36 Constitutionalism is both possible and necessary in the European Union. But its constitutional claim of authority is not a consequence of the powers the Union has acquired. I am stating this because rather often the constitutional argument for the European Union is based on the need to have constitutionalism as a form of governing and controlling EU power. Wherever there is power there ought to be constitutional limits. As a consequence: since the Union has developed autono­ mous forms of power it ought to be subject to constitutionalism. This might be true but it is not enough to support a constitutional claim to be opposed to national constitutionalism. It would make such claim a product of a circular reasoning that would require the power claimed by the Union to be subject to constitutional con­ trol and then use the latter to justify the power being claimed. A deeper justifica­ tion and legitimacy of European constitutionalism must be derived from its constitutional added value with respect to national constitutionalism. It is this that authorizes the former to be opposed to the latter and explains constitutional plu­ ralism. There is an important difference in recognizing that the European Union needs constitutional instruments to limit and govern its power and arguing that the Union has a constitutional claim of authority that can be opposed to that of national Constitutions. The latter requires demonstrating that the constitutional claim of the European Union has a legitimacy that can be opposed to that of the States. This does not imply that its constitutionalism is the same as that of the States or even has the same comprehensive social ambition. But it does require for EU constitutionalism to have, in certain respects, a constitutional added value with respect to State constitutionalism that may be opposed to it. In other works I have articulated where that added constitutional value may be.37 In short, we can identify three main sources of constitutional and democratic added value that the process of European integration and EU law can bring with respect to national political communities and their constitutional democracies. First, European constitutionalism promotes inclusiveness in national democra­ 35   I describe how this claim emerged in Poiares Maduro, n 11. For other narratives see JHH Weiler and U Haltern, ‘Constitutional or International? The Foundations of the Community Legal Order and the Question of Judicial Kompetenz-Kompetenz’, in Slaughter, Stone Sweet and Weiler, n 32, A Stone Sweet, Governing with Judges: Constitutional Politics in Europe (Oxford, Oxford University Press, 2000) 166 ff. 36   As explained by Mayer and Wendel behind the emergence of the similar concept of multilevel constitutionalism was precisely the attempt to argue that the concept of constitutionalism should not be limited to the state and could be applied to the European Union. See Mayer and Wendel, n 1. 37   Poiares Maduro (1998) and (2003), n 2 and Poiares Maduro, ‘Passion and Reason in European Integration’, Forum Constitutionis Europae No 03/10, http://ssrn.com/abstract=1709950.

Three Claims of Constitutional Pluralism  77 cies by requiring national political processes to take into account out-of-state interests that may be affected by the deliberations of those political processes. By committing to European integration, EU states accept to mutually open their democracies to the citizens and interests of other Member States. This amounts to an extension of the logic of inclusion inherent in constitutionalism. Second, European constitutionalism allows national democracies to collectively regain control over transnational processes that evade their individual control. While in the former case we can talk of outbounded democratic externalities (states impacting on out-of-state interests), the latter refers to inbounded democratic externalities (out-of-state decisions and processes that affect domestic interests). Third, European constitutionalism can also constitute a form of self-imposed external constitutional discipline on national democracies. There are many instances where domestic political malfunctions can be better corrected by exter­ nal constraints. These may force national political processes to rationalize national policies that have, for example, become path-dependent or captured by certain interests. In many such instances, EU law’s imposed discipline rationalizes and, often, reignites a more informed and genuinely open deliberation in the national political process. This does not ignore that national political communities are still, in many respects, the best forum for pursuing the values of constitutionalism. National constitutionalism is still, in many instances, the best proxy for constitutional val­ ues and also serves as a guarantee against possible concentrations and abuses of power from European constitutionalism. The highlighted forms of constitutional added value of European constitutionalism do not provide it with a general claim of supremacy over national constitutionalism. However, they do provide it with a claim. It is the constitutional added value arising from the mutual correction of each other’s constitutional shortcomings that requires pluralism to be maintained between the national and European constitutional orders. As long as the possible conflicts of authority do not lead to a disintegration of the European legal order, the pluralist character of European constitutionalism should be met as a welcome discovery and not as a problem in need of a solution. It is this that also explains why European constitutionalism brings us closer to the ideals of constitutionalism. It is not, in itself, a closer representation of consti­ tutionalism than national constitutionalism but their interplay is. This is what constitutional pluralism argues and therein lays its thicker normative claim, one that relates constitutional pluralism and constitutionalism in general.

IV.  The Thick Normative Claim The thicker normative claim of constitutional pluralism is that, in the current state of affairs, it provides a closer approximation to the ideals of constitutionalism than either national constitutionalism or a form of EU constitutionalism modelled after

78  Miguel Poiares Maduro state constitutionalism.38 In this way, the pragmatic concern that has dominated earlier writings on constitutional pluralism is turned upside down. Constitutional pluralism is not simply a remedy for the risks of constitutional conflicts of author­ ity; it’s the best representation of the ideals of constitutionalism for the current con­ text of increased pluralism and deterritorialization of power.39 To understand this we need to start by recognizing that pluralism is inherent in constitutionalism.40 In fact, constitutionalism aims to simultaneously guarantee and regulate such pluralism: a pluralism of interests, ideas and visions of the com­ mon good that is reflected in the paradoxes of constitutionalism and the balance between democratic deliberation and constitutional rights that its modern liberal form embraced. In a previous work I have argued that constitutionalism is related to three para­ doxes: the paradox of the polity; the fear of the few and the fear of the many;41 and the question of who decides who decides. They are paradoxical because they simultaneously embrace conflicting values in an attempt to reconcile them that is at the core of constitutionalism. With respect to all of them, national constitu­ tionalism could be seen as both a promoter of and a limit to constitutionalism. The polity is the basic assumption of a Constitution. Constitutional questions have always been addressed within a pre-existing polity (normally the Nation State). It is that polity that has served as the yardstick of constitutionalism. Relations within the polity are regulated by constitutional law. Relations among polities, instead, have been dominated by a different set of actors (the states) and a different set of rules (international law). The Constitution both defines and pre­ supposes a polity or political community whose members are bound by such Constitution. It is from this political community and its people that the demo­ cratic process draws its legitimacy and that of the majority decisions reached in the democratic representative process. The basis of the polity is normally referred to as ‘the people’. Constitutional and democratic theory scholars normally presuppose that ‘a people’ already exists.42 But what makes a people? And who has the right to be considered as part of the people? And why should participation and representa­ tion be limited by the requirement of belongingness to such a polity? It is the para­ dox of the concept of polity in its relation with constitutionalism and democracy. Isn’t a national demos a limit to democracy and constitutionalism? In fact, as dis­ cussed above, participation in national democracies is not granted to all those affected by the decisions of the national political process but only to those affected 38   It is unclear the extent to which most constitutional pluralists will support the thick normative claim. I am one who does. Mattias Kumm and Daniel Halberstam will likely share this view. Others, such as Neil Walker, appear to keep a more external and agnostic view on this question. 39   This notion is developed by Gustavo Zagrebelski, Il Diritto Mite (Torino, Einaudi, 1992). 40   I am referring in here to the holistic notion of constitutionalism described and discussed by Matej Avbelj, ‘Can European Integration be Constitutional and Pluralist – Both at the Same Time?’, Chapter 16 in this volume. 41   An expression originally crafted by N Komesar, Law’s Limits: The Rule of Law and the Supply of Demand Rights (Cambridge, Cambridge University Press, 2001) 60. 42   R Dahl, Democracy and Its Critics (New Haven, Yale University Press, 1989) 3.

Three Claims of Constitutional Pluralism  79 which are considered as citizens of the national polity. It is not the existence of democracy at national level that is contested but the extent of that democracy.43 There is a problem of inclusion faced by national polities.44 Such problem of inclu­ sion does not exist simply by not taking the others into account in decisions that affect them. National polities are often closed to many which would accept their political contract. National polities tend to exclude many which would accept their political contract and are affected by their policies simply because they are not part of the demos as understood in a certain ethnic, cultural or historical sense. In this way, if national polities can be seen as an instrument of constitutionalism, they also limit its ambitions of full representation and participation. The same occurs with the paradox of the fear of the few and the fear of the many. All major constitutional arguments and doctrines gravitate around a com­ plex system of countervailing forces set up by constitutional law to promote the democratic exercise of power (assuring that the few do not rule over the many) but, at the same time, to limit that power (assuring that the many will not abuse their power over the few). There are two basic fears underlying constitutional dis­ course and organization: the fear of the many and fear of the few. The core of constitutional law is the balance between the fear of the many and the fear of the few. Constitutional law sets up the mechanisms through which the many can rule but, at the same time, creates rights and processes for the protection of the few.45 The final paradox is that of who decides who decides? In reality, the question of ‘who decides who decides’, which appears to dominate debates about the rela­ tionship between national and EU constitutionalism, has long been around in constitutionalism. It is a normal consequence of the divided powers system inher­ ent in constitutionalism. In fact, it can be considered as an expected result of the Madisonian view of separation of powers as creating a mechanism of checks and balances. Though national Constitutions may have developed answers to that question historically, they are a contextual product of certain constitutional prac­ tices and not a systemic feature of constitutionalism. On the contrary, the nature of the organization of power inherent in constitutionalism requires the question to be open and periodically reassessed. 43   The difference between the existence of democracy and the extent of democracy is highlighted by J Elster, Deliberative Democracy (Cambridge, Cambridge University Press, 1988) 99. 44   Dahl points out that polities have a twofold problem: ‘1 – The problem of inclusion: What per­ sons have a rightful claim to be included in the demos; 2 – The scope of its authority: What rightful limits are there on the control of a demos’; n 42, 119. See also David Held, Democracy and the Global Order (Cambridge and Oxford, Polity Press, 1995) mainly Chapters 1 and 10. 45   Bellamy highlights three principles which have defined constitutionalism: rights, separation of powers and representative government. However, in his view, the first has come to predominate in recent years: ‘Rights, upheld by judicial review, are said to comprise the prime component of constitu­ tionalism, providing a normative legal framework within which politics operate’, Bellamy, ‘The Political Form of the Constitution: the Separation of Powers, Rights and Representative Democracy’ in R Bellamy and D Castiglione (eds), Constitutionalism in Transformation: European and Theoretical Perspectives (Oxford, Blackwell Publishers, 1996) 24. Note that Bellamy is, himself, critical of that dom­ inant view of constitutionalism. See R Bellamy, Political Constitutionalism: A Republican Defence of the Constitutionality of Democracy (Cambridge University Press, 2007).

80  Miguel Poiares Maduro These paradoxes are a consequence of the constitutional goal of both preserving and regulating political pluralism. At the same time, Constitutions provide a sub­ stantively partly fictional, but rationally useful, common platform on the basis of which political conflicts assume the nature of competing rational arguments on the interpretation of shared values and not the character of power conflicts.46 These conflicts are replaced by competing interpretations of the Constitution. But in order for constitutionalism to perform this role such pluralism of interpretations is sup­ ported by institutional pluralism: different institutions that guarantee that no set of interests acquires a dominant role and that any definition of the common good can be, at any moment, reassessed and contested. Such pluralism of interpretations and institutions ensures the simultaneous expression and accommodation of political pluralism in a particular political community.47 What constitutional pluralism notes is that such pluralism of interpretations and institutions embraced by constitution­ alism has now extended itself to different legal orders. To understand, however, both the promise and challenges of constitutional pluralism it is important to note that the paradoxes of constitutionalism embody two opposing pulls of modern constitutionalism. One, towards pluralism, linked to the values of freedom, diversity and private autonomy. The other, towards unity or hierarchy, linked with the ideals of equality, the rule of law and universal­ ity. Modern constitutionalism success has been founded on its capacity to recon­ cile both at the level of the state. These opposing pulls are reflected in a tension between the political project of pluralism endorsed by constitutionalism and its legal emphasis on hierarchy and primacy.48 They are, however, mutually dependent. Pluralism is ordered through democracy and in order to fulfil the idea of self-government requires a unified and closed political space. This entails, in turn, an ultimate source of political authority. State constitutionalism in its modern form made that political author­ ity reside in the people. The people is both the site and source of pluralism and the unified entity upon which rests ultimate political authority. This is also linked to a conception of constitutionalism as providing a comprehensive social ordering. Nico Krisch has recently labelled it ‘foundational constitutionalism’, the domi­ nant form of constitutionalism to emerge of political modernity. A conception, where a comprehensive and foundational constitutional settlement constitutes the basis for the realization of both public and private autonomy.49 46   Rational discourse through the Constitution is the guarantee of a common identity and the stabi­ lizer of the political community in a context of pluralism. 47   It is the fact that this point is often missed that explains many misunderstandings in the neverending debate about the role and legitimacy of judicial review. Also lawyers appear to be looking for the ultimate authority in determining the meaning of the Constitution when what is taking place can better be described as a form of institutional competition expressly intended by the embedded plural­ ism of the Constitution. I develop this point in M Poiares Maduro, ‘Interpreting European Law: Judicial Adjudication in the Context of Constitutional Pluralism’ (2007) 2 European Journal of Legal Studies 1–21. 48   See, in this volume, the Chapter by Matej Avbelj (n 40) and his discussion of the work of Dieter Grimm. 49   Krisch, n 24, at 47 ff, in particular p 52.

Three Claims of Constitutional Pluralism  81 The most powerful challenge to constitutional pluralism departs therefore from the association made between the values of constitutionalism and the existence of an ultimate source of political authority expressed, in legal terms, in the absolute primacy of the Constitution. These links are considered essential to protecting the constitutional values of the rule of law, equality and universalism. This challenge comes in two very different forms, however. A set of authors argues that the incompatibility between certain constitutional values and plural­ ism requires abandoning pluralism altogether and returning to either monism or dualism. Another set of authors argues that the solution is to be found, instead, in radically departing from constitutionalism as we know it. The first form of criticism was perhaps initiated by Baquero in a well-known piece where he accused constitutional pluralism of undermining the rule of law.50 It is also what largely underlies the empirical criticism made by Alexander Somek, discussed above. In practice, Somek does not conceive as possible for courts to both adhere to the conception of the law internal to their legal order and recog­ nize an external account that challenges that internal conception because that would undermine the authority of the law. The same legal order cannot claim to have authority to universally address all issues within its jurisdictional boundaries and, simultaneously, accept to negotiate that authority with others. As Alexander Somek puts it: ‘Law is intrinsically monistic.’51 In a recent article Pavlos Eleftheriadis makes a comprehensive critique of plu­ ralism focusing on the same perceived incompatibility with core values of consti­ tutionalism. Adopting Dworkin’s concept of integrity as an essential value of the legal order,52 Eleftheriadis argues that constitutional pluralism does not respect that value and, in fact, leads to a legal order of unpredictable results where courts and officials can freely choose between rival constitutional schemes.53 In his own words: The role of the constitutional doctrine is to impose some order on the way in which the various doctrines of law hang together. It is the nervous system of the law. If we say that our constitutional doctrine is relaxed about the coherence of its various frameworks, then we abandon coherence everywhere else in the system.54

Curiously, however, Eleftheriadis endorses a form of dualism. The reason is that integrity is always to be assessed from the point of view of an individual legal order.55 This may sound formalistic, particularly when he adds that any such legal order must be able to accommodate in law its relations with other states and international law.56 For him, in practice, such accommodation is achieved by 50   Baquero Cruz, n 5. I debated some of those criticisms with Baquero Cruz in M Avbelj and J Komárek (eds), ‘Four Visions of Constitutional Pluralism’ (2008) 2 European Journal of Legal Studies 325. 51   Somek, n 5, p 372. 52   R Dworkin, Law’s Empire (Cambridge, Harvard University Press, 1986). 53   P Eleftheriadis, ‘Pluralism and Integrity’ (2010) 23 Ratio Juris 365, 377–78. 54   Ibid, 378. 55   Ibid, 381. 56   Ibid, 381 and 388.

82  Miguel Poiares Maduro recognizing the other legal order under a doctrine of dualism. One may wonder, then, what exactly is the distinction from pluralism or how is political integrity really protected when conflicts occur between these legal orders. I think the differ­ ence from pluralism, for Eleftheriadis, is to be found in the fact that dualism does not impose an obligation of mutual accommodation on both legal orders. They co-exist and prevalence of one or the other is a simple function of jurisdictional power. Constitutional pluralism, on the other hand, requires mutual recognition but also communication and accommodation. As to political integrity, the reason why Eleftheriadis assesses it by reference only to each individual order must be found in the connection between the Dworkinian value of political integrity and the constitutional assumption of a unified and closed political space subject to an ultimate source of political authority. The obligation of political integrity is natu­ rally linked to that conception of political authority and stems from the unique position of the latter in a state. The problem occurs when, as in the postnational context we currently live in, it is difficult to continue to talk about unified and closed political spaces subject to an ultimate source of political authority. We can still do it in conceptual terms by artificially closing and insulating national polities under a self-referential notion of political authority that extends so far as the legal hierarchy and claim of supremacy of the constitutional order itself claims to extend. But this is a purely circular reasoning. More importantly, trust in political integrity will gradually erode as the purported coherence and universality of any particular legal order is increasingly challenged, in practice, by its interaction with other legal orders. In this respect, constitutional pluralism does nothing more than adapt constitu­ tionalism to the changing nature of the political authority and the political space.57 The challenge is to adapt it while protecting political integrity and the correspon­ dent ideals of coherence and universality of the legal order. This is what I have attempted to do in a previous work. I have developed a set of criteria to be adopted by all actors of the European legal system so as to preserve coherence and univer­ sality in a context of pluralism.58 I called them the principles of contrapunctual law so as to highlight that, as in musical counterpoint harmonization of different mel­ odies, so too in Europe’s constitutional pluralism it would be possible to preserve coherence and integrity if all actors, while preserving the internal viewpoint of their legal order, will commit to some common meta-methodological principles of substantive and procedural character in the protection of coherence and universal­ ity and broadly the promotion of political integrity beyond the state. Eleftheriadis recognizes my effort but considers that my principles require a unifying institutional basis. For that reason, my theory would be a form of implicit monism.59 To prevent the risk of entering into a discussion about the monist label I will limit myself here to two brief notes. I have no problem in recognizing that 57   Perhaps it will be even more appropriate to talk, as Nico Krisch does, of a change in the forms of public and private autonomy. 58   See Poiares Maduro, n 11. 59   Eleftheriadis, n 53, 387.

Three Claims of Constitutional Pluralism  83 my theory of constitutional pluralism is different from radical forms of plural­ ism60 or even from politically or international law arbitrated forms of pluralism.61 But it does not impose a unifying institutional basis. Moreover, the meta-­ principles of contrapunctual law do not put into question pluralism because they are themselves a product of pluralism: the rules of the game are entered into by playing the game according to the rules. The meta-principles or rules of discourse are not hierarchically imposed by an external authority. In fact, any of the partici­ pant courts can ‘propose a rule’ by acting or narrating the law in a certain way and that becomes a rule of the discourse to the extent that it becomes part of the dis­ course itself. Each judicial decision is an illocutionary act but, in this respect, their performative value is dependent on their discursive value. What I argue is that once courts are in a context of constitutional pluralism they ought to manage the risk of constitutional conflicts, in the context of the European Union, by embrac­ ing those principles. This is so, because these principles provide the best way for all courts involved to fulfil, in a pluralist context, the promise of constitutionalism that they are bound to pursue. Another group of authors believes, however, that constitutional pluralism requires a radical departure from constitutionalism itself. Matej Avbelj, in his contribution to this book, put this in the form of an alternative faced by pluralists: either to remain faithful to the conventional meaning of constitutionalism and give up on pluralism or remain genuinely pluralist and radically redefine consti­ tutionalism.62 In a similar vein, Nico Krisch also argues that the solution is to be found beyond constitutionalism, particularly if understood in its foundational modern form. Pluralism is the best way to pursue the ideals of constitutionalism in a postnational world, but that requires a radical departure from constitutional­ ism as we know it.63 However, as I stressed before, pluralism is inherent in constitutionalism, even in its conventional form. The tension between the unitary conception of the law and political authority, on the one hand, and the pluralist conception of society and the political space, on the other, is at the core of the constitutional project, at least in the modern liberal form that came to dominate its conventional under­ standing. As such, constitutional pluralists are not confronted with anything new. But we must address that tension in a new context. In other words, a new recon­ ciliation must be attempted between pluralism and unity to update constitution­ alism to the needs of postnational political communities.64 Constitutional pluralists try, in effect, to address the old constitutional para­ doxes in the new forms that result from the current practice of law in a pluralist   See, notably, J Komárek, in Avbelj and Komárek, n 50, 343–44.   I am thinking of the different versions of pluralism that can be found in Neil MacCormick’s work: ‘Beyond the Sovereign State’ (1993) 56 Modern Law Review 1 and Questioning Sovereignty: Law, State and Nation in the European Commonwealth (Oxford, Oxford University Press, 1999). 62   Avbelj, n 40. 63   Krisch, n 24. 64   Referring both to the emerging political communities beyond the state and to the challenges faced by traditional political communities in a context of regional and global integration. 60 61

84  Miguel Poiares Maduro context. As Neil Walker puts it: ‘The constitutional pluralist seeks to retain from constitutionalism the idea of a single authorising register for the political domain as a whole while at the same time retaining from pluralism a sense of the rich and irreducible diversity of that political domain.’65 To sum up, the promise of consti­ tutional pluralism lies in the success with which it will be able to reconcile again the opposing pulls of pluralism and unity that have always dominated constitu­ tionalism.

  N Walker, ‘Constitutionalism and Pluralism in Global Context’, Chapter 2 in this volume, p 17.

65

5 Systems Pluralism and Institutional Pluralism in Constitutional Law: National, Supranational and Global Governance DANIEL HALBERSTAM*

I. Introduction Constitutions are often seen as creating a closed and hierarchically organized system of law. Constitutional systems are taken as closed to claims of legality from outside the system and as setting forth a hierarchy of norms and institutions that governs within the system. This consolidation of authority, in turn, is predominantly associated with a radical political (re)founding of the state. Politics are framed by law and law is grounded in an act of collective politics on the part of an existing or aspiring community defined by shared histories, norms, processes, and politics. This unified vision of constitutionalism has increasingly come under stress. Especially as we confront globalization as a legal phenomenon, the question has become whether constitutionalism, as we (think we) know it, is coming to an end. Claims of legal authority from beyond the state, where law seems fragmented and a political community thin or absent, seem increasingly to infiltrate governance decisions within the state. At the level of global governance, the proliferation of functionally specific regimes in the absence of a unifying logic leads to a bric-à-brac that challenges our basic understanding of law as organization and settlement. There have been many proposals of how best to approach this emerging problem. Some argue for an administrative approach to global governance by bringing good governance practices from the national to the global level of governance.1 *  Eric Stein Professor of Law and Director, European Legal Studies Program, University of Michigan Law School. Thanks to Scott Hershovitz, Don Herzog, Dieter Grimm, Christoph Möllers, Ellen Katz and the participants in the SIAS Summer Institute on Federalism and Separation of Powers for helpful comments and discussions, to Michael Huston, Jonathan Fombonne, Sophia Henrich and Philippe Reyniers for research assistance, and to the libraries at the University of Michigan and, especially, at the Wissenschaftskolleg (Institute for Advanced Study) Berlin, where this Chapter was written. 1   See, eg, S Cassese et al (eds), Global Administrative Law: Cases, Materials, Issues 2nd edn (New York, IILJ, 2008); RB Stewart, ‘U.S. Administrative Law: A Model for Global Administrative Law?’ (2005) 68 Law and Contemporary Problems 63; DC Esty, ‘Good Governance at the Supranational Scale: Globalizing Administrative Law’ (2006) 115 Yale Law Journal 1490.

86  Daniel Halberstam Some argue for consolidating authority at the level of global governance in a unified constitutional order that spans the world.2 Some argue for a retreat into national Constitutions as the only source of legitimate legal authority while tapping into global norms only to the extent deemed convenient by local politics.3 And some argue for an understanding of global governance that abandons notions of order in favour of a free-floating plurality of legal claims.4 Despite the great diversity of views presented in these approaches, they all seem to share a common point of departure. One way or another, they seem to work from within the traditional view of constitutional law as the consolidation and hierarchical ordering of legal authority. This Chapter aims to challenge the idea of constitutionalism as complete consolidation and hierarchy. The Chapter teases out various ways in which the practice of constitutionalism may be open to claims from outside the system and lack certain forms of ordering through hierarchy within the system. This does not result in anarchy or chaos. Instead, under certain conditions, the lack of closure and hierarchy produces a principled practice of legal order. Indeed, as I shall suggest, it may produce a constitutional practice that is more true to the ideals of constitutionalism than the traditional model of consolidation and hierarchy itself. The hope is that this pluralist conception of constitutionalism will allow us to approach questions of national, supranational, and global governance afresh. The Chapter proceeds in three parts. The first part briefly describes the traditional narrative of constitutionalism as consolidation and hierarchy of legal authority. The second part discusses at somewhat greater length the pluralist alternative. This part focuses on two distinct aspects of pluralism and explores the extent to which each can be seen as operative within and beyond the state. The last part is the conclusion.

2   See, eg, C Tomuschat, ‘International Law: Ensuring the Survival of Mankind on the Eve of a New Century, General Course on Public International Law’ (1999) 281 Recueil des Cours 9; H Mosler, The International Society as a Legal Community (Alphen aan den Rijn, Sijthoff, 1980); B Fassbender, The United Nations Charter as the Constitution of the International Community (Leiden, Martinus Nijhoff, 2009). 3   See, eg, EA Posner, JL Goldsmith, The Limits of International Law (Oxford, Oxford University Press, 2005); J Yoo, The Powers of War and Peace: The Constitution and Foreign Affairs After 9/11 (Chicago, University of Chicago Press, 2005); CA Bradley, ‘International Delegations, the Structural Constitution, and Non-Self-Execution’ (2003) 55 Stanford Law Review 1557. 4   See, eg, G Teubner, ‘Global Bukowina: Legal Pluralism in the World Society’ in G Teubner (ed), Global Law Without a State (Aldershot, Dartmouth, 1997) 3.

Systems and Institutional Pluralism  87

II.  Constitutionalism, Hierarchy, and the Narrative of Progress A.  A Thumbnail Sketch of History Constitutionalism emerges from a rather diverse set of historical traditions of framing law through politics and politics through law. One can easily trace the origins of the idea of a Constitution to ancient societies, such as Greece and Rome,5 and perhaps even further back, for example to the biblical covenant between God and the Israelites.6 But the modern idea of constitutionalism as limited collective self-governance through law began in earnest around the time of the English Civil War and gained broad force towards the end of the eighteenth century with the classic revolutionary constitutions of the United States and France. Constitutionalism as a legal limitation of power is evident from the moment that Constitutions are first invoked in the struggles between King and Parliament that prefigured the English Civil War. James Whitlocke was perhaps the first to use the term in this way when he objected in 1610 to the impositions of James I on the grounds that they were ‘against the natural frame and constitution of the policy [that is, “polity”] of this kingdom, which is jus publicum regni, and so subverteth the fundamental law of the realm and induceth a new form of State and government’.7 Such arguments (albeit not always using ‘constitutional’ terminology) became a staple in revolutionary thought from the Petition of Right to the Glorious Revolution. Charles I, for instance, is formally accused in 1649 of having violated the ‘fundamental constitutions’ of the Kingdom8 and James II is finally evicted in 1689 because the latter had ‘endeavoured to subvert the constitution of the kingdom’.9 Subversion of these limits is what later motivated American Founders as well. If we take Bernard Bailyn’s account, for example, fears of the ‘destruction of the English constitution, with all the rights and privileges in it, . . . above all else . . . propelled [the Americans] into revolution’.10 5   H Mohnhaupt, ‘Verfassung I’ in R Koselleck et al (eds), Geschichtliche Grundbegriffe (Stuttgart, Klett, 1990) 831, 833–36. 6  Cf H Brunkhorst, Solidarität: Von der Bürgerfreundschaft zur globalen Rechtsgenossenschaft (Frankfurt am Main, Suhrkamp, 2002) 46; H Arendt, On Revolution (New York, Penguin Books, 2006) 190. 7  JR Tanner, Constitutional Documents of the Reign of James I: AD 1603–1625 (Cambridge, Cambridge University Press, reprint 1961) 260; see also G Stourzh, From Vienna to Chicago and Back: Essays on Intellectual History and Political Thought in Europe and America (Chicago, University of Chicago Press, 2007) 90. 8   SR Gardiner (ed), The Constitutional Documents of the Puritan Revolution 1625–1660 3rd edn (Oxford, Oxford University Press, 1906) 372. 9   House of Commons Journal, Vol 10, 28 January 1689, 14. 10   B Bailyn, The Ideological Origins of the American Revolution (Cambridge, Harvard University Press, 1967) 95.

88  Daniel Halberstam The idea of constitutionalism as a legal limitation on power also prefigures the French Revolution. Here, the thought was initially captured in different terms, in the notion of so-called ‘lois fondamentales’, that is basic laws that a prince could not violate because they are fundamental to his reign.11 As Diderot and d’Alambert’s Encyclopaedia puts it mid-century, fundamental laws serve as the foundation on which government is built and therefore even ‘prescribe boundaries to sovereign authority’.12 Or, to quote Vattel’s definition in the Law of Nations, which shortly after became influential throughout Europe, a Constitution is the set of ‘fundamental laws’ that determine the ‘manner in which public authority must be exercised’.13 Limiting power by subsuming it within a legal framework that governs power takes on rather different significance in different national traditions. In both England and Germany, for example, constitutional law was historically used to frame and temper previously existing power structures, whereas in the United States and France, constitutional law was ultimately deployed to break with the existing structure and give law and the administration of governance a new polit­ ical foundation.14 Constitutionalism in England, moreover, was part of a revolutionary yet inward-looking discourse of preserving, discovering, and making explicit legal principles that were thought to be immanent in the established system of governance.15 In Germany, by contrast, constitutionalism in the various states first came in the absence of revolutions and characteristically took the form of the reigning monarch unilaterally granting self-limiting constitutional treaties while, ultimately, consolidating and preserving all rights to the exercise of state power in the king or prince himself.16 Towards the latter part of the nineteenth century, the federal Constitution of the Reich imposed limitations and rules on the various component states via eternal interstate compact.17 Only in 1919 did Germany begin with a radical democratic transformation of its own.18   Mohnhaupt, n 5.   Encyclopédie de Diderot et d’Alambert 660 (1751) (s.v. ‘loi fondamentale’); see also A Vergne, La Notion de Constitution d’Après les Cours et Assemblées à la Fin de l’Ancien Régime (1750–1789) (Paris, de Boccard, 2006). 13   E de Vattel, Le Droit des Gens, ou Principe de la Loi Naturelle Appliqué a la Conduite et aux Affaires des Nations et des Souverains Vol 1 (London, 1758), Ch 3, paras 27, 29; see also Mohnhaupt, n 5 at 859; Vergne, n 12 at 36. 14   Cf C Möllers, ‘Verfassungsgebende Gewalt – Verfassung – Konstitutoinalisierung’ in A von Bogdandy (ed), Europäisches Verfassungsrecht: Theoretische und dogmatische Grundzüge (Berlin, Springer, 2002). 15   See, eg, M Rosenfeld, The Identity of the Constitutional Subject: Selfhood, Citizenship, Culture, and Community (London and New York, Routledge, 2010) 164. 16   D Grimm, Deutsche Verfassungsgeschichte 1776–1866 (Frankfurt am Main, Suhrkamp, 1988) 113– 16. 17   This compact was nonetheless in the nature of a Constitution. For perhaps the most prominent defence of this idea see P Laband, Das Staatsrecht des Deutschen Reiches (Tübingen, JCB Mohr, 1911) 55–102. 18   On these troubled beginnings, see PC Caldwell, Popular Sovereignty and the Crisis of German Constitutional Law: The Theory and Practice of Weimar Constitutionalism (Durham, Duke University Press, 1997). 11 12

Systems and Institutional Pluralism  89 Limiting government in this way does not mean making government ineffective. To the contrary, as the French experience with the demise of constitutionalism has taught, ‘limited’ government can be more stable and far more effective over the long term than any short reign of terror can hope to be. Jean Bodin had already drawn a similar lesson from the religious wars of sixteenth-century France in developing the idea that absolute sovereignty was nonetheless bound by certain rules. In all his capacity to declare law only for others and never for himself, Bodin’s sovereign was nonetheless bound by the law that established his own power.19 Self-limitation can take many forms, such as the consultation of parliamentary assemblies, full-fledged separation of powers, recognition of parental jurisdiction over their children, and toleration of private religious practice. Each of these can serve to enhance the power of the prince, whether in terms of solving epistemic problems in policy making or gaining the adherence of its citizens and calming civil discord. As Stephen Holmes has explained, modern democratic constitutionalism is the practical successor to this basic insight: constitutional government is stronger because of its ‘gag rules’ and owes its effectiveness and, indeed, its very existence to the various ‘precommitments’ that take the form of constitutional ‘limits’ on various actors.20 Who is to be served by all this? The national constitutional movement has been associated with the expression of a national political will ever since the Rump Parliament of 1648 claimed to speak for ‘the People of England and of all the Dominions and Territoryes thereunto belonging’.21 Emmerich de Vattel, perhaps the first scholar in modern times to put forth an exclusively legal definition of the term ‘Constitution’,22 similarly noted that ‘the chief interest of a Nation which forms a political society, and its first and most important duty to itself, is to choose the best possible constitution, and the one most suited to its circumstances’.23 Although eliding the question of democracy by distinguishing between ‘the people’ and ‘the Nation’, Vattel’s definition implies a decision of the Nation for itself. Vattel’s capacious definition could thus include even Germany’s imperial Constitution of 1871, which was enacted ‘for the welfare of the German nation’ on the part of the princes and kings who authored the interstate compact.24 The idea of a national choice, in any event, certainly accords with the classic invocation of subsequent Constitutions in the radical democratic tradition, as that of Massachusetts in 1780 (‘We, . . . the people of Massachusetts’), the United States in 1789 (‘We the People of the United States’), France in 1958 (‘le people 19  J Bodin, On Sovereignty: Four Chapters From Six Books of the Commonwealth (Cambridge, Cambridge University Press, 1992) 18. 20  S Holmes, Passions and Constraints: On the Theory of Liberal Democracy (Chicago, Chicago University Press, 1995) 109–31. 21   An Act Declaring England to Be a Commonwealth, May 19, 1649 (Eng.), reprinted in Gardiner (ed), n 9. 22   Mohnhaupt, n 5 at 859. 23   Vattel, n 13 at Book I, ch III, para 28. 24   Preamble, Constitution of the German Empire (Verfassung des Deutschen Reiches) (1871), in D Dosenwinkel and J Masing (eds), Die Verfassungen in Europa (Munich, CH Beck, 2006) 784.

90  Daniel Halberstam français’), or Poland in 1997 (‘We, the Polish Nation – all citizens of the Republic’). We can put to one side here the debates about whether the existence of a nation preceded such acts of Constitution making or whether the act of Constitution making itself created or promised to create the bounded polity in whose name the Constitution was proclaimed. One way or another, the modern tradition of constitutionalism as limited governance is based on the idea of self-expression and self-governance of a political community.

B.  A Narrative of Legal and Institutional Hierarchy Modern constitutionalism is often seen as an historical achievement. Of course, there is the real triumph of liberty and the idea of limited governance over absolutism and divine rule. More interesting, though, is the conceptual triumph associated with the radical republican traditions of the United States and France over what is considered to be a less-developed form of constitutionalism in England. Similarly, Germany’s progression from a monarchy to republic is associated with the movement of modern constitutionalism beyond an older, less-developed form of framing power through law. Bound up with this narrative are often two aspects of hierarchy. The first is the primacy of the Constitution’s legal system and legal norms over all other claims of public authority. The second is the primacy within the constitutional system of a single institution, such as a constitutional court, to serve as final arbiter of constitutional meaning. Dieter Grimm, for example, prominently gives voice to the first aspect of this narrative. England may have pioneered constitutional limitations of power in the modern sense but, Grimm suggests, England did not labour under the social and economic tensions that led to the embrace of radical constitutionalism in the United States and France.25 Reviewing the conceptual history of modern constitutionalism, he therefore concludes that England may well have ‘developed’ the modern constitutional idea, but that the new approach to governance was not ‘completed’ until it took its radical American and French forms.26 Modern constitutionalism, on this view, only comes to fruition in the United States when the people enact a documentary Constitution that is ‘antecedent to government’27 with substantive elements such as basic rights and a claim of authority that ‘takes primacy over all other acts of rule’.28 Niklas Luhmann similarly celebrates the historic achievement of French and American radical constitutional law.29 Luhman suggests an ‘evolutionary’ progression based on various elements including, for example, that ‘constitutional’   Grimm, n 16 at 33–36.   Dieter Grimm, ‘Verfassung II’ in R Koselleck et al (eds), n 5 at 859. 27   T Paine, The Rights of Man (London, Watts & Co, 1937) 36. 28   D Grimm, ‘The Achievement of Constitutionalism and its Prospects in a Changing World’ in P Dobner and M Loughlin (eds), The Twilight of Constitutionalism? (Cambridge, Cambridge University Press, 2010) 3; see also Grimm, n 26 at 866–67. 29   N Luhmann, ‘Verfassung als evolutionäre Errungenschaft’ (1990) 9 Rechtshistorisches Journal 176. 25 26

Systems and Institutional Pluralism  91 language simply lent itself well to charging authorities with the ‘unconstitutionality’ of their acts in England.30 Constitutionality and the lack thereof thus quickly became the centrepiece in the discourse surrounding legitimacy. This, in turn, conjured up a kind of paradox of authority concerning the ultimate basis of legality. The embrace of a radical republican Constitution was a ‘deparadoxification’ tool to respond to this problem.31 The radical Constitution is itself law, creates a legal system, defines and limits the powers of those within the system, sets its legal system off from other normative systems, and manages the interaction between its legal system and those other systems on its own terms. These other systems can be systems of politics or economics, systems of morality or ethics, and, most important for our purposes, other systems of law beyond the one the Constitution itself creates. Although grounded in politics, a Constitution, viewed thus, depends on no law outside itself for its claim of legal authority. The radical Constitution controls its own legality. It is an act, as Luhmann grandly calls it, of ‘autopoiesis’.32 The second movement towards settled hierarchy concerns the institutions of governance within the constitutional system. Here, the story is one in which British parliamentary supremacy is replaced by American-style judicial review, which is, in turn, replaced by the formal elevation of a high court with express authority to control the interpretation of the Constitution for all participants within the system. On this view, parliamentary supremacy is the hallmark of the older, incomplete version of British constitutionalism. Because in Britain (as compared to France) the historical conflicts surrounded excesses of the Crown against the Parliament, and a moderately liberal economic system had already empowered the bourgeoisie and mitigated the power of the aristocracy, the revolutionary movement united the aristocracy and bourgeoisie against the king but failed to attack the power of parliament.33 This led to what is portrayed as the ‘incomplete’ development of constitutionalism in England as compared to the United States or France. In the narrative of institutional hierarchy, however, the United States and France are no longer seen as completed or ideal forms of constitutionalism. To the contrary, with regard to judicial review these systems are understood to be largely stuck, each for their own historical reasons, in an intermediate stage of imperfection as well. Although judicial review has come about in these systems, it emerged only as the product of judicial improvisation and self-empowerment of the kind that has become synonymous the world over with the American case of Marbury v Madison.34 In the United States, this act of improvisation has led to endless debates about the counter-majoritarian difficulty and the constitutional warrant for courts to

  Ibid at 188–89.   Ibid at 185. 32   Ibid. 33   See, eg, Grimm, n 16 at 33–36. 34   Marbury v Madison, 5 U.S. 137 (1803). 30 31

92  Daniel Halberstam displace politics.35 In France, largely because of the deep mistrust of the judiciary and the association of judges with the ancien régime, judicial self-invention emerged only much later.36 Constitutional review arose in France only towards the end of the last century, and was not authoritatively embraced until rather recently with a constitutional amendment on judicial review.37 This brought France into the mainstream of what has been called the ‘new constitutionalism’, as distinguished from ‘old constitutionalism’ (putting aside for the moment that both of these are forms of what we have called ‘modern constitutionalism’ so far): In Europe, the move from the ‘old constitutionalism’ . . . to the ‘new constitutionalism’ depends heavily on constitutional review as a mechanism of governance. The precepts of this new constitutionalism can be simply listed: (a) state institutions are established by, and derive their authority exclusively from, a written constitution; (b) this constitution assigns ultimate power to the people by way of elections; (c) the use of public authority, including legislative authority, is lawful only insofar as it conforms with the constitutional law; (d) the constitution provides for rights and a system of constitutional justice to defend those rights; and (e) constitutional courts not only have a duty but are fully empowered to manage this system of justice. As an overarching political ideology, or theory of the state, the new constitutionalism faces no serious rival today.38

The new (modern) constitutionalism that expressly grants courts such complete power to manage this system of justice is presented as overcoming not only parliamentary supremacy and the lack of a documentary Constitution but also the backward American angst about judicial review. As Stone Sweet has said: New European constitutions expressly provide for the supremacy of constitutional courts with respect to constitutional interpretation. European academics and constitutional judges will state as much in one breath, and then move on to more interesting issues.39

C.  Constitutional Blinders? The narrative of radical constitutionalism, with its story of modernization, consolidation, and hierarchy, can create conceptual blinders. Many shun the label of constitutionalism when describing or analysing phenomena that do not fit the ideal of complete consolidation and hierarchy. Others, at times, appeal to constitutionalism as an ideal that all legal systems should somehow seek to achieve. 35   B Friedman, ‘The Birth of an Academic Obsession: The History of the Countermajoritarian Difficulty, Part Five’ (2002) 112 Yale Law Journal 153. 36   A Stone Sweet, The Birth of Judicial Politics in France: The Constitutional Council in Comparative Perspective (Oxford, Oxford University Press, 1992). 37   See, eg, Federico Fabbrini, ‘Kelsen in Paris: France’s Constitutional Reform and the Introduction of A Posteriori Constitutional Review of Legislation’ (2008) 9 German Law Journal 1297. 38   A Stone Sweet, ‘The Politics of Constitutional Review in France and Europe’ (2007) 5 International Journal of Constitutional Law 69, 74–75. 39   A Stone Sweet, ‘Why Europe Rejected American Judicial Review: And Why It May Not Matter’ (2003) 101 Michigan Law Review 2744, 2779.

Systems and Institutional Pluralism  93 The European Union, for example, has long been described as either inter­ national or ‘sui generis’ by those who resist the conceptual move to understand the Union in constitutional terms.40 Some scholars seem to deny that the United Kingdom has a Constitution in large part because there is no single written document that expressly claims foundational status and supremacy over all other claims of rule.41 And in the world of global governance some argue (within the traditional narrative) for the consolidation of international law under one global constitutional umbrella,42 while sceptics note (also from within the traditional narrative) that ‘the high degree of differentiation and the internal complexity of the international level that it reveals and takes as fundamental, is incompatible with the unified world and the holistic approach of constitutionalism’.43 Against this trend, a group of scholars are beginning to resist this narrative of consolidation and the equation of constitutionalism with a single, largely imagined brand of the phenomenon of constitutional governance. Some, for example, seek to recover earlier, less consolidated forms of federal constitutionalism to help understand the continuities between the European Union and other forms of federalism.44 Others have tried to rehabilitate a ‘Commonwealth’ form of constitutionalism that combines parliamentary supremacy with a responsible practice of rights protection and judicial involvement.45 Yet others have teased out various strands of constitutionalism, from the radical framing of law through politics to the gradual framing of politics through law, as each representing proper members of the family of modern constitutional projects.46 The idea of pluralism seeks similarly to contribute to these debates by challenging the traditional vision of completion and hierarchy in constitutional law.47 Focusing both on the normative and institutional dimensions of a legal order, pluralism challenges the idea that the practice of constitutionalism necessarily implies either a closed legal system or one that is hierarchically ordered internally. To the contrary, the challenge is nothing less than to reimagine constitutionalism as an inherently pluralist practice. 40   Eg T Hartley, ‘International Law and the Law of the European Union’ (2001) 72 British Yearbook of International Law 1. 41   See, eg, E Wicks, ‘A New Constitution for a New State? The 1707 Union of England and Scotland’ (2001) 117 Law Quarterly Review 109, 114 (‘[W]ithin the United Kingdom . . . the existence of a “Constitution” remains an issue of heated debate’). 42   See, eg, Tomuschat, n 2; Fassbender, n 2. 43   R Wahl, ‘In Defense of “Constitution”’, in P Dobner and M Loughlin (eds), n 28 at 220. 44  See O Beaud, Théorie de la Fédération (Paris, Presses Universitaires de France, 2007); C Schönberger, Unionsbürger (Tübingen, Mohr Siebeck, 2006); R Schütze, From Dual to Cooperative Federalism: The Changing Structure of European Law (Oxford, Oxford University Press, 2009). 45   S Gardbaum, ‘The New Commonwealth Model of Constitutionalism’ (2001) 49 American Journal of Comparative Law 707. 46   See, eg, Kumm, M, ‘Beyond Golf Clubs and the Judicialization of Politics: Why Europe Has a Constitution Properly So Called’ (2006) 54 American Journal of Comparative Law 54; Möllers, n 14; Rosenfeld n 15. 47   This Chapter provides one particular take on pluralism. Others writing in this still-emerging tradition are well represented in the pages of this book. See, eg, the contributions in this volume by Jan Komárek, Mattias Kumm, Miguel Maduro, and Neil Walker.

94  Daniel Halberstam

III.  The Pluralist Alternative Multiple claims of legal authority can arise because of multiple legal systems or because, within a given legal system, there are multiple legal actors, sources, or norms. We may, of course, find all of these or a combination of these present at the same time. Either way, whenever we confront a plurality of systems, actors, sources, or norms in the absence of a settled hierarchy or neat division of labour among them, we are confronted with potentially rivalling legal authorities. This may give rise to the practice of pluralism. To unpack pluralism as a legal and constitutional practice, it is useful to distinguish between two distinct kinds.48 The first is a pluralism of legal systems, or systems pluralism for short. The second is a pluralism of institutional actors within a single legal system, or institutional pluralism for short. For both of these – systems pluralism and institutional pluralism – we can set out in simple terms the conditions under which they arise and why it makes sense to speak of a coherent practice of pluralism as opposed to a hopeless cacophony of authorities. We will also see why it makes sense to distinguish between these two different types of pluralism and why it yet makes sense to consider them together.

A.  Systems Pluralism A multiplicity of legal systems does not necessarily mean that their claims of legal authority conflict (in the proper sense of that term). For example, multiple legal systems may have authority only over mutually exclusive subjects by virtue of an overarching system of law to which they both belong. This is true in the case of Austrian federalism, for instance, where federal and component state jurisdictions are mutually exclusive by virtue of the constitutional law of the system as a whole.49 So, too, multiple systems with potentially conflicting claims of authority may be hierarchically ordered. This is true for the United States, in which the Supremacy Clause of Article VI of the United States Constitution is (today) accepted by all participants and sometimes even formally reaffirmed in component state provisions such as California’s Article III, Section 1, which provides: ‘The State of California is an inseparable part of the United States of America, and the United States Constitution is the supreme law of the land.’50 And the claims of multiple legal systems may be ordered by a simple timing rule, as in the case of

48   For the initial articulation of this distinction, see D Halberstam, ‘Constitutionalism and Pluralism in Marbury and Van Gend’, http://ssrn.com/abstract=1103253, published as amended in MP Maduro and L Azuolai (eds), The Past and the Future of EU Law: Revisiting the Classics on the 50th Anniversary of the Rome Treaty (Oxford, Hart Publishing, 2010). 49   See Articles 10–15 of the Austrian Federal Constitutional Law (Bundes-Verfassungsgesetz). 50   California Constitution, Article III, § 1.

Systems and Institutional Pluralism  95 German federalism where in certain matters central or Länder law takes precedence depending on whichever comes last.51 To the extent we can speak of multiple legal systems in the case of such synchronized systems at all, no true systems conflict arises. Conflict about authority may exist, but such conflict is, strictly speaking, not one that concerns a clash of systems. It is, instead, a dispute about the proper understanding of the common legal norms that in each case comprehensively control the various claims of authority. In these systems, where the legality and terms of application of common norms are clear to everyone involved, no conflict of authority should arise. To put the matter clearly: if all participants agree that a particular federal law is constitutional and applicable according to, say, the US Constitution or the German Grundgesetz, no component state in either system would today seriously maintain that such a federal law violates the component state’s legal constitutional order and will therefore not be applied within that component state. The contention that otherwise valid central law is inapplicable because of a violation of the component state’s legal order would amount (today) to a misunderstanding of these federal systems, a revolutionary act, or an act of legal (or civil) dis­ obedience. It is not a feasible interpretation of these legal systems to maintain that this kind of clash of authorities can legally exist. Where the relationship among multiple systems is not controlled by mutually accepted tie-breakers or an overarching legal order, however, conflicting claims to final legal authority can indeed arise. This may, of course, happen in the ordinary conflicts-of-laws situation, where two unrelated states have conflicting laws or jurisdictional claims regarding the same contract, tort, or other transaction.52 And it may happen among multiple jurisdictions that are situated within a common federal system such as the United States, where the overarching legal system to which the component states belong controls important aspects of the conflict-oflaws among those states.53 The idea of systems pluralism, however, concerns neither the ordinary nor the federal/constitutional conflicts of laws scenario but lies somewhere in between. Systems pluralism, as we shall see, begins with a true conflict of final legal authority. In addition to conflicting claims of authority, however, each system that is party to the conflict also contains an inherent openness to the claim of authority of the other. And yet, there is no mutually accepted tie-breaker. Neither claim automatically trumps the other, for that would simply replace pluralism with hierarchy once more. Instead, it recognizes the normative pull of the claim of legality of the other without simply submitting to that claim. Finally, for this combination of independence and mutual embedded openness to work productively, we also need some sense of common cause among the various sites of governance.   See Article 72(3) of the Basic Law.   See, eg, WM Reisman, Introduction to Jurisdiction in International Law (Aldershot, Ashgate, 1999) xi. 53   Eg, in the United States, the Constitution controls such things as the mutual recognition of legal judgments, the extradition of fugitives from justice and certain aspects of the extraterritorial reach of component state law. See US Constitution Art IV, §§ 1 and 2. 51 52

96  Daniel Halberstam At the most abstract level, this shared purpose is a common commitment to limited collective self-governance through law.

i.  The Case of the European Union a.  Multiple claims to final legal authority In the European Union, for example, both the EU and the Member States make claims to final legal authority in the sense that each insists on a superior claim of authority for its own legal system over that of the other. To cut a long and all-toowell-known story short, the European Court of Justice has long interpreted the Treaties as a constitutional charter, holding that the subjects of the European legal order are not only Member States but also citizens and that the European Union’s ‘autonomous legal order’ has normative bite that is supreme over Member State law (including Member State constitutional law) and directly effective within the Member State’s legal orders.54 The ECJ’s constitutional reasoning radiates well beyond this summary and influences its interpretation of the law from the rights of the European Parliament to citizenship, fundamental rights, and social policies.55 Indeed, as far as EU law is concerned, it governs even modifications to the Treaty itself as well as a Member State’s exit from the European Union altogether.56 Opposing the EU’s claim to final legal authority, Member State constitutional systems have, by and large, maintained the superiority of their own national legal systems.57 Perhaps most prominently in this regard, the German Bundesverfassungsgericht has repeatedly insisted that German institutions will give effect to EU law only to the extent allowed by the German Constitution.58 These rival claims to final legal authority are not just paper claims. They are real in the sense that Member State and EU legal systems are each separately genera-

54   Case 26/62 Van Gend en Loos [1963] ECR 1; Opinion 1/91 of 14 December 1991 [1991] ECR I-6079, para 21; Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat International Foundation v Council and Commission [2008] ECR I-6351, paras 316 and 317. See also JHH Weiler, ‘The Transformation of Europe’ (1991) 100 Yale Law Journal 2403; E Stein, ‘Lawyers, Judges and the Making of a Transnational Constitution’ (1981) 75 American Journal of International Law 1; GF Mancini, ‘The Making of a Constitution for Europe’ (1989) 26 Common Market Law Review 595. 55   For an elaboration, see D Halberstam, ‘The Bride of Messina: Constitutionalism and Democracy in Europe’ (2005) 30 European Law Review 775. 56   Article 50 TEU. 57  The only exception seems to be a potentially far-reaching 2004 decision on the part of the Netherlands Supreme Court (Hoge Raad, Strafkamer), which (somewhat inexplicably) declared that an EC Regulation was applicable in the Netherlands not because of anything in the national Constitution but simply by virtue of the EC Treaty itself. See A 13V v Public Prosecutor, Supreme Court, 2 November 2004, NJ (2005) No 80, LJN No AR1797. The decision seems to be in considerable tension with the constitutional basis of the Hoge Raad’s own jurisdiction as well as with Art 91(3) of the Dutch Constitution (requiring a supermajority vote in Parliament to ratify treaties that conflict with the Constitution) and should therefore probably be read as a more mundane assertion of absence of any need for routine national legislation to implement EC regulations. 58   See Case 2 BvE 2/08, judgment of 30 June 2009, Lisbon, English translation http://www.bverfg.de/ entscheidungen/es20090630_2bve000208en.html; Case 2 BvR 2134, 2159/92 Maastricht [1993] BVerfGE 89, 155.

Systems and Institutional Pluralism  97 tive sites of governance. Neither Member State nor Union activity is subject to the complete control by the other (or some third party). With regard to Member State legal activity the argument is undisputed. The EU may set agendas and shape domestic politics and law in important ways, but Member States still respond to Europe in ways that are not predetermined by the Union level of governance itself. But even with regard to the European Union, the analogous claim holds true. Even here, we can speak of a separate legal (and political) system not just in formal but also in real terms. The European Union is not the Member States’ puppet – neither that of a single Member State nor that even of the collectivity of Member State governments taken as a whole. b.  Mutual embedded openness The European Union and any Member State legal system are each deeply embedded in the constitutional operating system of the other. As for the European Union, several key aspects of Union law suggest that it has not overcome its structural reliance on Member State legality. From the Preamble’s invocation of Member State authorship, the explicit recognition of Member State constitutional identities, and the central role of Member State transposition and execution of Union law to the ratification procedure for the Treaties themselves, the Union expressly depends on the legal systems of the Member States for its own legal authority. Even the ambitious Treaty on a Constitution for Europe did not seek to overcome this structural reliance on Member State law. Perhaps at its most basic level, what Bruno de Witte has called the ‘national dimension’ of Treaty amendment and ratification has been a steady feature of EU law both as a formal and as a politically consequential operative fact.59 Member States’ systems are similarly open to the European Union. Their embrace of the Union is not a circumscribed political convenience but a broadly operative legal principle embedded in the deep structure of their legal systems. Member State Constitutions, as the laws constituting each national claim of legal authority, contain provisions expressly opening Member State systems to European law. The provisions are well known, such as Italy’s and Germany’s, which expressly allow for the ‘limitation’ or ‘transfer’ of ‘sovereignty’ or ‘sovereign rights’ to the European Union,60 or France’s, which provide for a commitment to, and the ‘transfer of powers necessary’ for the ‘participat[ion]’ in, the 59   See B de Witte, ‘European Treaty Revision: A Case of Multilevel Constitutionalism’ in I Pernice and J Zemanek (eds), A Constitution for Europe: The IGC, the Ratification Process and Beyond (Baden Baden, Nomos, 2005); I Pernice, ‘Multilevel Constitutionalism and the Treaty of Amsterdam: European Constitution-Making Revisited?’ (1999) 36 Common Market Law Review 703, 750; I Pernice, F Mayer, ‘De la constitution composée de l’Europe’ (2000) 36 Revue Trimestrielle de Droit Européen 623, 648; J Ziller, ‘L’Elargissement Change-t-il les Données du Problème?’ in J Ziller et al, L’Européanisation des Droits Constitutionnels à la Lumière de la Constitution pour l’Europe (Paris, l’Harmattan, 2003) 311, 315. 60   See Art 11 of the Constitution of the Italian Republic [Costituzione della Repubblica italiana] and Art 23 of the Basic Law.

98  Daniel Halberstam European Union.61 Such Member State provisions bind the constituted branches of government as well as all legal officials – and, to some extent, citizens – to the project of European integration and to the recognition of at least some claims of legal authority on the part of the European Union. Embedded mutual openness makes way for officials (and citizens) of one system to respect the claims of authority of the other system, but this does not mean the two legal systems blend into one.62 Neither system (today) depends for its legal claim of authority on the legal recognition of authority by the other. More concretely, the constitutional openness of the Member States to the European Union is not (or, perhaps better, no longer) the legal foundation of the EU’s claim of authority over the Member States. Nor is the embedded openness of the EU’s legal system to the constitutional authority of the Member States the foundation of the legal authority of the Member State systems. For instance, a Member State could (as a matter of Member State law) repeal its constitutional openness to the European Union, but the EU’s independent claim of legal authority to have that Member State obey EU law would not thereby legally dissolve. As the European Court of Justice has long held, the Union lays claim to legal authority that takes primacy even over Member State constitutional law.63 This is not an act of judicial solipsism, as we can see from the Lisbon Treaty, which expressly ratifies the Court’s understanding of EU law on this point.64 Moreover, the EU’s foundational Treaties, which govern the act of withdrawal itself, provide that a Member State’s decision to withdraw does not automatically expunge the EU’s claim to legality. EU law will continue to operate with regard to that Member State until the conclusion of a withdrawal agreement or, if such an agreement should fail to come about, for two years after the Member State notified the European Council of its intention to withdraw.65 This suggests that, from the perspective of EU law, the normativity of EU law is no longer a matter of Member State choice. Even the collectivity of Member States seems bound by the Treaty and cannot abolish the Union’s legal system immediately at will – at least not as a matter of EU law. The provisions regarding Treaty revision,66 as well as 61   See Arts 88-1 to 88-7 of the Constitution of the French Republic [Constitution de la République française]; Art 90 of the Constitution of the Republic of Poland [Konstytucja Rzeczypospolitej Polskiej]; Art 10a and 10b, Constitution of the Czech Republic [Ústava České Republiky]. 62  But cf I Pernice, ‘The Treaty of Lisbon: Multilevel Constitutionalism in Action’ (2009) 15 Columbia Journal of European Law 349. For an insightful critique of Pernice’s conception of a Verfassungsverbund, see M Jestaedt, ‘Der Europäische Verfassungsverbund – Verfassungstheoretischer Charme und rechtstheoretische Insuffizienz einer Unschärferelation’ in R Krause et al (eds), Recht der Wirtschaft und der Arbeit in Europa: Gedächtnisschrift für Wolfgang Blomeyer (Berlin, Duncker & Humblot, 2004) 637–74. 63   Case 6/64 Costa v ENEL [1964] ECR 585; Case 11/70 Internationale Handelsgesellschaft [1970] ECR 1125, para 3. 64   See Declaration (No 17) concerning primacy, annexed to the TEU and TFEU by the Treaty of Lisbon of 13 December 2007. 65   Art 50 TEU. 66   Eg, Art 48 TEU demands that any such proposal be submitted first to the Council, and that the European Parliament would have to participate either in the Convention to consider this change or

Systems and Institutional Pluralism  99 withdrawal, would presumably govern this collective decision, too. The converse point needs no elaboration here: a Member State’s legal system does not depend for its own existence on the recognition of that Member State’s legal system in any Treaty or in any legal system beyond its own. c.  Shared Purpose and the Grammar of Legitimacy The European Union and its Member States are not randomly joined systems of law. The lack of hierarchy among them does not lead to mutual destruction. And the mutual embedded openness is not some tragic mistake. Instead, this network of constitutional systems functions productively as a principled form of shared governance. Each Member State and the European Union can see the other as a legal system that is complementary to its own project of governance. These judgments are not imposed from above but emerge from a decentralized process of conflict and accommodation among the actors themselves. In the crucible of these conflicts, each system’s actors give in to the other system’s claim of authority only when they can see the other’s claim as being legal from the point of view of its own system as well. But this may require adjusting one’s own interpretation of one’s own system to accommodate the claim of the other. This accommodation, in turn, will be made only if the claim of the other can be seen both as legally possible and as substantively legitimate. As I have described in greater detail elsewhere, the substantive legitimacy claims in the conflicts of authority between the European Union and its Member States have fallen into three categories: voice, rights, and expertise.67 These three values (or categories of values) form a kind of grammar of legitimacy. Claims of legit­ imacy of the various systems are made and understood in these terms. Take expertise first. An original and persistent theme in the legitimation of European supranationalism has been the regulatory character of integration and the claim that technocratic sectoral integration can deliver a desired output more effectively than can either the Member States acting on their own or any broadly inclusive political process at the European level.68 Conversely, the European Union has built subsidiarity limitations into the exercise of its powers, limiting the invocation of its own powers to those for which the common institutions are really needed.69 And the various subsidiarity protocols provide for an increasingly ‘consent’ to the European Council’s decision not to call a Convention. Moreover, EU law would require that the heads of state or government act as members of the ‘European Council’, which itself is subject to some (albeit minimal) procedural requirements. See, eg, Art 235 TFEU. 67   D Halberstam, ‘Constitutional Heterarchy: The Centrality of Conflict in the European Union and the United States’ in JL Dunoff, JP Trachtman (eds), Ruling the World? Constitutionalism, International Law and Global Governance (Cambridge, Cambridge University Press, 2009) 326. 68  See, eg, D Dinan, Ever Closer Union (Boulder, Lynne Rienner Publishing, 1999); G Majone, ‘Regulatory Legitimacy’ in G Majone (ed), Regulating Europe (London, Routledge, 1996) 284–301. 69   Art 5 TFEU.

100  Daniel Halberstam institutionalized forum for contest and accommodation surrounding these claims of expertise.70 Next, a concern for rights protection was the focus of perhaps the most notorious standoff between the European Court of Justice and the Member States. In the wellknown Solange saga, the Member States refused to recognize the Union’s claim to authority unless and until the Union had developed an adequate system of rights protection. In response, the ECJ adjusted its claim of authority by ‘discovering’ a commensurate system of rights protection at the Union level of governance. After the Union had thereby developed a system of protection, Member States gave way to Union authority by adjusting their previously held understanding of rights protection. As long as the Union would generally protect rights within reasonable limits, the Member States would let the Union’s claim of authority stand. The reverse idea of the ECJ respecting the Member States’ claim of authority only as long as the latter protect rights adequately may be underway in the case law as well.71 Finally, even collective political decisions or the expression of a political will beyond the state (as in EU directives and regulations) can be seen as legitimate from within the project of governance that each participating state has set for itself. To command allegiance, supranational commitments to collective governance need not vindicate only the independently conceived interests of the participating state, as some scholars would have it.72 Member States need not control the political outcome at the European level of governance as a way to make sense of their respect of EU law. National constitutional commitments to Europe (whether in terms of constitutional provisions as in Germany or France, or constitutionally significant statutes as in the United Kingdom),73 suggest that European governance can be seen as an extension of the Member State’s own constitutional project of limited collective self-governance. Moreover, the European Union, for its part, has increasingly involved the European Parliament and Member State parliaments as a way to provide for additional voice-based legitimacy at the Union level of governance.

ii.  Systems Pluralism Beyond Europe? Can the idea of systems pluralism be extended to the international realm? There is little doubt about (some form of) multiplicity of legal systems, but we may have 70   For the latest iteration, see Protocol No 2 on the application of the principles of subsidiarity and proportionality, annexed to the TEU and the TFEU by the Treaty of Lisbon of 13 December 2007. For a discussion, see J-V Louis, ‘National Parliaments and the Principle of Subsidiarity – Legal Options and Practical Limits’ (2008) 4 European Constitutional Law Review 429. For an account of the evolution of the protocols, see K van Kersbergen and B Verbeek, ‘Subsidiarity as a Principle of Governance in the European Union’ (2004) 2 Comparative European Politics 142. 71   See, eg, Case 380/05 Centro Europa 7 Srl v Ministero delle Comunicazioni e Autorità per le Garanzie nelle Comunicazioni et al, [2008] ECR I‑349 (Opinion of Advocate General Maduro). 72   Andrew Moravcsik has been the principal proponent of this idea. See, eg, A Moravcsik, ‘Preferences and Power in the European Community: A Liberal Intergovernmentalist Approach’ (1993) 31 Journal of Common Market Studies 473. 73  See Art 23 of the Basic Law; Title XV of the Constitution of the French Republic; and the European Communities Act 1972 (c 68).

Systems and Institutional Pluralism  101 greater difficulty finding here the kind of mutual embedded openness and shared purpose that characterizes systems pluralism within the European Union. I will not attempt to give definitive answers to this question here. Instead, I shall sketch out some preliminary thoughts as well as the line of inquiry that needs to be pursued in greater detail to decide whether a given clash of systems in the context of global governance properly amounts to ‘systems pluralism’ as defined here or whether such a practice could perhaps lie in the future. a.  One system or more? Not much shall be said here about the multiplicity of national legal systems or the basic question about the existence of some kind of legal system at the level of international law. This is the old debate about monism versus dualism, which has run its course in a rather unhelpful manner over far too long a period of time.74 That national legal systems exist independently from one another would be disputed only by a kind of global Kelsenian monism75 or hard global constitutionalism76 that we shall not spell out in greater depth here. Suffice it to say for present purposes that even Kelsen himself ultimately rejected his earlier thesis that grounded all national legal systems in a single global Grundnorm.77 So, too, it would go beyond the scope of this Chapter to revisit in any depth the sceptical claim whether international law is law at all78 or whether international law amounts to a legal ‘system’.79 The latter of these questions is the more difficult of the two, but even here there are good reasons to believe that we can cautiously proceed for present purposes on the assumption that a ‘system’ of international law indeed exists. With the increasing density of international legal norms, the existence of norms that designate the validity of international law (for example, pacta sunt servanda, rules on the emergence of customary law, and rules on the sources of law and interpretive principles) as well as norms that are of higher ranking status than others (such as ius cogens and, perhaps, UN Article 103), international law – or at least large segments of international law – can be seen as a loosely coherent legal ‘system’.80 74   See A von Bogdandy, ‘Pluralism, Direct Effect, and the Ultimate Say: On the Relationship between International and Domestic Constitutional Law’ (2008) 6 International Journal of Constitutional Law 397. 75   See, eg, H Kelsen, Reine Rechtslehre (Wien, Österreichische Staatsdruckerei, 1960) 328–43. 76  See G de Búrca, ‘The Constitutional Challenge of New Governance in the European Union’ (2003) 28 European Law Review 814, 817. 77   See H Kelsen in Allgemeine Theorie der Normen (K Ringhofer and R Walter, eds, Wien, Manz, 1979) 169, 330; cf J Kammerhoffer, ‘Kelsen – Which Kelsen? A Reapplication of the Pure Theory to International Law’ 22 (2009) Leiden Journal of International Law 225. 78   See JL Goldsmith and EA Posner, n 3; J Austin, The Province of Jurisprudence Determined (1832) in WE Rumble (ed) (Cambridge, Cambridge University Press, 1995) 171; T Hobbes, ‘Leviathan’ (1651) in CB Macpherson (ed) (New York, Penguin, 1981) 189–217. 79   Hart, for instance, was cautiously sceptical about the claim that international law amounts to a legal system. See HLA Hart, The Concept of Law (Oxford, Clarendon Press, 1994) 236–37. 80   For a brief discussion of this question, see, eg, Y Shany, The Competing Jurisdictions of International Courts and Tribunals (Oxford, Oxford University Press, 2003) 86–94.

102  Daniel Halberstam There is even an incipient social component to this system in the sense that these norms do not merely serve as individual tools to bind individual states but are often considered as bearing significance for, and expressing concerns of, the ‘international community’. As Bruno Simma cautiously put it in his Hague Lectures: ‘[A] rising awareness of the common interests of the international community, a community that comprises not only states, but in the last instance all human beings, has begun to change the nature of international law.’81 This means that from peace and security to environmental regimes and human rights, there is a growing understanding that international law serves not only collections of governments but a broader common ‘humanity’.82 To pick just one example, with regard to ius cogens the argument can be made that it is not derived from a simple consent-based view of state practice.83 Nor is it applied arbitrarily from some outside (divine or natural) authority. Instead, it expresses the understanding that ‘non-compliance would shock the conscience of mankind’ and be ‘contrary to elementary considerations of humanity’.84 Ius cogens, then, claims to represent a collective judgment of the international community. It is our ius cogens that governs the world. The far more difficult question is whether there are multiple legal systems at the international level of governance or just one. This is sometimes discussed in terms of the existence of so-called ‘self-contained regimes’85 (following the PJCC’s decision in the Wimbledon case86 and the ICJ’s decision in the Tehran case)87 or, as the ILC has later called them, ‘special regimes’.88 The question – from the point of view of international law – is to what extent special regimes constitute lex specialis that precludes recourse to principles of general international law. This position ultimately sees special regimes and general international law as what we have discussed here in terms of synchronized systems, not competing systems. From the point of view of the ‘special system’ itself, however, the question may be presented not as one about lex specialis and the synchronization of systems but about the special system’s autonomy, that is, emancipation from the general rules of international law. In the strong sense, a special regime means ‘interrelated wholes of 81   B Simma, ‘From Bilateralism to Community Interests in International Law’ in Collected Courses of The Hague Academy of International Law (1994) 234, 250. 82   See, eg, RJ Dupuy, La communauté internationale entre le mythe et l’histoire (Economica, Unesco, 1986) 160, 180; S Gardbaum, ‘Human Rights as International Constitutional Rights’ (2008) 19 European Journal of International Law 749, 762. 83   M Koskenniemi, ‘The Pull of the Mainstream’ (1990) 88 Michigan Law Review 1946, 1954. 84   Ibid at 1952 (quoting Reservations to the Convention on the Preservation and Punishment of the Crime of Genocide [1951] ICJ Rep 15, 23 (Advisory Opinion of 28 May); Corfu Channel Case (UK v Albania) (Merits) [1949] ICJ Rep 4. 85   For a list of regimes that might be so considered, see, eg, M Noortmann, Enforcing International Law: From Self-Help to Self-Contained Regimes (Aldershot, Ashgate, 2005) 141. 86   Case of the SS Wimbledon Series A No 1 (1923) PCIJ Rep 25, at 23–24. 87   Case concerning the United States Diplomatic and Consular Staff in Tehran (United States of America v Iran) [1980] ICJ Rep 3, at 41, para 86. 88   Fragmentation of International Law: Difficulties Arising From the Diversification and Expansion of International Law (A/CN.4/L.682), para 123 ff, available at http://daccess-dds-ny.un.org/doc/ UNDOC/LTD/G06/610/77/PDF/G0661077.pdf?OpenElement.

Systems and Institutional Pluralism  103 primary and secondary rules, sometimes also referred to as “systems” or “subsystems” of rules that cover some particular problem differently from the way it would be covered under general law’.89 Such systems of rules may even regulate ‘whole fields of functional specialization’, as in the case of the ECHR, the WTO, or the EU.90 The extent of this separation is often hotly debated, as in the case of the WTO. The question is whether the WTO presents a special regime or, in more grand terms, whether it should be viewed as being separately ‘constitutionalized’.91 Some scholars have argued that all such subsystems retain their link to general international law, that is, that they simply function as special rules that must be exhausted before recourse can be had to the more general system of international law.92 At the other end of the spectrum, scholars have suggested that even customary international law and international treaty law ‘may not be normatively connected’.93 This would mean that international law would be ‘a number of different normative orders merely “held together” by an empirical classification’.94 The idea of pluralism as presented here does not aim at settling these disputes. For that, each of the arguments about monism, dualism, and systems proliferation would need to be addressed on their own terms. As a preliminary matter, however, we should take note that pluralism can add to this debate by suggesting that rival claims of legality among multiple legal systems can nonetheless yield a principled legal practice of conflict and accommodation. With regard to the relationship between domestic and international law, pluralism can thus serve to blunt some of the force of the local sceptic’s as well as the global enthusiast’s positions by suggesting a viable third way.95 With regard to the relationship between the EU (or other systems) and general international law, pluralism can serve to calm the fears of those who are concerned that by moving to a series of self-­ contained or even ‘constitutionalized’ regimes, we eliminate shared legal commitments, principled legal communication, and rivalry among the various systems. b.  Embedded Mutual Openness? Does the relationship between the European Union and its Member States present a special case of embedded mutual openness, or can we find similar relationships   Ibid at para 128.   Ibid.   Eg KA Nicolaidis and R Howse, ‘Legitimacy and Global Governance: Why a Constitution for the WTO is a Step Too Far?’ in R Porter et al (eds), Equity, Efficiency and Legitimacy: The Multilateral System at the Millennium (Washington DC, Brookings Institution Press, 2001). 92   See B Simma, ‘Self-Contained Regimes’ (1985) XVI Netherlands Yearbook of International Law 111, 128–9. Similarly, MacCormick argued for subordinating the plurality of jurisdictional claims within the European Union to international law as a kind of tie-breaker; see N MacCormick, Questioning Sovereignty (Oxford, Oxford University Press, 1999). 93   Kammerhofer, n 76 at 247. 94   Ibid. 95   D Halberstam, ‘Local, Global, and Plural Constitutionalism: Europe Meets the World’ in G de Búrca and JHH Weiler (eds), The Worlds of European Constitutionalism (forthcoming), available at http://ssrn.com/abstract=1521016. 89 90 91

104  Daniel Halberstam beyond the European Union as well? This question would need to be examined in detail with regard to each of the systems that are considered to be part of a constellation of pluralism. Two preliminary observations can nonetheless be made here as well: one regarding the relationship between domestic constitutional law and international law, and the other regarding the relationship between European Union law and general international law. Both suggest that the possibility of systems pluralism may extend beyond the special case of the relationship between the European Union and its Member States. For example, even the radical democratic tradition of national constitutionalism did not make as clean a break with its normative surroundings as the traditional story of constitutional progress might suggest.96 One could, of course, theoretically conceive of a radical democratic tradition that did so break with all other forms of legality. And yet, the American and French constitutional traditions that are held up as the model of modern constitutionalism recognize and deliberately invoke a commitment to legality beyond the state at the very moment of their inception. Perhaps unsurprising, this fusion of universalism and particularism is already inherent in Vattel. His discussion of a Nation’s choice of a Constitution notably appears in his magisterial treatise on law beyond the state, that is, ‘The Law of Nations or Principles of Natural Law’. Accordingly, he cautions: ‘The duties of a Nation towards itself are here considered mainly to determine what its conduct should be in that larger society which nature has established among the peoples of the world.’97 But Vattel’s legal concern for humanity both within and beyond the state was not merely the hobbyhorse of a global natural law enthusiast. The basic idea proved resonant within the radical democratic constitutional tradition that was soon to come. For instance, if we take the Declaration of Independence as representing the founding spirit of the American Revolution, we find universalizing aspirations in the various commitments to the idea of equality, natural rights, and a ‘decent respect for the opinions of mankind’.98 Here, the national polity is not an end in itself but a far more contingent entity. As Mark Tushnet has put it: If we see the Constitution and the Declaration working together, we would conclude that the people of the United States are constituted by our commitment to the realization of universal human rights, which when realized would render the community defined as ‘the people of the United States’ politically unimportant’.99

This is, as Tushnet adds laconically, ‘not an entirely unattractive selfunderstanding’.100 At a more functional level, from references to the ‘offenses against the law of nations’ to the inclusion of treaties in the supremacy clause and other provisions,   G Teubner, Recht als Autopoietisches System (Frankfurt am Main, Suhrkamp, 1989) at 30.   Vattel, n 13, Bk 1: Ch III, ¶ 29.   The Declaration of Independence (U.S. 1776). 99   M Tushnet, Taking the Constitution Away from the Courts (Princeton, Princeton University Press, 2000) 53. 100   Ibid. 96 97 98

Systems and Institutional Pluralism  105 the Constitution recognizes and incorporates various norms of international law.101 This recognition does not impose upon the United States a slavish adherence to international law. At the same time, however, it expressly acknowledges the separate existence of the international legal system. And it envisions the United States as an international legal actor and legal subject playing the role of a (supporting) participant in that other legal system. Similarly, the French revolutionaries grounded their constitutional aspirations in a Declaration of Rights that combines a right to national sovereignty with universal aspirations of justice beyond the state.102 Here, too, the radical democratic tradition combined the particularism of the nation with a universal concern for a global community of law and justice beyond the state. This latter feature of French constitutionalism continues into the present with commitments not only to Europe but to international governance beyond Europe as well. More broadly, especially in Europe after the Second World War, and again in the post-­communist tradition, this kind of embedded openness to both European as well as inter­ national forms of collective governance has proved to be not an opportunistic add-on but part of the core of the constitutional enterprise itself. As a result, national commitments to governance mechanisms beyond the state should be taken seriously as part of the project of limited collective self-­ governance. By engaging a legal system beyond the state, as in joining the European Union or signing the UN treaty, national systems may do so as constitutionally committed participants in that other system of law. To be sure, treaties and international agreements can still be the result of opportunistic quid pro quo bargains, and breach of a treaty – like breach of contract – is possible. But the value of the legal system beyond the state is often grounded in the national constitutional tradition itself and therefore carries normative weight for legal officials beyond that of a simple tool to be used or discarded at will. With regard to the European Union’s relationship with general international law, we can also see an entanglement of legal systems similar to that between the EU and its Member States. Claims of autonomy notwithstanding, the European Union’s normative links with international law still run deep. Its founding charters have all taken the form of international treaties, as have several intergovernmental instruments that are integrated into the functioning system of what we know as European Union law today. Moreover, the founding charters have always expressly acknowledged the importance of maintaining proper forms of ‘co-­ operation’ with international institutions such as the UN, the Council of Europe, and other international organizations.103 Along these lines, the Union’s treaty provisions have long recognized that Member States may have competing international legal obligations ‘for the purpose of maintaining peace and international security’104 which may cause Member States to take actions that ‘have the effect of   SH Cleveland, ‘Our International Constitution’ (2006) 31 Yale Journal of International Law 1, 13.   Déclaration des droits de l’Homme et du Citoyen (1789). 103   Art 220 TFEU. 104   Art 347 TFEU. 101 102

106  Daniel Halberstam distorting the conditions of competition in the internal market’.105 The Union’s founding treaties expressly proclaim that none of its treaty provisions shall affect the rights and obligations that any given Member State has incurred under international agreements entered into prior to that Member State’s joining the Union.106 It seems too easy to take these links as an invitation to integrate Union law with international law into a single structured whole. Union law is not merely lex specialis that applies (in accordance with international law) wherever it applies, with general international law taking over where Union law leaves off.107 The practice of the European Court of Justice has always been more ambitious than that. The ECJ’s ‘new legal order of international law’108 quickly dropped the reference to ‘international law’109 and has become an ‘autonomous legal order’110 ever since. Any doubt about the straightforward integration of international and European legal orders was recently laid to rest by the ECJ in the Kadi judgment. The Court reiterated there that international legal obligations cannot ‘prejudic[e] the principles of the EC Treaty’.111 Even former Article 307 EC (now 351 TFEU) cannot derogate from the Union’s fundamental commitments to ‘liberty, democracy, and respect for human rights and fundamental freedom’.112 As Advocate General Maduro put it: ‘[I]n the final analysis, the Community Courts determine the effect of international obligations within the Community legal order by reference to conditions set by Community law.’113 To be sure, the Union did not simply reject international legal norms, but it has not invariably embraced them either (for example, by giving them automatic direct applicability within the Union or by automatically bending to serve international legal obligations).114 Instead, the ECJ has come closer to occupying a ‘middle ground’115 of charitable openness, which has entailed such legal principles as a presumption that the Union intends to act in ways that are compatible with its international legal obligations. Whether this will ultimately lead to a principled   Art 348 TFEU.   Art 351 TFEU. 107   For an elaboration of the view of EU law as a kind of lex specialis, see B Simma and D Pulkowski, ‘Of Planets and the Universe: Self-Contained Regimes in International Law’ (2006) 17 European Journal of International Law 483. 108   Case 26/62 Van Gend and Loos v Netherlands Inland Revenue Administration [1963] ECR 1. 109   Case 90/63 Commission v Luxembourg and Belgium [1964] ECR 625. 110  Case 6/64 Flaminio Costa v ENEL [1964] ECR 585; 11/70 Internationale Handelsgesellschaft [1970] ECR 1125. 111  Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat International Foundation v Council and Commission [2008] ECR I-6351, para 285. 112   Ibid at para 303; Opinion of Advocate General Poiares Maduro on C-402/05 P (Kadi), para 31. 113   Ibid at para 23. 114   See, eg, A Nollenkaemper, ‘The Direct Effect of Public International Law’ in JM Prinssen and A Schrauwen (eds), Direct Effect: Rethinking a Classic of EC Legal Doctrine (Groningen, Europa Law Publishing, 2002) 157–80; J Wouters and D Van Eeckhoutte, ‘The Enforcement of Customary International Law Through EC Law’ ibid at 183. 115   R Schütze, ‘On “Middle Ground”: The European Community and Public International Law’ in S Muller and M Louth (eds), Highest Courts and the Internationalisation of Law: Challenges and Changes (The Hague, Hague Academic Press, 2009) 35–73. 105 106

Systems and Institutional Pluralism  107 practice of pluralism on the part of the European Court of Justice, however, has yet to be seen. Finally, and very briefly, international law reciprocates the structural openness to domestic legal systems, on the one hand, and even to the European Union, on the other. The United Nations, for example, depends in myriad ways on the cooperation of domestic systems and, by extension, on the European Union as well. If the European Union remains centrally reliant on, and open to, its Member States, then a fortiori the UN remains so as well. From the creation of Treaties to the formation of customary international law and the relative lack of an independent institutional architecture, the United Nations system – and international law more generally – remains structurally dependent on states. And as functional or regional systems such as the EU develop with an institutional apparatus of their own, this structural reliance on states extends to reliance on such non-state systems as well. At the UN level, as Eric Stein and I have suggested elsewhere, we might also be witnessing phenomena, such as the ‘constitutional absorption’ of legitimating principles, that we have seen at the EU level of governance in the past.116 c.  A Common Grammar of Legitimacy? If the lack of ordering between international law and domestic law or among various separate international and supranational regimes is to lead to a sustained form of governance through law, we need more than sporadic or happenstance mutual co-operation. Some minimally shared purpose or basic compatibility is necessary to ensure that the independent sites of governance with conflicting claims of authority and embedded mutual openness can interact positively over the long term here, too. Without some common substrate of shared principles, actors who are fundamentally committed to one system will view the demands of the other system as simply foreign and reject the formal act of linkage as a frolic or mistake. On the other hand, the common substrate of shared principles need not be so thick or substantively determined as to render the independence of the other system negligible and the multiplicity of systems imperceptible. For each system, the legal claim of the other can remain recognizably ‘other’ and yet retain some normative pull despite the conflict of authorities. Legality and legitimacy run together here. Joseph Raz has suggested, for example, that in the case of conflicting authorities, ‘the question whether a given authority’s power extends to exclude the authority of another is to be judged in the way we judge the legitimacy of its power on any matter.’117 Without subscribing to Raz’s service conception of authority, we can agree with him that when particular authorities conflict with one another, considerations of their relative 116   D Halberstam and E Stein, ‘The United Nations, the European Union, and the King of Sweden: Economic Sanctions and Individual Rights in a Plural World Order’ (2009) 46 Common Market Law Review 1, 14–18, 54–56. 117   J Raz, Between Authority and Interpretation: On the Theory of Law and Practical Reason (Oxford, Oxford University Press 2009) 143.

108  Daniel Halberstam legitimacy will enter the calculus of whom to obey. To be sure, authorities can also simply be ignored. But to the extent that a principled practice of mutual accommodation among independent yet interconnected systems can and does develop, it must build on some common ground by which actors who are primarily committed to one system can recognize the legal claims of the other system as legitimate. And to feel this normative pull of a foreign legal system, local actors must ultimately be able to see the foreign legal claim as vindicating the aspirations of their local system as well. Limiting one legal system by recognizing the legal claims of another can fit well into the project of limited collective self-governance through law. Within the United States, for example, constitutional limitations on component state powers in matters of interstate trade prevent destructive trade wars and thereby promote a more efficient system of economic self-governance even for the component states themselves. Similarly, the effectiveness of American self-governance was strengthened by the collective commitment via the Constitution to honour international treaties as the ‘supreme law of the land’, with direct effect (where appropriate, as John Marshall would later qualify)118 within the federal and state legal systems. In Europe, the post-war commitment to a supranational union that was or soon became embedded in Member State constitutions, formally limited Member State powers but only to gain a more stable, peaceful, and effective system of governance within Europe than the continent had ever seen. At the global level, trade regimes, for instance, allow state governments to combat interest group politics on the part of well-organized producers by shifting a good deal of the law and politics of international trade to a realm of collective decision beyond the state.119 From peace and human rights to trade and financial security, collective systems can augment the power of self-governance through self-binding.120 A mutual legal precommitment on the part of two systems that stand in a relationship of pluralism to one another can, by a similar logic, enhance – not reduce – the effectiveness of either system of governance. Call this global checks and balances, if you will. This is not a Panglossian vision that every limitation or every added veto point inevitably enhances power and effectiveness; sometimes limitations can indeed threaten the capacity to deliver effective policy output and to shift policies away from the status quo.121 But limitations should not be viewed as inherently diminishing effectiveness either; they can enhance the possibilities of governance as well.122 118   See, eg, Foster v Neilson, 27 U.S. 253 (1829); JJ Paust, ‘Self-Executing Treaties’ (1988) 82 American Journal of International Law 760. 119   E Benvenisti, ‘Exit and Voice in the Age of Globalization’ (1999) 98 Michigan Law Review 167. 120   See, eg, D Held, Democracy and the Global Order (Stanford, Stanford University Press, 1995) 21. 121   FW Scharpf, Governing in Europe (Oxford, Oxford University Press, 1999); I Shapiro, The State of Democratic Theory (Princeton, Princeton University Press, 2005) 109–10, 149; G Tsebelis, Veto Players: How Political Institutions Work (New York, Russel Sage Foundation, 2002); R Hirschl, Toward Juristocracy: The Origins and Consequences of the New Constitutionalism (Cambridge, Harvard University Press, 2004). 122   S Holmes, ‘Precommitment and the Paradox of Democracy’ in J Elster and R Slagstad (eds), Passions and Constraint: On the Theory of Liberal Democracy (Chicago, University of Chicago Press, 1995) 134.

Systems and Institutional Pluralism  109 If the shared purpose is the liberal enlightenment idea of limited collective selfgovernance through law, then for claims of authority from beyond the state (or beyond one specialized system such as the European Union) to be respected, they must be shown to further this project as well. Claims of authority from beyond the system must give voice to what is seen as a legitimate political will, respect or advance the rights of individuals and groups, and promote the effectiveness of governance. Unless the claim from the outside can be made in these terms, it will fall on deaf ears.

B.  Institutional Pluralism The kind of plurality of institutions relevant to pluralism arises when multiple actors carry out the same function within a single system, that is, when they make out claims to the same kind of authority within any given system of law. Conversely, pluralism does not come into play (at least not in the sense in which pluralism is used here) where the functions of multiple institutions or the kinds of authority they invoke remain distinct. For example, where execution and legislation are distinct activities within a given legal system, institutional pluralism (again, in the sense in which I use the term here) does not arise between executive and legislative actors. Where the functions of legislation, execution, and adjudication are imagined as being truly distinct, we may indeed have a multiplicity of institutions, but the institutions are simply doing different things. The main focus of Montesquieu’s idea of separation of powers was to protect liberty by limiting any given actor to performing one kind of function in the exercise of public powers. As Montesquieu put it: ‘When legislative power is united with executive power in a single person, or in a single body of the magistracy, there is no liberty.’123 The same is true if ‘the power of judging is not separate from legislative power and from executive power’.124 On the traditional model, then, legislation, execution, and adjudication are distinct functions to be carried out by different actors. ‘All would be lost,’ Montesquieu writes, ‘if the same man or the same body of principal men, either of nobles, or of the people, exercised these three powers: that of making the laws, that of executing public resolutions, and that of judging the crimes or the disputes of individuals.’125 As far as pluralism is concerned, Montesquieu may have set forth good prescriptions for liberal governance, but they amount to only half the story about multiple actors in modern liberal governance. Institutional pluralism picks up where the traditional understanding of the division of powers leaves off. Institutional pluralism focuses on the protection of liberty that inheres in granting multiple actors not different but the same kind of legal authority. The idea of 123   Montesquieu, ‘The Spirit of the Laws’ in AM Cohler, BC Miler and HS Stone (trans and ed) (Cambridge, Cambridge University Press, 1989) Book 11, Ch 6, 157. 124   Ibid. 125   Ibid.

110  Daniel Halberstam pluralism does not deny or replace the more traditional conception of separation of powers. Instead, pluralism can be seen as an overlay over traditional power divisions by focusing on a different aspect of the multiplicity or, to use Robert Cover’s word, ‘redundancy’ of actors.126

i.  Institutional Pluralism and the Centrality of Interpretation Pluralism focuses on the fact that multiple actors can at times carry out the same function within a single legal system. Taking the traditional separation of powers as our point of reference, this would mean that any one of the tasks of, say, legislation, execution, or adjudication can be carried out by more than one actor. The conceptual heart of pluralism, however, concerns the act of interpretation and, in particular, that of interpreting the secondary (or constitutional) rules of a legal system. There are many situations in which multiple actors, for example, execute the same law in the same territory regarding the same matters and the same individuals or targets. If multiple states acting at the same time would each take it upon themselves to enforce a particular UN Security Council resolution unilaterally (as the United States did with regard to Iraq),127 we would wind up with a multiplicity of executive actors. So, too, we have multiple executive actors when the State of Arizona takes it upon itself to enforce US immigration laws alongside federal agents.128 And the same situation exists when, in the European Union, national regulatory authorities, national competition authorities, and the European Commission all enforce some of the same antitrust laws.129 The chief concern in each of these situations is not the multiplicity of executive actors as such, but the danger that the interpretive decisions about the content of the law, that is, about what the law allows and demands, will differ from one actor to the other. After all, if executive actors in a given legal system proceed with a harmonized and co-ordinated understanding of the law, the multiplicity of actors does not raise particularly troubling problems. To the extent that multiple executive actors must engage in a strategically co-ordinated enterprise, a unified understanding of the law can set forth the relevant duties that each actor must obey.130 Similarly, many systems have multiple legislative institutions that enact legal norms independently from one another. Taking public international law as a single legal system, for example, scholars have raised concern about the multiplicity 126   RM Cover, ‘The Uses of Jurisdictional Redundancy: Interest, Ideology, and Innovation’ (1980) 22 William and Mary Law Review 639. 127   Authorization for the Use of Military Force against Iraq Resolution 2002 (Iraq War Resolution) as enforcement of UN Security Council Resolution 1441, Public Law 107–243. 128   Arizona Senate Bill 1070 (2010), discussed in Randal C Archibold, ‘Arizona Enacts Stringent Law on Immigration’ (New York Times, 23 April 2010) A1. 129   K Wissenbach, ‘Systemwechsel im europäischen Kartellrecht: Dezentralisierte Rechtsanwendung in transnationalen Wettbewerbsbeziehungen durch die VO 1/03’ (2005) 36 Beiträge zum Transnationalen Wirtschaftsrecht 1. 130   See, eg, EF McGarrell et al, ‘Project Safe Neighborhoods – A National Program to Reduce Gun Crime: Final Project Report’ (2009), available at http://www.ncjrs.gov/pdffiles1/nij/grants/226686.pdf.

Systems and Institutional Pluralism  111 of actors in terms of so-called ‘fragmentation’.131 From treaty making to the development of customary international law, the process of legislation in the inter­ national arena is notoriously decentralized with multiple actors engaged in legislation at the same time. Different treaties create competing legal regimes each of which may, in addition, stand in tension with certain demands of customary international law. In the European Union, scholars have raised a similar concern about the variety of instruments and procedures that different combinations of institutions can employ to create legal norms.132 Even the basic EU legal instruments, such as directives, regulations, and decisions are not always governed by a uniform procedure. And in domestic legal systems as well there is a variety of institutions, from local and regional legislatures to central legislatures and executive bureaucracies, all of whom may issue legal norms. In all of this, however, the chief concern is never the multiplicity of actors as such, but the potential effect on our ability to determine reliably the content of the law, that is, a concern about interpreting (or constructing) what the law is in light of the legislative output of these various actors. If component states legislate in the shadow of federal pre-emption, as they do in the United States, the existence of separate federal and state legislatures can often be unremarkable. By contrast, if under international law (taken as a single legal system) the legal priority among conflicting treaty regimes is unsettled, or under EU law the priority of conflicting European measures passed according to different legislative pro­ cedures is unclear, the multiplicity of legislative actors can turn out to be a challenge to legal determinacy. The challenge or absence thereof, in each case, depends therefore not on the multiplicity of actors as such. It depends on whether secondary law133 exists that sets forth a legal hierarchy of sources and norms and whether the interpretation of primary law against the background of this secondary law is shared among the various actors. This is usually where courts come in. We often charge courts with making sense of all this, especially where the content of the law is disputed. In the case of multiple executive or legislative actors, for instance, courts will apply rules of hierarchy and pre-emption as a way to determine the content of the law in light of the normative output of the various legislative institutions. On Hans Kelsen’s view, which by now has inspired the majority of post-war Constitutions throughout Europe along with many others around the world, we put a constitutional 131  See, eg, M Koskenniemi and P Leino, ‘Fragmentation of International Law? Postmodern Anxieties’ (2002) 15 Leiden Journal of International Law 553. 132  K Lenaerts and M Desomer, ‘Towards a Hierarchy of Legal Acts in the European Union? Simplification of Legal Instruments and Procedures’ (2005) 11 European Law Journal 744, 764. For a discussion of the same issue after passage of the Lisbon Treaty, see B de Witte, ‘Legal Instruments and Law Making in the Lisbon Treaty’ in S Griller and J Ziller (eds), The Lisbon Treaty: EU Constitutionalism without a Constitutional Treaty? (Vienna and New York, Springer, 2008) 79. 133   Unfortunately the terms ‘secondary’ and ‘primary’ mean opposite things in different scholarly contexts. By ‘secondary law’ I mean the legal rules that govern and order the production of law within a given system along the lines of HLA Hart’s ‘secondary rules’. What I am calling here secondary law would, in EU parlance, be part of ‘primary law’, ie Treaty law.

112  Daniel Halberstam court in charge of determining the meaning of the Constitution, which, in turn, orders the remaining subordinate forms of law within the system.134 The constitutional court, in Kelsen’s pure vision, is a single highest court that controls the meaning of the Constitution for all actors and participants within the system. A unified vision of primary law thus ultimately co-ordinates the activities of all. Even where there is an absence of a formal hierarchy of norms, this kind of central high court will – one way or another – order potentially conflicting norms. A central high court may do so by resorting to an analysis of mutually exclusive competences,135 the application of a last-in-time rule,136 consideration of the democratic pedigree of the various institutional actors,137 or some form of balancing the intrinsic significance of the various norms involved.138 Regardless of the doctrinal ordering mechanism, once such a privileged constitutional court has settled this matter of interpretation, the law is happily unified. But what if it isn’t? What if we have multiple claims of interpretive authority? If different actors claim the power to interpret the law, we cannot impose a uniform interpretation of that particular law (or set of laws) without denying the various claims of interpretive autonomy. This problem comes into plain view in public international law, for example, which lacks a central Kelsenian high court. The International Court of Justice has special jurisdiction to settle certain disputes among nations, but it does not control the interpretation of international law for all participants in the international legal system. From the United States Supreme Court to the International Criminal Tribunal for the Former Yugoslavia, fellow interpretive actors do not find themselves bound by the International Court of Justice’s interpretation of international law.139 Similarly, states have disagreed with the international human rights law decisions of the UN Human Rights 134   H Kelsen, Wesen und Entwicklung der Staatsgerichtsbarkeit, Veröffentlichungen der Vereinigung deutscher Staatsrechtslehrer (1929) at 30. 135   This is the case in Austria, for example, where conflicts between federal and component state law are not solved by recourse to a rule of federal supremacy but by weighing mutually conflicting claims to exclusive powers. See S Lebitsch-Buchsteiner, Die Bundesstaatliche Rücksichtnahmepflicht (Vienna, Braumüller, 2001). 136   This rule governs in the case of conflicting treaties and federal statutes in the United States. See, eg, Head Money Cases, 112 U.S. 580 (1884); cf JG Ku, ‘Treaties as Laws: A Defense of the Last-in-Time Rule for Treaties and Federal Statutes’ (2005) 80 Indiana Law Journal 319. 137   This may come into play in the European Union, where different combinations of legislative actors have been able to create conflicts among the same or different species of legislation. See, eg, Lenaerts, Desomer, n 131 at 744. 138   This kind of reasoning arises in Germany, for example, when potentially conflicting constitutional norms are read in light of one another to produce a coherent whole. See, eg, 1 BvR 1257/84 Herrnburger Bericht [1987] 77 BVerfGE 240; Case 1 BvR 1215/87 Nationalhymne [1990] 81 BVerfGE 298; see also K Hesse, Grundzüge des Verfassungsrechts in der Bundesrepublik Deutschland 20th edn (Heidelberg, CF Müller, 1999) Rn 72. 139   See the dispute between the ICJ and the International Tribunal for the former Yugoslavia on state responsibility: Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v Serbia and Montenegro) [1996] ICJ Rep 595, 1996; Prosecutor v Tadic (Jurisdiction) (1996) 3 Intl Human Rights Rep 578; see also E Cannizzaro, ‘Interconnecting International Jurisdictions: A Contribution from the Genocide Decision of the ICJ’ (2007) 1 European Journal of Legal Studies 1.

Systems and Institutional Pluralism  113 Committee.140 Even with the proliferation of adjudicative institutions – or perhaps in part because of it – we can often find a multitude of actors interpreting the same set of legal rules independently from one another. Some international lawyers lament this lack of a single authoritative interpreter and, sometimes selfservingly, protest the lack of deference that is given around the world to the interpretations of the International Court of Justice.141 The luring vision operating in the background of such complaints is that of domestic constitutional law in which a constitutional high court controls the legal interpretation for the system as a whole. As we shall see, however, this vision may be an unattainable, and perhaps even undesirable, ideal in any setting.

ii.  Madisonian Pluralism in the United States The United States Constitution does not grant the Supreme Court exclusive authority to determine constitutional meaning. From Andrew Jackson, Abraham Lincoln and Franklin D Roosevelt right through the current administration, Presidents have often asserted their independent authority – and indeed duty – to interpret the United States Constitution. To be sure, when Ronald Reagan’s Attorney General, Edwin Meese asserted this authority forcefully, he was met with a firestorm of protest (mostly from the political left). But the basic claim was a rather traditional – if largely forgotten – one: that each department, that is, the Executive, the Legislature, and the Judiciary, can determine the meaning of the Constitution for itself.142 a.  Multiple Interpretive Institutions The claim that each co-ordinate branch of the federal government has the incidental power to interpret the Constitution with ‘coequal status’ goes back to the American Founders.143 One might even say that this was Madison’s key innovation in the idea of checks and balances over that of Montesquieu’s functional separation of powers. As President Jefferson explained in a letter to Abigail Adams after refusing to enforce the notorious Alien and Sedition Acts and issuing pardons for sentences handed down during President Adams’s administration: You seem to think it devolved on the judges to decide on the validity of the sedition law. But nothing in the Constitution has given them a right to decide for the Executive, 140  See United Nations Human Rights Committee, General Comment 24, UN Doc CCPR/C/21/ Rev.1/Add. 6 (11 April 1994); Report of the Human Rights Committee, UN Doc A/51/40 117–19 (16 September 1996) (objections by France); Report of the Human Rights Committee, Vol 1, UN Doc A/50/40[Vol.1] (Supp) 126–34 (4 February 1996) (objections by the US and the UK); AT Guzman, ‘International Tribunals: A Rational Choice Analysis’ (2008) University of Pennsylvania Law Review 157, 171, 232 n 168. 141   See Koskenniemi and Leino, n 130. 142   S Levinson, ‘Looking at the Constitution: Could Meese Be Right This Time?’ (1986) 243 The Nation 689. 143   MS Paulsen, ‘The Most Dangerous Branch: Executive Power to Say What the Law Is’ (1994) 83 Georgetown Law Journal 217, 240.

114  Daniel Halberstam more than to the Executive to decide for them. Both magistracies are equally independent in the sphere of action assigned to them. The judges, believing the law constitutional, had a right to pass a sentence of fine and imprisonment; because the power was placed in their hands by the Constitution. But the executive, believing the law to be unconstitutional, was bound to remit the execution of it; because that power has been confided to them by the Constitution. That instrument meant that its co-ordinate branches should be checks on each other. But the opinion which gives to the judges the right to decide what laws are constitutional, and what not, not only for themselves in their own sphere of action, but for the legislature and executive also, in their spheres, would make the judiciary a despotic branch.144

American Presidents since Jefferson have consistently invoked this power to interpret the Constitution for themselves, even if this leads to the refusal to execute a Congressional statute or to disregard the Supreme Court’s interpretation of the law. From Jackson’s refusal to sign the Bank Bill145 to Lincoln’s first inaugural address146 and from President Roosevelt’s court packing plan147 to the regular signing statements on unconstitutional provisions in federal statutes,148 Presidents have asserted their independent authority of constitutional interpretation. Congress, too, has pushed ahead with its own interpretations of the Constitution from the attempted impeachment of Andrew Johnson for violating the Tenure in Office Act149 to the enactment of the Voting Rights Act150 and the Religious Freedom Restoration Act (RFRA).151 The court has not always reacted with favour to this idea, but there is some indication that this view of interpretive power finds some acceptance even here, too. To be sure, the court famously insisted in the school desegregation battle of the 1950s that ‘the interpretation of the Fourteenth Amendment enunciated by this Court . . . is the supreme law of the land.’152 And the court fought back in City of Boerne v Flores153 by striking down RFRA and claiming a monopoly on interpreting the Bill of Rights. But at other times, the court has taken a more sanguine perspective that implicitly approves of the co-ordinate branches’ independent authority to interpret the Constitution.

144   Letter from Thomas Jefferson to Mrs John Adams (11 September 1804) in AE Bergh (ed), The Writings of Thomas Jefferson (1905) 49, 50–51. 145   Paulsen, n 142. 146   Ibid at 262–64, 275, 334. 147   Ibid at 326, n 377. 148   Department of Justice, Office of Legal Counsel, ‘The Legal Significance of Presidential Signing Statements’ 17 U.S. Op. Off. Legal Counsel 131 (1993), available at http://www.justice.gov/olc/signing. htm. 149   HL Trefousse, Andrew Johnson: A Biography (New York, Norton, 1989) 311–29. 150   42 U.S.C. § 1973–1973aa-6. 151   42 U.S.C. § 2000bb–2000bb-4. In the Act’s introductory section, the Congress expressed its disagreement with the Supreme Court’s decision ‘in Employment Division v Smith, 494 U.S. 872 (1990) [which] virtually eliminated the requirement that the government justify burdens on religious exercise imposed by laws neutral toward religion’. 152   Cooper v Aaron, 358 U.S. 1, 18 (1958). 153   City of Boerne v Flores, 521 U.S. 507 (1997).

Systems and Institutional Pluralism  115 In Myers v United States,154 for example, the Supreme Court upheld the President’s unilateral dismissal of a Postmaster (an executive branch official), despite the fact that the dismissal violated a Congressional statute. Here, the court implicitly held that the President had no obligation to wait until the statute was declared unconstitutional by the court before disregarding its provisions. In Agostini v Felton,155 the court accepted the government’s petition to reopen an existing injunction and change Establishment Clause jurisprudence on sending public teachers into parochial schools. Here, the court implicitly held that the government was not bound to accept the court’s earlier interpretation of the Establishment Clause in the very same case. And the court may have followed Congress’s lead in the past when holding that voting is a fundamental right under the Constitution.156 Finally, it has treaded gingerly into the territory of reconsidering the constitutionality of the most onerous provisions of the Voting Rights Act, the continued effectiveness of which challenges the court’s current interpretation of the Constitution.157 If the court strikes down the Voting Rights Act in the coming year(s), it will be only after having given the political branches the chance to review, reaffirm, and possibly reshape their considered commitment to the law and its constitutionality.158 All this is perhaps as it should be. If the Constitution were what the court said the Constitution means, then the President and Congress would have no legal warrant to act contrary to the court’s current interpretation even in a simple effort to induce the court to change its views. We would be stuck with a particular constitutional meaning until the court, acting on its own, were to change its mind and somehow find a fortuitous vehicle for authoritatively announcing its new view. President Lincoln went even further. During the Civil War, he defied Chief Justice Taney’s order in a habeas corpus action to produce secessionist Lieutenant John Merryman, who was in federal custody at the time.159 Lincoln sent a deputy to court to argue that the President had suspended the writ, whereupon Taney, holding that only Congress could suspend habeas corpus, ordered the President to release Merryman. Lincoln refused. Subsequently explaining his actions in a   Myers v United States, 272 U.S. 52 (1926).   Agostini v Felton, 521 U.S. 203 (1997). 156  Compare Lassiter v Northampton County Board of Elections, 360 U.S. 45 (1959) (holding that North Carolina’s voter literacy test did not violate the 14th Amendment’s Equal Protection clause and stating that ‘[t]he States . . . have broad powers to determine the conditions under which the right of suffrage may be exercised’), with Harper v Virginia State Board of Elections, 383 U.S. 663 (1966) (holding that Virginia’s $1.50 poll tax violated the Equal Protection clause and stating that ‘[u]ndoubtedly, the right of suffrage is a fundamental matter in a free and democratic society’, so that ‘any alleged infringement of the right of citizens to vote must be carefully and meticulously scrutinized’). Cf B Ackerman and J Nou, ‘Canonizing the Civil Rights Revolution: The People and The Poll Tax’ (2009) 103 Northwestern University Law Review 63. 157  See Northwest Austin Mun. Utility Dist. No. One v Holder, 129 S. Ct. 2504 (2009). 158   ED Katz, ‘From Bush v Gore to NAMUDNO: A Response to Professor Amar’ (2009) 61 Florida Law Review 991. 159   Ex Parte Merryman, 17 F. Cas. 144 (1861). 154 155

116  Daniel Halberstam message to Congress, Lincoln asserted his good faith belief that the Constitution did give the President the power to suspend the writ in cases of rebellion and insurrection, at least until Congress could convene and make a legislative judgment on the matter. Noting that ‘it was not believed that any law had been violated,’ he implied there was no need to obey what the President saw as an unlawful order of the court.160 Michael Stokes Paulsen has referred to this as the ‘Merryman Power’. Europeans, in particular, may be wary of this and quickly think of Carl Schmitt and his treatise on the President as the ‘guardian of the Constitution’.161 Schmitt, too, took issue with the judicialization of the Constitution. He argued that the ultimate responsibility to vindicate constitutional norms lies with the President, whom Schmitt saw as the only true representative of the people as a whole. Hans Kelsen, in characteristic fashion, savagely and cogently destroyed that position. His chief objection was that Schmitt’s position violated the most fundamental legal postulate, ‘that nobody shall be judge in his own case’.162 But note that Lincoln’s position is different from Schmitt’s. Lincoln’s position comes down to this: the President is not the exclusive arbiter of constitutional meaning; but neither is the court. As Madison had already said in the Federalist Papers: ‘The several departments being perfectly co-ordinate by the terms of their common commission, [none] of them, it is evident, can pretend to an exclusive or superior right of settling the boundaries between their respective powers.’163 And so, Lincoln’s position, which built upon that of the American Founders, might be said to vindicate Kelsen’s fundamental postulate even better than placing a con­ stitutional court in charge of constitutional meaning for all. And that is what interpretive pluralism is all about. b.  Embedded Mutual Openness Interpretive pluralism depends on more than the existence of multiple claims to final interpretive authority. As in the case of a pluralist constellation of systems, a pluralist constellation of interpreters depends also on important connections among the various claims of authority. In the United States, the various actors with competing claims to final interpretive authority are not merely independent of one another but also structurally connected to one another in important ways. The President, the Congress, and the Supreme Court claim to be interpreting the 160   A Lincoln, ‘Message to Congress in Special Session (July 4, 1861)’, in RP Basler (ed), The Collected Works of Abraham Lincoln (New Brunswick, Rutgers University Press, 1953) 421, 429–30; see also Paulsen, n 142 at 279. This was not the first time that a President had defied the order of a court. President Jefferson, for example, refused to comply with a subpoena duces tecum in Aaron Burr’s treason trial and ordered James Madison not to appear in Marbury v Madison. See, eg, R Fallon, ‘Executive Power and the Political Constitution’ (2007) Utah Law Review 1, 8 (ultimately arguing against the legality of such action). 161   C Schmitt, Der Hüter der Verfassung (Tübingen, Mohr, 1931). 162   H Kelsen, ‘Wer soll der Hüter der Verfassung sein?’ (1930–31) Die Justiz 576, 577. 163   Madison, ‘The Federalist No 49’ in JE Cooke (ed) (Middletown, CT, Wesleyan University Press, 1961) 339.

Systems and Institutional Pluralism  117 rules and principles of a common legal system; their interpretive authority is grounded in this common legal system; and they claim interpretive authority with respect to a common audience. In short, the interpretive actors inhabit the same interpretive space. Each actor draws its own authority from a common set of legal rules and principles that are, themselves, subject to interpretation by all and for all. In that spirit, for example, even President Lincoln’s interpretation was offered as one that, in principle, the Congress and the Court could both have accepted as a proper interpretation of the US Constitution. Lincoln did not simply rest his case on the power of the President to disregard the Constitution in times of crisis. To be sure, Lincoln did famously say: ‘[A]re all the laws, but one, to go unexecuted, and the government itself go to pieces, lest that one be violated?’164 But even in this, he made a more limited argument grounded in law by invoking his oath to uphold the Constitution: ‘Even in such a case, would not the official oath be broken, if the government should be overthrown, when it was believed that dis­ regarding the single law, would tend to preserve it?’165 Perhaps most important, the President ultimately made his case by arguing that his actions were con­ stitutional in that they did not violate the Suspension Clause at all. A proper understanding of that Clause, Lincoln suggested, meant that the President can suspend the writ until Congress can consider the matter.166 The Merryman case ultimately ended in an interpretive stalemate, in which the court never officially accepted the President’s (and, implicitly, Congress’s) interpretation of the Constitution. But it does not follow from this that the Chief Justice was right and that the President simply violated the Constitution with the acquiescence of Congress. Nor does it follow that Lincoln was right. It just means that this time around, the President happens to have had the last word. As Laurence Tribe’s magisterial treatise put it in 1978: [T]he Constitution remains a fundamentally democratic document, open to competing interpretations limited only by the values which inform the Constitution’s provisions themselves, and by the complex political process that the Constitution creates – a process which on various occasions gives the Supreme Court, Congress, the President, or the states, the last word in constitutional debate.167

c.  A Grammar of Legitimacy The structural connection of the various actors imposes a legal obligation to interpret the common system in a way that can, in principle, be accepted by all participants. For this to work in practice, the various actors must be able to speak to one another productively. Whether the actors subscribe to a set of understandings to constitute an ‘interpretive community’,168 or a set of ‘common meanings’   Lincoln, n 159 at 430.   Ibid. 166   Ibid. 167   LH Tribe, American Constitutional Law (New York, Foundation Press, 1978) 33. 168   SE Fish, ‘Interpreting the Variorum’ (1976) 2 Critical Inquiry 465, 483–85. 164 165

118  Daniel Halberstam that underpin a shared practice,169 or simply an ‘ideology’,170 their claims about a preferred interpretation (and about their own institutional authority to put forth that interpretation) must be made in terms that others can understand. Moreover, there must be more than a mere coincidence of subjective beliefs on these matters. As Charles Taylor puts it when discussing common meanings, it is important that ‘this sharing be shared’.171 Common meanings – at least in the sense of some shared understandings about authority – would seem to be necessary if law is to be a form of shared activity or shared plan for action.172 Participants in this shared activity can still disagree about the substantive interpretation of the common framework and even about the method of interpretation itself, but they must share a commitment to the basic project of governance. That is, in the case of constitutional governance, they must share a commitment to the enterprise as being one of limited collective self-governance through law. Disagreements among the actors about interpreting the legal framework (and about their own institutional authority to do so) must therefore be put in terms of furthering this project, that is, of furthering the three basic values of constitutionalism: voice, rights, and expertise. Lincoln faced this task of overcoming rival interpretations of the Constitution to an extent that no other President has either before or since. Working within the Union to preserve the Union in the greatest moment of uncertainty, Lincoln repeatedly returned to first principles of constitutionalism to make sense of the American Constitution. In making out his claim of authority to call out the ‘war power’ of government to resist the rebellion, for example, Lincoln made clear: [T]he issue embraces more than the fate of these United States. It presents the whole family of man, the question, whether a constitutional republic, or a democracy – a government of the people, by the same people – can, or cannot, maintain its territorial integrity, against its own domestic foes. It presents the question, whether discontented individuals, too few in numbers to control administration, according to organic law, in any case, can always, upon the pretences made in this case, or on any other pretences, or arbitrarily, without any pretence, break up their Government, and thus practically put an end to free government upon the earth. It forces us to ask: ‘Is there, in all republics, this inherent, and fatal weakness?’ ‘Must a government, of necessity, be too strong for the liberties of its own people, or too weak to maintain its own existence?’173

Limited collective self-governance meant vindicating these three values: a government must ‘maintain its own existence’, preserve a ‘government of the people, by the same people’, and protect the ‘liberties of its own people’ all at the same time. Lincoln specifically grounded his claim to authority in the American people’s

169   C Taylor, Philosophy and the Human Sciences, Philosophical Papers 2 (Cambridge, Cambridge University Press, 1985) 34–40. 170   D Gauthier, ‘The Social Contract as Ideology’ (1977) 6 Philosophy and Public Affairs 130. 171   Taylor, n 168 at 39. 172   S J Shapiro, Legality (Cambridge, Harvard University Press, 2010). 173   A Lincoln, n 159 at 426.

Systems and Institutional Pluralism  119 ‘popular demand’174 as well as the ‘public necessity’175 to take action, all the while maintaining that his actions remained within the bounds of the Constitution. He also dealt at length with the claim of ‘right’ on the part of the secessionist states. Presenting his construction of constitutional history and meaning, Lincoln argued to Congress that the states’ rights claim was no more than the ‘sugar-coat[ing]’ of a ‘rebellion’ and attempted revolution.176 Their constitutional rights, as Lincoln had already carefully explained in his First Inaugural Address, had not been violated. To the contrary, the rights in peril were the rights of equality of individuals, which the confederate declaration of independence seemed to deny, as well as the right to a republican form of governance, which could no longer be assured after secession.177 His argument about the Presidential suspension of the writ taps into these principles by insisting on the legality of the suspension ‘until Congress could be called together; the very assembling of which might be prevented, as was intended in this case, by the rebellion’.178 If the President were unable to act in such a case, the entire project of constitutional governance would be doomed. Similarly, in his fireside chat to defend the court packing plan, FDR argued from within the Constitution and by appeal to the values of constitutionalism.179 FDR urged his listeners to ‘re-read’ the Constitution.180 He grounded his interpretive disagreement with the Supreme Court’s majority by appealing, for example, to the dissent of Chief Justice Hughes who, writing for four judges, had accused the majority of ‘a departure from sound principles’, by placing ‘an unwarranted limitation upon the commerce clause’.181 Appealing to the ‘high purposes’ of the framers,182 FDR, too, backed up his institutional resistance to the court’s majority by returning to the values of the constitutional enterprise itself. Here, FDR reminded listeners of the ‘mandate’183 he had received from having been elected for the third time in a landslide victory. He emphasized that the ‘American people’ ‘are in the driver’s seat’ expecting the three branches of government to come together on a common vision of the Constitution.184 And he came back repeatedly to the value of expertise. He explained that the Constitution was created in order to provide for ‘National Government with power enough to handle national problems’.185 The current justices, however, were thwarting the ability of the government to ‘protect us against catastrophe by meeting squarely our modern social and economic conditions’.186   Ibid at 429.   Ibid. 176   Ibid at 433. 177   Ibid. 178   Ibid at 431. 179   See FD Roosevelt, President of the U.S., ‘Fireside Chat on the Reorganization of the Judiciary’ (9 March 1937), transcript available in the Franklin D Roosevelt Presidential Library and Museum and at http://docs.fdrlibrary.marist.edu/030937.html. 180   Ibid. 181   Ibid (quoting RR Ret. Bd. v. Alton R.R. Co., 295 U.S. 330, 375 (Hughes CJ dissenting)). 182   Ibid. 183   Ibid. 184   Ibid. 185   Ibid. 186   Ibid. 174 175

120  Daniel Halberstam And so, the federal judiciary was in need of ‘a steady and continuing stream of new and younger blood’ especially to ‘bring to the decision of social and economic problems younger men who have had personal experience and contact with modern facts and circumstances under which average men have to live and work’.187 In short, the plan was to provide for justices who ‘understand . . . modern conditions’ and therefore will interpret the Constitution properly to allow for effective government. Lincoln and FDR provide perhaps the two most dramatic instances of an interpretive challenge to the judiciary. But we see mutual accommodation between the court and other interpretive actors from dramatic to far more sanguine times. As a routine matter, the political branches will obey the court’s rulings; but so, too, will the court frequently adjust its reading of the Constitution based on the interpretive authority of other actors. As Madison explained in a letter to then US Representative Edward Everett: ‘In the case of disputes between independent parts of the same Government, neither part being able to consummate its will, nor the Government to proceed without a concurrence of the parts, necessity brings about an accommodation.’188

iii.  Kelsen v Madison, or, the Hydraulics of Institutional Pluralism Is the unsettled nature of interpretive authority in the Madisonian arrangement yet another form of American exceptionalism or are there other reasons why we do not see a similar play of conflict and accommodation in Kelsenian systems? After all, if constitutionalism means limited government, then the idea of constitutionalism ought to be opposed, in principle, to monopolies of authority – even those held by Kelsenian constitutional courts. Perhaps surprising to some, even systems following the general Kelsenian mould may contain elements of contested hierarchy regarding the interpretation and application of the Constitution. For example, Germany follows the general idea of elevating its constitutional court to a privileged position in the domestic constitutional constellation. And yet, the German Grundgesetz grants the Bundestag and Bundesrat broad power to determine by simple law the structure and procedure of   Ibid.   See ‘Letter from James Madison to Edward Everett’ (August 1830), in Letters and other Writings of James Madison 98 (Philadelphia, JB Lippincott & Co, 1867). Madison distinguished this situation from that of a conflict between the central government and the states: ‘In disputes between a State government and the Government of the United States, the case is, practically as well as theoretically, different; each party possessing all the departments of an organized Government, Legislative, Executive, and Judiciary; and having each a physical force to support its pretensions’ (ibid at 98). It is not clear, however, why the latter situation should not also result in mutual accommodation, at least if neither side intends to follow through with its capacity to exercise physical force. Justice Story argues in support of Madison’s position here on the basis that an individual State does not represent the whole of the American people and is therefore an inappropriate interpretive authority of the US Constitution. This need not be an argument in favour of federal interpretive supremacy, but only an argument that the federal government’s voice-based claim to authority should often be accepted over that of the States. That this does not always hold true – especially when the question is one of the protection of rights – is perhaps evidenced best by Madison’s and Jefferson’s claims to component state interpretive authority in the Virginia and Kentucky Resolutions. 187 188

Systems and Institutional Pluralism  121 the court as well as the cases in which the decisions of the Constitutional Court shall have the force of law.189 After all, the Bundesverfassungsgericht itself was set up by a simple law two years after ratification of the Grundgesetz.190 And it is only a simple legislative act that grandly proclaims: ‘The decisions of the Federal Constitutional Court bind [all] federal and state constitutional bodies as well as all courts and authorities.’191 Although this law enjoys special status in Germany’s constitutional culture, its continued survival is, as a formal matter, subject to the ordinary legislative process.192 There is some room, then, even in a system such as Germany’s, that explicitly grants the formal power of judicial review to a constitutional court, to challenge the exclusivity (and finality) of the judiciary in matters of interpretation. Germany has featured several skirmishes over the years in which the political branches – and other actors – have indeed resisted or pushed back against the Bundesverfassungsgericht’s constitutional interpretation. Konrad Adenauer’s government notably threatened to pass legislation stripping the court of jurisdiction in early disputes surrounding the Treaties on the European Defense Community; the BVG encountered significant resistance to its first abortion decision; and the Bundestag pushed back hard against the BVG’s early decision limiting campaign finance contributions.193 The constitutional court won the first battle and adjusted its jurisprudence in the latter two. And yet, in Germany, these instances do not figure prominently in any standard account of German constitutional interpretation as analogous conflicts among the federal branches of government do in the United States. The overwhelming perception in Germany is that the Bundesverfassungsgericht is the only true guide to constitutional meaning. Why? Even though the potential for interpretive pluralism might ultimately exist in a Kelsenian as well as a Madisonian system, the unsettled nature of interpretative authority may be more salient in the latter given the more particular institutional and political dynamics involved. Put another way, institutional and political dynamics in the United States may invite inter-institutional confrontation to a degree that those in other systems do not.   Art 94(2) of the Basic Law.   K Niclaus, ‘Der Parlamentarische Rat und das Bundesverfassungsgericht’ in RC van Ooyen and MHW Möllers (eds), Das Bundsverfassungsgericht im Politischen System (Wiesbaden, VS Verlag, 2006) 117. 191   Federal Constitutional Court Law [Bundesverfassungsgerichtsgesetz], § 31(1) (‘Die Entscheidungen des Bundesverfassungsgerichts binden die Verfassungsorgane des Bundes und der Länder sowie alle Gerichte und Behörden’). Despite this law, a scholarly and judicial debate exists whether, for example, the Bundestag can introduce legislation identical to a law the Court has previously struck down as unconstitutional. See A Heusch, ‘Commentary on § 31, Margin no 64’ in DC Umbach, T Clemens, F-W Dollinger (eds), in Bundesverfassungsgerichtsgesetz: Mitarbeiterkommentar und Handbuch (Heidelberg, CF Müller, 2005). 192   The sole exception seems to be that if, in a case of emergency, a joint legislative committee were to exercise the full powers of the Bundestag and Bundesrat, that committee could not alter the Federal Constitutional Court Act absent the consent of the Federal Constitutional Court. See Art 115g of the Basic Law. 193   G Vanberg, The Politics of Constitutional Review in Germany (Cambridge, Cambridge University Press 2005) 67–77, 126–28, 143–66; see also C Landfried, Parteifinanzen und Politische Macht (BadenBaden, Nomos, 1994, 2nd edn) 78–84. 189 190

122  Daniel Halberstam Consider, first, the enormous difficulty of amending the United States Constitution.194 This creates immense pressure on the practice of interpretation given that only the act of constitutional interpretation or ‘construction’195 can ultimately preserve the document as a viable framework for modern governance. All constitutional systems must adapt to new realities, from the need for administrative agencies and the demands of an expanding market to the problems posed by technological innovation and the emergence of environmental pollution as a distinct object of governmental concern. Many other systems can – and regularly do – adapt by altering their foundational document, even if only to experiment with a new approach to an old problem.196 The United States, by contrast, cannot. The inordinate difficulty of changing our Constitution has meant that in the United States adaptation comes, if at all, via change in constitutional interpretation, not in constitutional text. By the same token, once the Supreme Court has spoken on a question of constitutional interpretation, the political branches cannot easily ‘overturn’ the court by formal amendment. Having set the stakes of constitutional adjudication high, the United States appoints judges in a process that reflects the American winner-take-all system of political elections and then renders judges fiercely independent by granting them life tenure. The appointment process gives the President – a single political actor – overriding power to select a given Supreme Court justice. Senate confirmation has done little to blunt the force of Presidential nomination. Even the emergence of modern confirmation battles have not changed matters significantly, as Presidents have become savvy enough to nominate candidates who lack a written record that could derail the candidacy. Also, each appointment is considered a one shot event. There is little logrolling among political groups and across nominees for different vacancies. To be sure, at some point pressure will mount to appoint a woman or a person of a particular racial, religious, or ethnic background. But in general, there is no broadly representative political body that ensures a broad spectrum of political or interpretive representation among the various judges on the court. Life tenure then augments the significance of the President’s choice by allowing the Administration to embed its preferred nominees on the Court for the long haul. As a result, with average tenure on the Court

194   Compared to the US demand of a proposal by a two-thirds majority of both Houses of Congress or two-thirds majority of state legislatures followed by adoption of the amendment by either threequarters of state legislatures or three-quarters of state ratifying conventions, Germany’s amendment procedure of a two-thirds vote in both houses of the legislature is a relatively small hurdle to overcome. Compare Art V of the Constitution of the United States of America with Art 79 of the Basic Law. 195  KE Whittington, Constitutional Construction: Divided Powers and Constitutional Meaning (Cambridge, Harvard University Press, 1999). 196   An example is Germany’s federalism reforms of 1994, 2006, and 2009. See, eg, A Benz, ‘From Joint Decision Traps to Over-Regulated Federalism: Adverse Effects of a Successful Constitutional Reform’ (2008) 17 German Politics 440; AB Gunlicks, ‘German Federalism and Recent Reform Efforts’ (2005) 6 German Law Journal 1283; D Heinz, ‘Federal Reform II in Germany’ (2010) 2 Perspectives on Federalism; C Moore et al, ‘German Federalism in Transition?’ (2008) 17 German Politics 393.

Systems and Institutional Pluralism  123 topping 25 years,197 the judges on the Supreme Court do not necessarily represent any general interpretive consensus among political actors – let alone among the people. Nor do they provide the American people with a sense that the views of the general public are broadly represented on the Court. Germany, by contrast, elects its constitutional court judges by the votes cast within a proportional system of political elections and then grants these judges only a limited tenure of 12 years subject, in addition, to a mandatory retirement age.198 Each house of parliament elects half of the constitutional court’s members by a supermajority vote. Both the nomination as well as the election of German constitutional court justices is a politically inclusive process. This is not to suggest that the German nomination process is democratically laudable; much of the nomination process happens behind closed doors. But in terms of the political dynamics of representation, the Judicial Selections Committee of the Bundestag199 and the plenary vote in the Bundesrat200 reflect a broad-based compromise among the various parties represented in the two houses of parliament. In Germany, the appointments committees over time and across candidates generally respect the idea of proportional representation of the various political parties among the justices on the court. Moreover, limited tenure has ensured the regular turnover of constitutional court judges in Germany and thereby further supported the broad-based political control of the make-up of the court. Without sacrificing judicial independence in any individual case, it also means that judges remain conscious of the need to return to their former careers upon their ‘retirement’ from the court. The legal requirement of selecting six judges from the judiciary201 and the tradition of selecting some members from the middle echelons of politics and a large number of judges from academia means that the court has a strong communicative link to the world beyond its hallowed chambers. Legal academia is the most forceful interlocutor in this regard. By law, the only profession a constitutional court judge may exercise alongside her judicial office is that of law professor at a German public university.202 This professional connection is further fostered by the fact that German law professors – the former, current, and future colleagues of the overwhelming number of justices on the court – regularly appear before the court to represent the full spectrum of litigants from individual claimants and officials to the federal parliament and the German government itself. 197   SG Calabresi, J Lindgren, ‘Term Limits for the Supreme Court: Life Tenure Reconsidered’ (2006) 29 Harvard Journal of Law and Public Policy 770, 771. 198  For a brief review in English, see D Kommers, The Constitutional Jurisprudence of the Federal Republic of Germany (New York, Duke University Press, 1997), 22; Vanberg, n 192 at 81–86. For a short description in German, see GM Sierck, C Sinnokrot, Die Wahl von Richtern des Bundesverfassungsgerichts (Deutscher Bundestag Wissenschaftliche Dienste), available at http://www.bundestag.de/dokumente/ analysen/2006/Die_Wahl_von_Richtern_des_Bundesverfassungsgerichts.pdf For a brief political analysis, see C Landfried, Bundesverfassungsgericht und Gesetzgeber 2nd edn (Baden-Baden, Nomos, 1996) 15–22. 199   Art 6 of the Federal Constitutional Court Law, cf Umbach, n 190 at 209. 200   Art 7 of the Federal Constitutional Court Law. 201   Art 2(3) of the Federal Constitutional Court Law. 202   Art 3 of the Federal Constitutional Court Law.

124  Daniel Halberstam All this renders US Supreme Court justices less connected than their German Constitutional Court counterparts to the relevant political mainstream over time. It also renders US Supreme Court justices taken as a group less reflective of broad coalitions across the political spectrum than their colleagues on the German Constitutional Court. Even when the court happens to render decisions by margins that roughly reflect the distribution of public opinion, there is no strong sense that the court as a collective entity is structurally or systematically representative of this spectrum of belief. In sum, whereas structural elements of interpretive pluralism exist even in a system like Germany’s, the institutional dynamics surrounding court composition and constitutional amendment contribute to the fact that pitched battles of inter-institutional conflict are less likely to be fought there as compared to the United States. Other factors may contribute to this dynamic as well, such as a willingness to neutralize politics by deferring to the judiciary in Germany to an extent that may not match American democratic sensibilities.203 Either way, the apparent American exceptionalism may be less a matter of exceptional ground rules regarding interpretive pluralism in the United States than a consequence of inter-institutional dynamics, that is, the hydraulics of constitutional change, composition of the judiciary, and political culture.

C.  Summary and Outlook The pluralist alternative questions two traditional assumptions of constitutionalism. First, it questions the view that a constitutional system is always entirely closed to legal claims from beyond the system. Second, it questions the view that a constitutional system necessarily establishes an absolute institutional hierarchy within the system. Pluralism suggests that constitutional systems – both traditional national systems and more loosely ‘constitutional’ systems at the supra­ national or international level – may be structurally open to multiple claims of legal authority. This discussion highlighted two analytically distinct forms of multiplicity. The first is a multiplicity of legal systems that stand in a pluralist relation to one another. The second is a multiplicity of interpretive institutions that stand in a pluralist relation to one another within a single legal system. As we have seen, even though these two forms of pluralism are analytically distinct, they seem to have certain characteristics in common. Systems pluralism and institutional pluralism are marked by a similar constellation of multiplicity, connectedness, and shared understanding of legitimacy. They are both marked by a similar practice of conflict and accommodation over time. And they can both be understood as a principled legal practice, as opposed to a practice of fortuitous or happenstance political opportunism. 203   See, eg, Landfried, n 197 at 86–7; V Curran, ‘Racism’s Past and Law’s Future’ (2004) 28 Vermont Law Review 683.

Systems and Institutional Pluralism  125 Finally, systems pluralism and institutional pluralism often come together. And when they do, each tends to mask the presence of the other. In the United States, for example, during the nullification and interposition debates of the early nineteenth century, the two claims of pluralism were frequently joined and perhaps even conflated. Claims of autonomous state-based interpretation of the Constitution were at times joined with a compact theory of the US Constitution that ultimately challenged the unification of federal and state law into a single system with federal legal supremacy. Similarly, in the European Union, Member State resistance to European law has often taken the form of autonomous Member State interpretation of Union law that, in turn, was ultimately based on the autonomy of the Member State’s own legal system. Conversely, state accommodation of legal claims from a legal system beyond the state has taken the form of adjustments in interpretation of the demands of the state’s own legal system. Accordingly, it is useful to understand both phenomena together even when (seemingly) confronted with only one or the other.

IV. Conclusion Global governance is not a threat but an opportunity. Not the kind of opportunity generally talked about, that is, to reach a more just society by tempering power and politics through reason and law on a global scale. That is an opportunity we should pursue as well. But even more important, and crucial to reaping these real gains from global governance, is a different kind of opportunity – a conceptual one. We should take the challenge of global governance to rethink our traditional understanding of law and, in particular, of constitutional law as the consolidation and settlement of authority. The challenge of law beyond the state helps highlight the shortcomings of the traditional state-based vision of law and constitutional order. Spurred by the problem of understanding law beyond the state, we may come to a better understanding of law within the state as well. Traditional theories of constitutional law as hierarchy have neglected the central lack of settlement that inheres in the practice of constitutionalism as laid out in this chapter. Indeed, mutual openness and lack of settlement – the “constitutional heterarchy”204 described in this chapter – is as it perhaps should be, if we are to live up to the constitutional idea of limited collective self-governance through law. Only when leaving the baggage of the traditional narrative of closure and consolidation behind can we truly bring out the productive capacity of constitutionalism in national, supranational, and global governance.

  Cf Halberstam, n 67.

204

6 Multilevel Constitutionalism and Constitutional Pluralism Querelle Allemande or Querelle d’Allemand? FRANZ C MAYER AND MATTIAS WENDEL*

I. Introduction ‘Multilevel constitutionalism’ was introduced in 19981 as the English translation of Verfassungsverbund, a concept suggested by Ingolf Pernice in 1995 in order to describe and explain the constitutional specificity of European integration.2 The term was supposed to articulately contrast the approach of the German Federal Constitutional Court, which had construed the EU as a Staatenverbund (literal translation: a compound of states) in its 1993 Maastricht decision. 15 years ago, insisting on a constitutional law approach to European integration was, if not a provocation, a challenge on any account.3 Soon this approach became part of an intense and widespread debate on whether Europe can actually have a Constitution at all, and whether the use of constitutionalism really makes sense in the EU context. In French, the expression querelle d’Allemand means a quarrel about nothing, a fight without sense, as opposed to the expression querelle allemande, which simply means a German quarrel.4 Could it be that the controversy about the concept of * The authors wish to thank Imke Stanik, University of Bielefeld, for her help in finalising this text. 1  I Pernice, ‘Constitutional Law Implications for a State Participating in a Process of Regional Integration. German Constitution and Multilevel Constitutionalism’ in E Riedel (ed), German Reports on Public Law Presented to the XV International Congress on Comparative Law (Baden-Baden, Nomos, 1998); I Pernice, ‘Multilevel Constitutionalism and the Treaty of Amsterdam: European ConstitutionMaking Revisited?’ (1999) 36 Common Market Law Review 703. 2   I Pernice, ‘Bestandssicherung der Verfassungen: Verfassungsrechtliche Mechanismen zur Wahrung der Verfassungsordnung’ in R Bieber and P Widmer (eds), The European Constitutional Area (Zürich, Schulthess, 1995) 261 ff. 3   A constitutional law reading of European integration was suggested much earlier, though; just see E Stein, ‘Lawyers, Judges, and the Making of a Transnational Constitution’ (1981) 75 American Journal of International Law 1. 4   The expression has nothing to do with Germans, though, but seems to refer to a family with the name of Alleman.

128  Franz C Mayer and Mattias Wendel multilevel constitutionalism is – or even was – just a domestic problem of the Germans, something related to a particular constitutional psychology, in short: a querelle allemande? Has the debate even turned into a querelle d’Allemand, in particular after the failure of the Treaty establishing a Constitution for Europe in 2005? Or is the conceptual value of multilevel constitutionalism rooted deeper and therefore still – perhaps even more – relevant today? What about its relationship with the increasingly popular paradigm of constitutional pluralism? In the following, we will briefly revisit the concept of multilevel constitutionalism and take a closer look at the way it was received (I). Secondly, we will explore its relationship with constitutional pluralism (II), before finally coming back to the initial questions which will be addressed in the light of the Lisbon decision of the German Constitutional Court of 2009 (III). Let us emphasise that this contribution is not on the concept of multilevel constitutionalism as such. Considering that the present volume is on constitutional pluralism, we will not address such misunderstandings of or objections to multilevel constitutionalism which are not directly linked to the core issue of constitutional pluralism.

II.  Multilevel Constitutionalism A.  The Concept The starting point of multilevel constitutionalism was a simple question: could it be that the concept of Constitution is neither linked nor limited to the state?5 Obviously, this was not a debate on the formal use of the term ‘Constitution’. After all, the founding document of the World Health Organization is called the ‘Constitution of the WHO’. Rather it was about a substantive concept of Constitution. The historical starting point of the debate is the German Constitutional Court’s Maastricht decision of 19936 and its focus on state and statehood, captured by the term Staatenverbund, compound of states, which the Court uses to characterise the European construct.7 The term had previously been coined by the judge rapporteur of the Maastricht decision, Paul Kirchhof.8 In 1995, at a conference in 5   Which is, of course, just one possible conceptualisation of European constitutionalism. See, for other concepts beyond those discussed in this volume, the principle approach of Bogdandy in A von Bogdandy and J Bast (eds), Principles of European Constitutional Law 2nd edn (Oxford, Hart, 2009), or for the concept of a ‘dual constitution’, T Öhlinger, ‘Die Verfassung im Schmelztiegel der europäischen Integration: Österreichs neue Doppelverfassung’ in Öhlinger, Verfassungsfragen einer Mitgliedschaft zur Europäischen Union (Wien, Springer, 1999) 165 ff; P Pernthaler, ‘Die neue Doppelverfassung Österreichs’ in H Haller et al (eds), Staat und Recht (Wien, Springer, 1997); see in this context the Austrian Constitutional Court, Decision of 9 December 1999, Reverse discrimination, (2001) Europäische Zeitschrift für Wirtschaftsrecht 219, 222; for more references on the debate see A Peters, Elemente einer Theorie der Verfassung Europas (Berlin, Duncker and Humblot, 2001). 6   Case 2 BvR 2134, 2159/92 Maastricht [1993] BVerfGE 89, 155. 7   Ibid, at 181, 190 and headnote 8. 8  P Kirchhof, ‘Contribution to the Discussion’ (1991) Beiheft 1 Europarecht 47; Kirchhof, ‘Der deutsche Staat im Prozeß der europäischen Integration’ in P Kirchhof and J Isensee (eds), Handbuch

Multilevel and Constitutional Pluralism  129 Lausanne,9 Ingolf Pernice suggested to conceptualise the European Union as Verfassungsverbund, a system of composite Constitutions. Obviously, the terminology chosen was supposed to contrast the Constitutional Court’s terminology, and also its focus: Verfassung, Constitution, instead of Staat, state. In the translations, the contrast gets lost. Staatenverbund and Verfassungsverbund are examples of the difficulty of translating new terms and concepts: the closest literal translation of Verfassungsverbund is composite (or compound) of Constitutions, while the substance of the concept is better captured by ‘multilevel constitutionalism’. Whereas multilevel constitutionalism10 carries at least the idea behind the concept, the English translation of Staatenverbund, ‘compound of states’,11 remains clumsy. Nuances between Verbund (compound or composite)12 and Verband (association) pale into obscurity. Paul Kirchhof actually initially suggested Staatenverband.13 The latter concedes at least an original form of its own to the European construct; the Verbund leaves this autonomy in the dark and deliberately emphasises that the Member States are the principal entities in European integration – the masters of the treaties. The French translation, constitution composée,14 is more literal. But it is still different from the German term, as the German term does not say whether there is one compound Constitution or 27 of them. The constitu­ tion composée is clearer on that. The debate around the Verfassungsverbund is to a large extent a debate on the question whether there has to be a state first, and then there may be a Constitution, or whether state and Constitution are independent concepts. According to the Verfassungsverbund approach, a Constitution is both foundation and limitation of public authority over those who are subjects of and thus subject to that authority. There can be no legitimate public authority beyond the one created by the Constitution. In other words: public authority is created by the des Staatsrechts. Volume VII (Heidelberg, CF Müller, 1993) § 183, para 69. B Kahl, ‘Europäische Union: Bundesstaat – Staatenbund – Staatenverbund?’ (1994) 33 Der Staat 241, 245 tells us that in terms of content, system and telos, the Staatenverbund corresponds to the Staatenbund (confederation). U Di Fabio, Das Recht offener Staaten (Tübingen, Mohr Siebeck, 1998) 140 ff, describes the Staatenverbund as a transitory model in times of change. 9  Pernice, n 2, 261 ff and ‘Die Dritte Gewalt im europäischen Verfassungsverbund’ (1996) Europarecht 27 ff. See also Pernice, ‘Europäisches und nationales Verfassungsrecht’ in (2001) 60 Veröffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer 163 ff, with further references in Das Verhältnis europäischer zu nationalen Gerichten im europäischen Verfassungsverbund (Berlin, de Gruyter, 2006). For a recent account see Pernice, ‘The Treaty of Lisbon: Multilevel constitutionalism in action’ (2009) 15 Columbia Journal of European Law 349 ff. 10   See n 1. 11  G Wegen and C Kuner, ‘Germany: Federal Constitutional Court Decision Concerning the Maastricht Treaty’ (1994) 33 International Legal Materials 338; note in this context L Siedentop, Democracy in Europe (London, Penguin Books, 2000) 27, though, pointing to Madison in the Federalist papers, speaking of the ‘compound republic’. 12  On Verbund as a concept of order E Schmidt-Aßmann, ‘Europäische Verwaltung zwischen Kooperation und Hierarchie’ in Cremer et al (eds), Festschrift Steinberger (Berlin, Springer, 2002) 1380 ff. 13   P Kirchhof, ‘Deutsches Verfassungsrecht und Europäisches Gemeinschaftsrecht’ (1991) Beiheft 1 Europarecht 11, 16. 14   I Pernice and FC Mayer, ‘De la constitution composée de l’Europe’ (2000) Revue Trimestrielle de Droit Européen 623.

130  Franz C Mayer and Mattias Wendel Constitution, it does not exist outside the Constitution.15 Based on this assumption, it is possible to conceptualise the European Union in ‘non-statal’ terms: the core elements of European constitutionalism are, on the one hand, the public authorities that exist both at European and Member State level and, on the other hand, the individuals subject to those authorities. The respective frameworks for these national and supranational public authorities can be understood as two distinct, albeit related, constitutional levels which are complementary rather than in a hierarchical relationship, forming a single constitutional system of multilevel con­ stitutionalism (Verfassungsverbund).16 It is this system as a whole that is ‘the European Constitution’, which thus turns out to be a composite Constitution. ‘The’ European Constitution is depicted as a complementary structure of national and European Constitutions. This concept of a European Constitution is not affected by the failure of the Constitutional Treaty and the fact that the Lisbon Treaty refrains from using the constitutional rhetoric,17 because it does not rely on this rhetoric. A European Constitution already existed before the Constitutional Treaty. It arises from both national and European constitutional levels. European and national constitutional law form two levels of a ‘unitary system’ in terms of substance, function and institutions. One may identify the individual as the deeper basis of validity for this European composite of Constitutions, to whom the public powers allocated to both the national and European component Constitutions may be traced back.18 This is different from classical international law constructs, and this is also how the concept of primacy may be justified.19 On this reading, the principle of primacy in application (Anwendungsvorrang) does not imply a hierarchy of norms in the sense of the general hierarchical superiority – supremacy – or inferiority of either European or national (constitutional) law:20 ‘The hallmark of the Verfassungsverbund is its non-hierarchic structure.’21 It is here that it turns out to be useful that the image of distinct levels is not necessarily linked to ‘superordination’, that is, supervision and subordination. 15   Cf A Arndt, ‘Umwelt und Recht’ (1963) Neue Juristische Wochenschrift 24, 25: ‘In einer Demokratie gibt es an Staat nicht mehr, als seine Verfassung zum Entstehen bringt’ (‘In a democracy, there is no more to a State than established by the constitution’); see also P Häberle, Verfassungslehre als Kulturwissenschaft 2nd edn (Berlin, Duncker and Humblot, 1998) 620. 16   See I Pernice et al, ‘Renewing the European Social Contract. Can Europe Have a Constitution?’ (2001) 12 King’s College Law Journal 60; for the level metaphor, see FC Mayer, Kompetenzüberschreitung und Letztentscheidung (München, CH Beck, 2000) 36. 17   See on the elimination of the constitutional rhetoric FC Mayer, ‘Die Rückkehr der europäischen Verfassung?’ (2007) 67 Zeitschrift für ausländisches öffentliches Recht und Völkerrecht 1141. 18   In this sense I Pernice, ‘Die Europäische Verfassung’ in Cremer et al (eds), Festschrift Steinberger (Berlin, Springer, 2002) 1319, 1324; see also Pernice et al, n 16. 19   This cannot be explored in more detail at this point. 20   This may be the difference between supremacy and primacy. On hierarchies see R Bieber and I Salomé, ‘Hierarchy of Norms in European Law’ (1996) 33 Common Market Law Review 907, 912; see on this Spanish Constitutional Court, Declaration of 13 December 2004, 1/2004, [2005] 3 BOE 5. 21  Pernice, Europäisches und nationales Verfassungsrecht, n 9, 185, emphasising the difference between Verbund (compound or composite) and Verband (association). Critical of this M Nettesheim, ‘German Report for the XX. FIDE Conference 2002’ (2004) Beiheft 2 Europarecht 7.

Multilevel and Constitutional Pluralism  131 Levels may also be understood as platforms that may be at equal height in one case, at different heights in another, or even circling freely around each other. Public power is not primarily defined by the monopoly of power, the traditional concept used inter alia to define elements of sovereignty,22 but rather by the mere decision-making power (leaving aside the question of enforcement capacity), typically expressed in the specific form of norm- or law-making capacity. The decision-making power represents a subset of the elements that characterise the traditional concept of state and public power: the monopoly of force plus exclusive law-making powers.23 Levels in the present context are decision-making levels.24 Decision in this context is a cipher for decision-making operating under the rule of law, that is, determined by and organised according to law.25

B.  The Reception of the Concept It is probably not an exaggeration to claim that multilevel constitutionalism became one of the most debated concepts designed to capture the legal and institutional peculiarities of the European legal space.26 The reactions range from critical to affirmative. There are a number of aspects of the Verfassungsverbund approach that may be challenged. The Verfassungsverbund approach’s emphasis on the individual has been highlighted as rather problematic by, inter alia, Ulrich Preuss.27 Matthias Jestaedt deconstructs the concept from the perspective and with the tools of a Kelsenian positivist28 – without offering an alternative for conceptualising the European construct, though. This is also true for René Barents in this volume. And of course, the critics do query the decoupling of the concept of Constitution from the concept of ‘state’.29 22   The monopoly on the (legitimate) use of force as the basis for state and public authority structures is emphasised, inter alia, by M Weber, Wirtschaft und Gesellschaft 5th edn (Tübingen, Mohr, 1985) 835 ff. 23   A von Bogdandy, ‘Supranationale Union als neuer Herrschaftstypus’ (1993) Integration 210, 215. 24   For a similar concept see FW Scharpf, Optionen des Föderalismus in Deutschland und Europa (Frankfurt am Main, Campus, 1994) 25 and 29, and R Mayntz, ‘Föderalismus und die Gesellschaft der Gegenwart’ (1990) 115 Archiv des Öffentlichen Rechts 232. 25   See in this context the concept of the state suggested by H Heller, Staatslehre (Leiden, Sijthoff, 1934) 228 ff, according to whom the state is an organised entity of effective decision making (‘organisi­ erte Entscheidungs- und Wirkungseinheit’). 26   See A von Bogdandy and J Bast, ‘Introduction’ in von Bogdandy and Bast (eds), Principles of European Constitutional Law 2nd edn (Hart, Oxford 2009). 27  U Preuss, ‘Contribution to the Discussion’ (2001) 60 Veröffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer 384 ff. 28  M Jestaedt, ‘Der Europäische Verfassungsverbund – Verfassungstheoretischer Charme und rechtstheoretischer Insuffizienz einer Unschärferelation’ in R Krause et al (eds), Recht der Wirtschaft und der Arbeit in Europa (Berlin, Duncker and Humblot, 2004) 638, 662 and 664 ff. 29  See eg J Isensee, ‘Staat und Verfassung’ in J Isensee and P Kirchhof (eds), Handbuch des Staatsrechts. Volume 2 (Heidelberg, CF Müller, 2004) § 15 paras 1, 3 ff, and references in C Grabenwarter, ‘Europäisches und nationales Verfassungsrecht’ (2001) 60 Veröffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer 290, 292.

132  Franz C Mayer and Mattias Wendel There is at least one critique that touches the core of the concept. Although the term and the concept of Verfassungsverbund (multilevel constitutionalism) seek to avoid the trappings of traditional terms and concepts, it encounters objections that point to the danger of potential ‘blueprint traps’.30 ‘Blueprint traps’ means that the use of terms and concepts, developed for the nation state, can lead to confusion in a European integration context, as the EU is not a state. ‘Constitution’, as in ‘multilevel constitutionalism’, however, is precisely one of these traditional terms and concepts. If one wanted to go beyond the traditional typologies in an even more principled way, one could consider using the term ‘multilevel system’.31 In spite of the critique, the term and the concept have been quite successful. Success in this context means that a term and a concept are taken up, debated and developed further.32 But Verfassungsverbund and multilevel constitutionalism are also examples of how terms and concepts develop a life of their own – not always the way the inventors of a concept expect. Consider that although conceived in 1995 as an alternative and counter-concept to the Staatenverbund of the German Constitutional Court in the Maastricht decision, the notion is increasingly used detached from its original context by merging Staatenverbund and Verfassungsverbund and claiming that the EU is both:33 Staatenverbund and Verfassungsverbund at the same time.

III.  Multilevel Constitutionalism and Constitutional Pluralism One of the main reasons for the relative success of the concept of multilevel constitutionalism might be its epistemic and explanatory value. As Andreas Voßkuhle, the vice-president of the German Federal Constitutional Court, put it, the concept of 30   For the term ‘Blaupausenfalle’ see GF Schuppert, ‘Anforderungen an eine Europäische Verfassung’ in H-D Klingenmann and F Neidhardt (eds), Zur Zukunft der Demokratie (Berlin, Edition Sigma, 2000) 207, 216 and 226. See also GF Schuppert, ‘Contribution to the Discussion’ (2001) 60 Veröffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer 352. 31   The comparative law context which is inherent in European integration speaks in favour of referring to an analytical concept as neutral as possible. The variety of legal and constitutional concepts in Europe arising out of differences in language and legal culture (as may easily be illustrated by the different understandings of state, federalism, sovereignty and Constitution), necessitates an enormous amount of conceptual and terminological clarification before one uses these terms and notions in the EU context. 32   See eg L Bourgorgue-Larsen, ‘La démocratie au sein de l’Union européenne. De la “constitution composée” à la “démocratie composée”’ in H Bauer and C Calliess (eds), Verfassungsprinzipien in Europa. Constitutional Principles in Europe. Principes Constitutionnels en Europe (Athens, Sakkoulas, 2008) 83 ff. See also the references in Pernice (Lisbon), n 9, fn 2. The German Constitutional Court remains unconvinced, though, and explicitly sticks to the Staatenverbund in Case 2 BvE 2/08, judgment of 30 June 2009, Lisbon, English translation http://www.bverfg.de/entscheidungen/ es20090630_2bve000208en.html. 33   See eg PM Huber, ‘Europäisches und nationales Verfassungsrecht’ (2001) 60 Veröffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer 194 ff and C Calliess (ed), Verfassungswandel im europäischen Staaten- und Verfassungsverbund (Tübingen, Mohr Siebeck, 2007).

Multilevel and Constitutional Pluralism  133 Verbund can be helpful in describing the functionality of a complex multilevel system without relying on oversimplified terms such as hierarchy or subordination. According to Voßkuhle, the Verbund concept draws on the more adequate categories of unity-difference-plurality as well as separation-­interaction-interdependence, thus taking into account both the idea of respective autonomy as well as the capacity of joint and uniform action.34 This leads directly to the question of how multilevel constitutionalism relates to the concept of constitutional pluralism. The many faces of constitutional pluralism are discussed by other contributions to this volume. In the present context, a short summary of its basic assumptions shall suffice in order to highlight similarities and possible differences compared to multilevel constitutionalism. Constitutional pluralism, in the words of Neil Walker, recognises that the European order . . . has developed beyond the traditional confines of international law and now makes its own independent constitutional claims, and that these claims exist alongside the continuing claims of states. The relationship between the orders, that is to say, is now horizontal rather than vertical – heterarchical rather than hierarchical.35

If one agrees that these elements constitute basic assumptions of constitutional pluralism within the EU context,36 then one also agrees on significant similarities of constitutional pluralism and multilevel constitutionalism as far as their theoretical foundation is concerned.

A.  Formally Separated Legal Orders Multilevel constitutionalism acknowledges that national and supranational legal orders form two formally separated legal orders or two different centres of legislation.37 This formal separation is already indicated by the differing scope of application of national and supranational law. Although overlaps exist in various ways, the supranational constitutional level exceeds by definition the normative frameworks of the Member States. Supranational public authority transcends the national dimension and is, therefore, not to be confounded with a mosaic of fragmented state competences.38 This may be evidenced by the existence of genuine 34  A Voßkuhle, ‘Der europäische Verfassungsgerichtsverbund’ (2010) Neue Zeitschrift für Verwaltungsrecht 1. 35   N Walker, ‘The Idea of Constitutional Pluralism’ (2002) 65 Modern Law Review 317, 337. 36   See also N MacCormick, ‘Beyond the Sovereign State’ (1993) 56 Modern Law Review 1 ff; Peters, n 5 at 269 ff; NW Barber, ‘Legal Pluralism and the European Union’ (2006) 12 European Law Journal 306 ff; M Poiares Maduro, ‘Contrapunctual Law: Europe’s Constitutional Pluralism in Action’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 511 ff; M Kumm, ‘The Jurisprudence of Constitutional Conflict: Constitutional Supremacy in Europe before and after the Constitutional Treaty’ (2005) 11 European Law Journal 262, 281 ff; F Giorgi and N Triart, ‘National Judges, Community Judges: Invitation to a Journey through the Looking-glass’ (2008) 14 European Law Journal 693, 709 ff; J Baquero Cruz, ‘The Legacy of the Maastricht-Urteil and the Pluralist Movement’ (2008) 14 European Law Journal 389, 391. 37   See eg Pernice, n 9 at 374. 38   C Tomuschat, in Bonner Kommentar zum Grundgesetz 42nd edn (Hamburg, CF Müller, 1981) Art 24 para 15.

134  Franz C Mayer and Mattias Wendel supranational competences that cannot, per definitionem, exist at national level. A recent example is the new competence to promote the interconnection of (transnational) energy networks which can, by definition, only exist at a supranational level.39 In short, the conferral of competences on a supranational entity is not a ‘zero-sum game’. Both national and supranational constitutional levels provide for their own rules of rule production or ‘secondary rules’.40 The formal separation of the national and supranational legal orders is also acknowledged by the jurisprudence. As early as 1962 the ECJ stated in its Bosch judgment that national and supranational law constitute two ‘separate and distinct legal orders’.41 In principle, national constitutional courts agree on the formal separation, even if they construe the supranational legal order as being ‘derived’ from the Member States. The jurisprudence of the German Federal Constitutional Court is an illustrative example in this respect. Contrary to its early case law of the 1960s which seemed to have shared the ECJ’s view on the autonomy of Community law,42 the Federal Constitutional Court has, during the last decades, established a well settled case law which has been recently confirmed by its Lisbon judgment. According to the Court, ‘the foundation and the limit of the applicability of European Union law in the Federal Republic of Germany is the [normative] order to apply the law which is contained in the Act Approving the Treaty of Lisbon’.43 In other words, national law constitutes the normative basis or ‘bridge’ for the applicability of EU law within the German legal space.44 Nevertheless, while the Federal judges construe supranational law as a ‘derived order’, they still acknow­ ledge its formal distinctness from national law.45 Further examples illustrating the general acknowledgement of a formal separ­ ation can be found in the jurisprudence of the constitutional courts of Italy,46   Art 194(1)(d) and Art 194(2) TFEU.   HLA Hart, The Concept of Law 2nd edn (Oxford, Oxford University Press, 1994) 94 ff. The theoretical category of secondary rules in the sense of Hart means rules that specify the ‘ways in which the primary rules [that are concerned with the actions of individuals] may be conclusively ascertained, introduced, eliminated, varied, and the fact of their violation conclusively determined’, ibid 94, and must not be confounded with secondary EU law (regulations, directives and decisions). 41   Case 13/61 De Geus [1962] ECR 45, 99. 42   Case 1 BvR 248/63, 216/67, EWG-Verordnungen [1967] BVerfGE 22, 293 at 296; Case 2 BvR 255/69 Milchpulver [1971] BVerfGE 31, 145 at 173 ff; Case BvL 52/71 Solange I [1974] BVerfGE 37, 271 at 277 ff. 43   Lisbon, n 32, para 343. For previous judgments affirming the necessity of a normative fundament in national law see Case 2 BvR 499/74, 1042/75 Rückwirkende Verordnungen [1977] BVerfGE 45, 142 at 169; Case 2 BvL 6/77 Vielleicht-Beschluss [1979] BVerfGE 52, 187 at 199; Case 2 BvR 197/83 Solange II [1986] BVerfGE 73, 339 at 367, 375; Case 2 BvR 687/85 Kloppenburg [1987] BVerfGE 75, 223 at 244; Case 2 BvL 12, 13/88, 2 BvR 1436/87 Absatzfonds [1990] BVerfGE 82, 159 at 193; Case 1 BvR 1025/82, 1 BvL 16/83, 10/91 Nachtarbeitsverbot [1992] BVerfGE 85, 191 at 204; Maastricht, n 6, [1993] BVerfGE 89, 155 at 190. 44   Kirchhof (Deutsches Verfassungsrecht), n 13; Kirchhof (Der deutsche Staat) n 8, § 183, para 65; P Kirchhof, ‘The Legal Structure of the European Union as a Union of States’ in A von Bogdandy and J Bast (eds), Principles of European Constitutional Law 2nd edn (Oxford, Hart, 2009) 743 ff. 45   Lisbon, n 32, para 343, para 301, 231. 46   Italian Constitutional Court, 183/1973 Frontini v Amministrazione delle finanze dello Stato [1974] GU 2; 170/1984 Granital v Ministero delle finanze [1984] GU 169. 39 40

Multilevel and Constitutional Pluralism  135 Spain47 or Hungary.48 Even the few Member States that, with all precaution, may still be regarded as ‘classically monistic’ countries – as, in particular, the Netherlands – consider national and supranational law to be two formally separated parts of a monist system.49

B.  Multiple, Independent and Incommensurable Claims of Constitutional Autonomy As formal separation can even be construed within the monist theory, the core assumption of constitutional pluralism as well as of multilevel constitutionalism is the existence of multiple, independent and incommensurable claims of constitu­ tional autonomy.50 In terms of multilevel constitutionalism, public authority is constituted auto­ nomously at the national as well as at the supranational level. These claims of constitutional autonomy exist alongside each other without being forced into a hierarchical relationship. Thus, on the one hand, the question of ultimate authority is left open from a legal point of view. Sovereignty is conceptually abandoned or at least said to have transformed into shared or pooled sovereignty.51 On the other hand, multilevel constitutionalism assumes the possibility of direct collisions between norms as regards their applicability – not their validity. This non-hierarchical and horizontal approach associates multilevel con­ stitutionalism with legal and sociological approaches, which conceptualise the relationship of different orders or regimes in terms of the network theory,52 the conflict of laws,53 the systems theory54 or the paradigms of fragmentation55 and ‘regime collisions’.56 47   Spanish Constitutional Court, n 20. See F Castillo de la Torre, ‘Tribunal Constitucional (Spanish Constitutional Court), Opinion 1/2004 of 13 December 2004 on the Treaty Establishing a Constitution for Europe’ (2005) 42 Common Market Law Review 1169 ff. 48  Constitutional Court of Hungary, decision of 22 June 1998, 30/1998 (VI 25) AB, Europe Agreement, [1998] MK 1998/55. 49   See R Wessel and W Van de Griendt, ‘Offene Staatlichkeit’ in A von Bogdandy and PM Huber (eds), Ius Publicum Europaeum. Volume II (Heidelberg, CF Müller, 2008) § 19 Niederlande, para 13. 50  See also Peters, n 5, 268 ff. For the different meanings of the term autonomy see ibid, and R Barents, The Autonomy of Community Law (The Hague, Kluwer, 2004) 245 ff. In the Lisbon decision (n 32), the German Constitutional Court displays an understanding of autonomy as a derived status, opposed to sovereignty. 51   See recently Pernice (Lisbon), n 9, 372 ff. 52   K-H Ladeur, ‘Towards a Legal Theory of Supranationality – The Viability of the Network Concept’ (1997) 3 European Law Journal 33, 48. 53   C Joerges, ‘(Re-)Conceptualizing the Supremacy of European Law: A Plea for a Supranational Conflict of Laws’ in B Kohler-Koch and B Rittberger (eds), Debating the Democratic Legitimacy of the European Union (Lanham, Rowman and Littlefield, 2007) 311 ff. 54   N Luhmann, Das Recht der Gesellschaft (Frankfurt-am-Main, Suhrkamp, 1995) 440 ff. 55   M Koskenniemi and P Leino, ‘Fragmentation of International Law? Postmodern Anxieties’ (2002) 15 Leiden Journal of International Law 553 ff. See also M Koskenniemi, From Apology to Utopia (Cambridge, Cambridge University Press, 2005) 390 and 609 ff. 56   A Fischer-Lescano and G Teubner, ‘Regime-Collisions: The Vain Search for Legal Unity in the Fragmentation of Global Law’ (2004) 25 Michigan Journal of International Law 999 ff.

136  Franz C Mayer and Mattias Wendel It also contrasts multilevel constitutionalism with unitarian approaches based on a logical monism in the sense of Hans Kelsen and his proponents.57 As is well known, Kelsen’s Pure Theory of Law offers two theoretically equal possibilities of construing the relationship between national and (public) international law: monism with supremacy of international law and monism with supremacy of national law.58 If this was applied to the European legal space, either national law would have to be derived in its legal validity from EU law or EU law would have to be derived in the same way from national law. According to Kelsen, the choice between the two opposing views would not be a legal, but a mere political one. In other words, there would be no pluralism in terms of normative sources or constitutional claims, but a mere pluralism of interpretation.59 However, neither of the two monist alternatives is fully convincing. First, national law can hardly be seen as a derivative of EU law. EU law contains no rule of rule production for national law,60 nor is there a ‘genetic EU footprint’ on national law, which, by the way, is a condition not even fulfilled within federal states, if one agrees that the federal units are able to constitute genuine public power.61 Not even the ECJ claims that national law is being derived from EU law. Therefore, the ECJ’s concept of primacy is only a primacy in application.62 Second, one can also hardly construe supranational law as being derived from national law. Indeed, at first glance it seems that monism with supremacy of national law corresponds widely to the claims of national constitutional courts to ultimate constitutional supremacy. However, national constitutional courts – if at all – base only the domestic applicability of EU law on the normative fundament of national law. The normative existence of supranational law per se, that is, irrespective of its applicability within the internal sphere of the Member States, cannot be captured from a pure national perspective. As already mentioned, supranational authority per definitionem goes beyond a simple addition of national authorities. The monistic theory, requiring a genetic derivation of law in a hierarchically structured legal pyramid (Stufenbau),63 thus cannot capture the genuine supranational character of European public authority.64 In addition, it 57   H Kelsen, Reine Rechtslehre 2nd edn (Wien, Verlag Franz Deuticke, 1960) 328 ff. See especially Jestaedt, n 28, 637 ff. 58   Kelsen, n 57, 328 ff. 59   Not ‘Rechtsquellenpluralismus’, but ‘Rechtsdeutungspluralismus’. See Jestaedt, n 28, 666 ff. 60   This does not, of course, exclude affections deriving from the legal obligations of EU membership. In some cases, these obligations may even necessitate the change of a national Constitution, Case C-285/98 Kreil [2000] I-69. But EU law cannot, by itself, change national (constitutional) law. 61   See 2 BvG 1/51 Südweststaat [1951] BVerfGE 1, 14 at 34 and Case 2 BvH 1, 2/82, 2 BvR 233/82 Startbahn West [1982] BVerfGE 60, 175 at 207. 62   The ECJ also made clear in its judgment in Joined Cases C-10/97 to C-22/97 IN.CO.GE.’90 and Others [1998] ECR I-6307, para 21 that it ‘cannot . . . be inferred from the judgment in Simmenthal that the incompatibility with Community law of a subsequently adopted rule of national law has the effect of rendering that rule of national law non-existent. Faced with such a situation, the national court is, however, obliged to disapply that rule . . .’. 63   AJ Merkl, Die Lehre von der Rechtskraft, entwickelt aus dem Rechtsbegriff (Leipzig, Deuticke, 1923) 207 ff. 64   See also W Schroeder, Das Gemeinschaftsrechtssystem (Tübingen, Mohr Siebeck, 2002) 230.

Multilevel and Constitutional Pluralism  137 presupposes a false idea of structural equality of both levels. Through the national lens, EU law can be perceived only in an incomplete and distorted manner. However, multilevel and pluralist approaches are not – e contrario – dualist systems either. Multilevel constitutionalism and legal pluralism are not simply legal dualism renamed. A fundamental difference from the classic concept of dualism already results from the fact that the latter does not accept an overlap of public international law and national law, but a tangency at the most.65 Even if one took the perspective of moderate dualism in the sense of ‘dualisme, mais interpénétration’,66 there would still be a difference from multilevel constitutionalism. If an act of public international law is to produce legal effects within the domestic legal order, dualism, also in its moderate form, demands a legal basis in domestic law. Either international law is being transcribed into national law by transformation, or domestic law must order the internal application of international law as international law.67 In short, dualism requires legal order A to provide an internal legal basis for law deriving from legal order B, whereas pluralism – accepting the parallelism of incommensurable claims of constitutional autonomy – does not. In terms of legal pluralism, law deriving from B is not only valid, but also applicable in its own terms, even within the (territorial) scope of legal order A. The assumption of a heterarchical relationship of different constitutional claims and levels allows us not only to construe both union and national law in their normative autopoiesis or self-logic. It also takes into account the peculiar state of openness and abeyance that characterises the question of ultimate authority within the European legal space. Against this background, it is not surprising that a comparative survey reveals clearly that the importance of the classic dichotomy between monism and dualism is increasingly vanishing from the perspective of national constitutional law. Ambiguous statements, like the following, are typical: ‘One could hardly classify the Greek legal system as either monistic or dualistic. It is rather a mixture.’68 Similar conclusions can be found for traditionally dualist states such as Italy, Denmark, Ireland, Sweden, Finland, the Czech and Slovak Republics as well as for classically monist countries such as France, Belgium, Austria or Poland (perhaps with the exception of the Netherlands).69 The shift from the classic monism-­ dualism dichotomy towards a pluralist approach is expressed concisely by the   H Triepel, Völkerrecht und Landesrecht (Leipzig, Hirschfeld, 1899) 111.   L Cavaré, Le droit international public positif. Volume I 3rd edn (Paris, Pedone, 1967) 155. 67   See W Rudolf, Völkerrecht und deutsches Recht (Tübingen, Mohr, 1967) 150 ff. 68   K Chryssogonos, ‘The European Union and the Greek Constitutional Order’ in AE Kellermann et al (eds), EU-Enlargement – The Constitutional Impact at EU and National Level (The Hague, TMC Asser Press, 2001). See also M Pogačnik et al, ‘Slovenia’ in AE Kellermann et al (eds), The Impact of EU Accession on the Legal Orders of New Member States and (Pre-) Candidate Countries (TMC Asser Press, The Hague 2006) for the legal situation in Slovenia: ‘Some case law exists that unfortunately cannot confirm either a monist or dualist perception of the relationship between the international and the Slovenian legal order (. . .) Nevertheless, according to commentators the difference between the monist and the dualist doctrine is less and less clear’ (at 181). 69   This cannot be dealt with in detail here. For a comprehensive comparative analysis see M Wendel, Permeabilität im europäischen Verfassungsrecht (Berlin 2010) Chapter 1, forthcoming. 65 66

138  Franz C Mayer and Mattias Wendel Polish Constitutional Tribunal in its judgment of 11 May 2005 concerning the Treaty of Accession: The concept and model of European law created a new situation, wherein, within each Member State, autonomous legal orders co-exist and are simultaneously operative. Their interaction may not be completely described by the traditional concepts of monism and dualism regarding the relationship between domestic law and international law. The existence of the relative autonomy of both, national and Community, legal orders in no way signifies an absence of interaction between them. Furthermore, it does not exclude the possibility of a collision between regulations of Community law and the Constitution.70

C.  Legal Pluralism, but Functional Coherence The Polish Constitutional Tribunal does not only underline the co-existence of national and supranational law and its conceptual detachment from the monismdualism dichotomy. The Tribunal also points to the potential risk that is inherent in a pluralist conception.71 Whereas the historical development of sovereignty has revealed the problems of a centripetal accumulation of power, the risk which is inherent to a pluralist conception lies in the possibility of centrifugal distraction and legal incoherence. In this context it becomes clear that multilevel constitutionalism goes beyond a purely descriptive approach and entails also a normative dimension. Multilevel constitutionalism acknowledges that national and supranational legal orders are (and will remain) formally separated, but claims that they are – by means of various instruments – a mutually interwoven unity especially in a functional sense.72 As regards the nature of these instruments, it must be differentiated between legal, institutional, political, economic and social mechanisms. The following considerations aim specifically at the legal dimension. One of the most significant instruments aiming at an adequate degree of legal coherence and mutual adjustment is the mutual legal permeability of national and supranational law. In legal terms, permeability can be defined as the capacity of a given legal order to limit its own claim of normative exclusivity in order to enable legal rules or principles which emanate from a formally separated legal order to penetrate or integrate.73 At the national level, permeability is provided for especially by so-called ‘integration clauses’, which today can be found in almost all European Constitutions, 70   Polish Constitutional Tribunal, judgment of 11 May 2005, K 18/04, Poland’s membership in the European Union (The Accession Treaty) [2005] OTK ZU 2005/5A/49. Para 12 of the English summary, www.ecln.net/documents/Decisions-Poland/2005_pol_constitutionality_of_eu_accession.pdf. 71   The Polish Constitutional Tribunal claims that the Polish Constitution enjoys, because of its Art 8, ‘precedence of binding force and precedence of application within the territory of the Republic of Poland’, in ibid, para 11. 72   Pernice (Lisbon), n 9, 374, who also highlights the material and institutional dimension. 73   See in detail Wendel, n 69, Chapter 1.

Multilevel and Constitutional Pluralism  139 for example Article 23(1) of the German Basic Law or Article 10a of the Czech Constitution.74 These clauses constitute the normative basis for the transfer of powers to the supranational level and – more importantly in this respect – simultaneously ‘open up’ the national legal order to the operation of EU law, as the Czech Constitutional Court has put it aptly in one of its recent leading cases.75 Permeability, however, is not limited to the legal orders of the Member States. It also can be found at EU level. The most striking example is Article 4(2) TEU, according to which the Union shall respect the national identities of its Member States inherent in their fundamental political and constitutional structures. This stipulation may be seen as the Union’s auto-limitative response to the national jurisprudence aiming at the protection of national constitutional identity.76 Another mechanism of legal permeability are the general principles, through which the constitutional traditions of the Member States flow into EU law. Being part of the (positive) law of the respective legal order, permeability cannot normatively exceed this legal framework. It is not an instrument settled at a meta-level from the perspective of which it could operate or govern the inter­ action of national and supranational legal orders in an overarching manner. Permeability produces legal effects only with respect to the legal order it belongs to. However, it entails an integrative effect because it partially dissolves the boundary between inside and outside from the perspective of this legal order. This order performs an auto-limitation by means of which it revokes (to some extent) its own claim of normative exclusivity within its scope of application and thus becomes permeable.77 Permeability, although legally confined to one single legal order, thereby contributes to the creation of a holistic entity. This entity is not to be confounded with a unitary legal order, but rather emerges out of mutual adjustment, accommodation and interaction.78 It is a unity, above all, in a func­ tional sense, normatively emanating from the participating legal orders. As permeability only covers a specific aspect of this interaction, it must necessarily be supplemented by other principles and techniques, especially by an appropriate institutional setting as well as a regime of delimited competences including a collision norm. Permeability is also usually combined with safeguard mechanisms aiming at the protection of specific substantive core principles of the respective legal order. 74   For an analysis comprising all 27 Member States as well as the candidate countries see Wendel, n 69, Chapters 4 to 11. 75   Czech Constitutional Court, judgment of 8 March 2006, Pl. ÚS 50/04, Sugar Quota Regulation III. For a comment see I Šlosarcˇík, ‘Czech Republic and the European Union Law in 2004–2006’, (2006) 13 European Public Law 367, 373 and W Sadurski, ‘ “Solange, Chapter 3”: Constitutional Courts in Central Europe – Democracy – European Union’ (2008) 14 European Law Journal 1, 6 ff. 76   See also FC Mayer, ‘Multilevel Constitutional Jurisdiction’ in A von Bogdandy and J Bast (eds), Principles of European Constitutional Law 2nd edn (Oxford, Hart, 2009) 423 ff. 77   In this respect permeability operates as a ‘one-sided collision norm’, see S Kadelbach, Allgemeines Verwaltungsrecht unter europäischem Einfluß (Tübingen, Mohr Siebeck, 1999) 219. 78   To speak of ‘one system’ or ‘one unit’, Pernice (Lisbon), n 9, 374, may therefore be mistakable, as it could easily be confounded with ‘one legal order’, which is not intended.

140  Franz C Mayer and Mattias Wendel

D.  The Compatibility of Multilevel Constitutionalism with Constitutional Pluralism Multilevel constitutionalism cannot legally eliminate its pluralist presupposition. It seeks coherence, but not for the price of a hierarchically structured legal order. Against this background, a contrast between multilevel constitutionalism and constitutional pluralism could only be construed if one mistook either legal pluralism for chaos or multilevel constitutionalism for a unitary legal order based on the idea of undivided sovereignty.79 But neither does multilevel constitutionalism presuppose the idea of undivided sovereignty, nor is constitutional pluralism to be confused with chaos. Pluralist tendencies may be identified even within federal states in the sense of mutually conflicting claims of authority between different state organs, being insofar part of the institutional system of checks and balances.80 Speaking of misunderstandings: when conceptualising Europe’s constitutional pluralism, one should also be very careful in associating it with the Bund approach suggested by Carl Schmitt.81 Even if some elements – especially the unresolved question of sovereignty – seem to coincide at first glance, several aspects of Schmitt’s ‘Theory of the Constitution’ of 1928 (Verfassungslehre)82 remain problematic as they may appear to be embedded in a specific theoretical framework which emphasised the pre-eminence of power politics (Machtpolitik) over law. Transplanting elements of this theory into the context of the Community of Law (Rechtsgemeinschaft)83 that the EU is, runs the risk of carrying fragments of this ‘dark side’ as a stowaway.84

IV.  Querelle Allemande or Querelle d’Allemand? Is the German debate on multilevel constitutionalism actually relevant for debates in other Member States or is it just a German quarrel? A simple answer would argue that whatever happens in the legal discourse of the largest Member State should matter. And indeed, there is some evidence that the debates and positions   M Avbelj, ‘Questioning EU Constitutionalisms’ (2008) 9 German Law Journal 1, 19.   In this respect, concerning the constitutional order of the United States, see D Halberstam’s Chapter in this volume. 81   See already H-P Folz, ‘Verfassungslehre des Bundes von Carl Schmitt und die Europäische Union’ in M Wittkopp-Beine (ed), Carl Schmitt in der Diskussion (Plettenberg, Stadtarchiv Plettenberg, 2006) 69, 83, and M Avbelj’s Chapter in this volume. The Schmittian idea of homogeneity as precondition for democracy appears equally problematic, cf JHH Weiler, ‘The State “über alles”’ in O Due et al (eds), Festschrift Everling (Baden-Baden, Nomos, 1995). 82   C Schmitt, Verfassungslehre (Duncker and Humblot, Berlin, 1928). 83   W Hallstein, Der unvollendete Bundesstaat (Econ, Düsseldorf, 1969) 228. 84   I Pernice, ‘Carl Schmitt, Rudolf Smend und die europäische Integration’ (1995) 120 Archiv des öffentlichen Rechts 100 ff. 79 80

Multilevel and Constitutional Pluralism  141 in the German legal academic debate (A) and the German judiciary (B) are closely watched from other Member States.

A.  The German Academic Debate in Legal and Political Science One may object that the German debate on multilevel constitutionalism and the European Constitution is just a very German debate, explained by a particular German constitutional psychology which does not exist elsewhere. Staat and Staatlichkeit (statehood) have been coined the ‘central complex of German constitutional psychology’ from an outside observer’s perspective.85 But maybe a ‘central complex’ of constitutional psychology can be detected in most countries. In Britain, for example, it seems to be Dicey.86 Dicey’s nineteenth century, Victorian view on public law is a central reference point even for those who are critical of his approach. The questions are the same, focusing on parliamentary sovereignty, the relationship between the courts and the administration, and, of course, the distinction between public law and private law. It is true that there is this strange conflation of state and Constitution in German law. Actually, it is not conflation, it is substitution. People mean Constitution, but they say state. The basics of constitutional law are not taught in ‘Constitutional Law’ classes, Verfassungsrechtsvorlesungen, but in Staatsrechtsvorlesungen, literally translated ‘classes on the law of the state’. Using Staatsrecht instead of Verfassungsrecht is a pattern: the impressive books of many thousand pages that embody the mainstream of German constitutional thought are not Handbooks of Verfassungsrecht, but rather a multi-volume Handbook of Staatsrecht, edited by Isensee and Kirchhof.87 Another example is the quite exclusive association of German public law scholars, founded under the Weimar Republic. In its name, it refers neither to public law nor to constitutional law, but to Staatsrecht – it is the Staatsrechtslehrervereinigung. Of course, there are textbooks today entitled ‘Verfassungsrecht’, but in the law faculties, the classes dealing with constitutional law remain classes in ‘Staatsrecht’. All this indicates that in Germany, the relationship between state and Constitution has been peculiar, even outside the European Union context. And indeed, one way to look at the fierce controversy on whether it is all about the state or all about the Constitution could be that it is just an old debate between liberal and conservative scholars taken to another, new arena, the European integration arena. 85   P Allott, ‘The Crisis of European Constitutionalism’ (1997) 34 Common Market Law Review 439, 444; see also Weiler, n 81, 1651; see on this C Möllers, Staat als Argument (München, CH Beck, 2000) 407. 86   AV Dicey, An Introduction to the Study of the Law of the Constitution 10th edn (London, Macmillan, 1959). 87   Klaus Stern’s impressive handbook on constitutional law has the simple title Staatsrecht. A counter concept, the Handbook of Constitutional Law, Handbuch des Verfassungsrechts, has remained less relevant than the Isensee and Kirchhof books.

142  Franz C Mayer and Mattias Wendel Unfortunately, things are more complicated. This is indicated by the participation in this debate of scholars such as Dieter Grimm, who do not fit into this narrative. In the domestic context, his writings and his opinions as a judge of the Constitutional Court are considered to be liberal. But he has been one of the most outspoken critics of the constitutionalist approach to European integration.88 So there must be something that brings together scholars such as Grimm and Kirchhof, who in the domestic context would rather be considered antagonists. Looking closer at the critique, it appears that the critics do not only query the decoupling of the concept of Constitution from the concept of ‘state’.89 They underline the risk of weakening the national Constitution, inherent in the idea of a European Constitution or European constitutionalism, since the structural security built into national Constitutions is called into question. Thus, the argument continues, a Constitution is the enactment of an existing legal culture, which must be developed to some degree, and this level of development has not yet been achieved by the EU.90 Such an emphatic approach to the concept of Constitution may have numerous advantages, not least the familiarity of the Constitution’s interpreters with this concept. The driving force in the German debate seems to be distrust, including distrust towards European law and European law scholars and their (lack of) understanding of the Staat. Scepticism against a constitutional approach to European integration turns out to be mainly about protection and safeguarding constitutional identity – and hierarchy. This kind of scepticism is actually not really new. Earlier attempts to establish a substantive understanding of ‘Constitution’ for non-statal entities were met with resistance as well: just re-read Carl Schmitt’s critique of Alfred Verdross’s attempt to establish a Constitution of the international community based on a Grundnorm ‘pacta sunt servanda’, which was published at the time of the Weimar Republic.91 This kind of approach brings about a debate not only on the link between statehood and Constitution, but also on the basis of a legal order, the exact location of 88   D Grimm, ‘Europäischer Gerichtshof und nationale Arbeitsgerichte aus ver­fas­sungs­rechtlicher Sicht’ (1996) Recht der Arbeit 66. English version: ‘The European Court of Justice and National Courts: The German Constitutional Perspective after the Maastricht Decision’ (1997) 3 Columbia Journal of European Law 229. Grimm categorises the Community somewhere between State and International Organisation, not being a state because of the lacking self-determination on form and content of its political existence; not being an International Organisation because of a surplus of public authority, ibid, 66 ff. 89   See above. 90   P Kirchhof, ‘Große Entwürfe und kleine Schritte’ Handelsblatt (Düsseldorf, 8 August 2001) 6. 91   A Verdross, Die Verfassung der Völkerrechtsgemeinschaft (Wien, Springer, 1926); Schmitt, n 82, 69 and 363 ff. For the concept of a Constitution of the international community see also A Verdross and B Simma, Universelles Völkerrecht 3rd edn (Berlin, Duncker and Humblot, 1984) § 75 ff, with further references. On the UN Charter as a Constitution of the international community see also B Simma, ‘From Bilateralism to Community Interest in International Law’ (1994) 250 Recueil des Cours de l’Académie de Droit International 217, 258 ff. For tendencies of constitutionalisation at the WTO level see R Howse and K Nicolaidis, ‘Legitimacy and Global Governance: Why Constitutionalizing the WTO is a Step too Far’, in RB Porter et al (eds), Equity, Efficiency and Legitimacy: The Multilateral Trading System at the Millenium (Washington, Brookings Institution Press, 2001).

Multilevel and Constitutional Pluralism  143 sovereignty, the core of federalism, etc. These questions as well as the question of how to preserve constitutional identity, it seems to us, are questions that keep coming back – not only in Germany – from slave-holding states in nineteenthcentury USA and their claim of ‘indivisible sovereignty’ to more recent debates on constitutional autonomy etc. in Hungary. Thus, the academic debate is not just a querelle allemande.

B.  The German Judiciary: the 2009 Lisbon Decision of the Federal Constitutional Court as an Example92 The 2009 Lisbon judgment of the German Federal Constitutional Court is a ‘litmus test’ about possible German peculiarities as far as the judiciary is concerned, and it also indicates to what extent German constitutional positions on European integration are noticed, debated, understood and taken up outside Germany.93

i.  German Peculiarities? Taking into account the above-mentioned academic debate on multilevel constitutionalism in Germany, it is striking that the Lisbon judgment highlights the concept of Verbund in its very first headnote. According to the federal judges, the EU is designed as an association of sovereign national states (Staatenverbund). The concept of Verbund covers a close long-term association of states which remain sovereign, an association which exercises public authority on the basis of a treaty, whose fundamental order, however, is subject to the disposal of the Member States alone and in which the peoples of their Member States, i.e. the citizens of the states, remain the subjects of democratic legitimisation.

Clearly, the Federal Constitutional Court does not pick up the concept of multilevel constitutionalism, but – on the contrary – continues along the path it had already taken in 1993 in its Maastricht decision.94 The specific focus on the state which is intrinsic to the concept of Staatenverbund can be seen as a German particularity, as we already pointed out. But the state-centred perspective of the German Federal Constitutional Court in the Lisbon judgment is not limited to the concept of Staatenverbund. Moreover, the Court now emphasises the principle of ‘sovereign statehood’ (souveräne Staatlichkeit) as a leading paradigm.95 92   Franz Mayer was counsel to the Bundestag (German Parliament) in the Lisbon case of the German Constitutional Court. The views expressed here on the Lisbon decision do not necessarily reflect positions of the Bundestag. 93   With 147 pages and more than 400 paragraphs, the judgment is not only voluminous, but also very complex in its structure and argument. It is clear that the following remarks can only briefly assess some of its key aspects which are particularly relevant for the present context. For the debate about the decision in Germany and beyond, see the references in Voßkuhle, n 34. See FC Mayer, ‘Rashomon in Karlsruhe’ (2010) Neue Juristische Wochenschrift 714 (English version http://www.whi-berlin.de) for an illustration of the decision, which may be interpreted differently, depending on the perspective of the reader. 94   Maastricht, n 6 at 184 ff. See also Kirchhof (‘The Legal Structure’), n 44, 743 ff. 95   Cf D Thym, ‘In the Name of Sovereign Statehood: A Critical Introduction to the Lisbon Judgment of the German Constitutional Court’ (2009) 46 Common Market Law Review 1795 ff.

144  Franz C Mayer and Mattias Wendel The emphasis on the principle of sovereign statehood is noteworthy, because for a long period of time sovereignty had played – if at all – only a minor part within the constitutional framework of the German Basic Law.96 This is partly explained by the fact that Germany regained full sovereignty only with reunification in 1990. Still, at a time when the demise or conceptual transformation of sovereignty seems to be increasingly accepted in legal sciences,97 one may wonder why the German Federal Constitutional Court reintroduces the notion of sovereignty just at the present moment and just within the context of European integration. ‘Sovereign statehood’ might, at first view, be taken as synonymous with classical ‘state sovereignty’.98 However, a closer look at the judgment soon reveals that the Federal Constitutional Court seeks to depart from a classical notion of sovereignty. The Court underlines that the German Basic Law ‘breaks with all forms of political Machiavellianism and with a rigid concept of sovereignty’. But first of all, the Court deduces the necessity to protect ‘sovereign statehood’ primarily from the principle of democracy99 which is, in procedural terms, protected by the right to vote (Article 38). Hence, according to the Court, sovereign statehood stands for a pacified area and the order guaranteed therein on the basis of individual freedom and collective self-determination. The state is neither a myth nor an end in itself but the historically grown and globally recognised form of organisation of a viable political community.100

Sovereign statehood thus appears as an auxiliary means to the end of individual freedom and collective self-determination. Herein lies, at first sight, a considerable shift in German constitutional doctrine, leading away from a state-centred perspective. If the state is detected to be a mere instrument in order to realise constitutionally founded individual freedom,101 then the field is open also to new modes of democratically legitimised governance beyond the state. It is here that the argument of the German Federal Constitutional Court becomes circular, as the Court assumes that only a state or a quasi-statal unit is able to ensure the necessary degree of democratic legitimation. In other words, according to the Federal Constitutional Court the protection of statehood is a precondition for the protection of democratic legitimation. Popular sovereignty thereby is inseparably tied to statehood. The consequence of this conceptual bondage is a binary, all-or-nothing-­ fashioned choice: either the existing association of sovereign national states 96   For first echoes see Kloppenburg, n 43 at 242 and Case 2 BvR 1481/04 Görgülü I [2004] BVerfGE 111, 307 at 319. 97   See N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003). 98   Cf CF von Gerber, Grundzüge eines Systems des deutschen Staatsrechts (Aalen, Scientia, 1969 [reprint of the 3rd edn 1880]) 22 and G Jellinek, Allgemeine Staatslehre 3rd edn (Berlin, Häring, 1914) 169 ff. 99   Lisbon, n 32, paragraph 248. 100   Ibid, para 224 (emphasis added). 101   See Möllers, n 85.

Multilevel and Constitutional Pluralism  145 (Staatenverbund) is to be continued, or a new European federal state would have to be founded, the participation in which would require a new Constitution in Germany. If the threshold to the federal state and to the waiver of national sovereignty were transgressed, which would in Germany require a free decision of the people beyond the present applicability of the Basic Law, democratic requirements would have to be complied with on a level which would have to completely fulfil the requirements placed on the democratic legitimisation of a union of rule organised by a state. This level of legit­ imisation could no longer be prescribed by national constitutional orders.102

To put it bluntly, despite claiming the contrary,103 the German Federal Constitutional Court ties the legal guarantee of democracy strictly to the (pre‑)existence of (quasi‑) statehood. Sovereignty of the people and sovereign statehood thus become two inseparable sides of one coin, of ‘sovereign statehood’. Almost symptomatic in this respect is the Court’s definition of ‘peoples’ which are equalised with ‘the citizens of the states’.104 The journey on the traces of the Court’s concept of sovereign statehood thus ends where it has previously started: the concept of statehood. Consequently, the Member States continue to be the ‘masters of the Treaties’,105 which constitute no more than a mere derived legal order. The ‘Constitution of Europe’, the law of international agreements or primary law, remains a derived fundamental order. It establishes a supranational autonomy which is quite far-reaching in political everyday life but is always limited factually. Here, auto­ nomy can only be understood – as is usual regarding the law of self-government – as an autonomy to rule which is independent but derived, i.e. is accorded by other legal entities. In contrast, sovereignty under international law and public law requires independence of an alien will particularly for its constitutional foundations.106

Hence, the Member States remain ‘the constituted primary political area of their respective polities’, while the EU enjoys only ‘secondary, i.e. delegated, responsibility for the tasks conferred on it’.107 According to the Federal Constitutional Court, the protection of sovereign statehood furthermore has to be ensured by a judicial claim to ultimate authority, concerning not only the question of ultra vires acts, but also the compatibility of (directly applicable) EU law with fundamental rights and the constitutional ‘identity’ of the German Basic Law.108

102   Lisbon, n 32, para 263 (emphasis added). The English translation of the second part of the first sentence seems to be rather obscure. The German original reads: ‘. . . müssten demokratische Anforderungen auf einem Niveau eingehalten werden, das den Anforderungen an die demokratische Legitimation eines staatlich organisierten Herrschaftsverbandes vollständig entspräche’. 103   Ibid, para 267. 104   Ibid, para 229. 105   Maastricht, n 6 [1993] BVerfGE 89, 155 at 199. 106   Lisbon, n 32, para 301. See also para 231. 107   Ibid, para 301. 108   Ibid, para 331 ff and 340.

146  Franz C Mayer and Mattias Wendel Moreover, the Federal Constitutional Court interprets the so called ‘eternity clause’ in Article 79(3)109 in a rather detailed and justiciable manner, identifying certain key areas which shall be immune against the transfer of competences to the EU level under the German Europe clause, Article 23(1) of the German Basic Law: European unification on the basis of a union of sovereign states under the Treaties may, however, not be realised in such a way that the Member States do not retain sufficient space for the political formation of the economic, cultural and social circumstances of life. This applies in particular to areas which shape the citizens’ circumstances of life, in particular the private space of their own responsibility and of political and social secur­ ity, which is protected by the fundamental rights, and to political decisions that particularly depend on previous understanding as regards culture, history and language and which unfold in discourses in the space of a political public that is organised by party politics and Parliament. Essential areas of democratic formative action comprise, inter alia, citizenship, the civil and the military monopoly on the use of force, revenue and expenditure including external financing and all elements of encroachment that are decisive for the realisation of fundamental rights, above all as regards intensive encroachments on fundamental rights such as the deprivation of liberty in the administration of criminal law or the placement in an institution. These important areas also include cultural issues such as the disposition of language, the shaping of circumstances concerning the family and education, the ordering of the freedom of opinion, of the press and of association and the dealing with the profession of faith or ideology.110

ii.  Reactions to the Lisbon Decision outside Germany It is not clear whether the German Constitutional Court has always been aware of the impact of its decisions throughout the rest of Europe. With the decision in the European arrest warrant case the Court issued an English translation of its ruling for the first time. A (preliminary) English translation of the Lisbon judgment was published the very day of its promulgation. This seems to indicate that the Court is not only aware of the fact that it is being closely watched from outside Germany – it almost appears as if the Court wants to be heard outside Germany, as well. There are several waves of reception or ‘migration’111 of the Court’s jurisprudence that can be identified within the context of European integration.112 Primarily, its famous Solange jurisprudence regarding the protection of fundamental rights as well as its claim of ultimate decision-making power on ultra vires acts has inspired constitutional and highest courts all across Europe. Distinct examples can be found in the jurisprudence of the Danish Highest Court,113 the 109   Art 79(3) reads: ‘Amendments to this Basic Law affecting the division of the Federation into Länder, their participation on principle in the legislative process, or the principles laid down in Articles 1 and 20 shall be inadmissible’. 110   Lisbon, n 32, para 249. 111   S Choudhry, ‘Migration as a new metaphor in comparative constitutional law’ in S Choudhry (ed), The Migration of Constitutional Ideas (Cambridge, Cambridge University Press, 2006) 1 ff. 112   See on that Mayer, n 16. 113   Danish Supreme Court, judgment of 6 April 1998, I 361/1997, Carlsen/Rasmussen [1998] UfR 800.

Multilevel and Constitutional Pluralism  147 Polish Constitutional Tribunal114 as well as the Czech Constitutional Court.115 Some elements have even found their way into constitutional texts. The Solange formula, for example, was adopted by a former version of the Swedish Europe clause (Chapter 10 § 5 of the Swedish Regeringsform). Further, the ultra vires reasoning is mirrored by the text of the Hungarian Europe clause (Article 2a of the Hungarian Constitution).116 However, the German Federal Constitutional Court’s approach concerning the eternity clause as a limitation to the integration process has, until today, remained unique. No other court relied on such a clause or concept as extensively as the German Federal Constitutional Court relied on Article 79(3) of the German Basic Law. This is true not only for Article 89(5) of the French Constitution and its interpretation by the Conseil constitutionnel in its Maastricht III judgment,117 but also for Article 139 of the Italian,118 Article 110 of the Greek,119 Article 288 of the Portuguese,120 Article 152(1) and (2) of the Romanian121 and Article 182 of the Cypriot Constitutions.122 Recently, the contrast to the German approach was expressed most explicitly by the second Lisbon judgment of the Czech Constitutional Court. The Czech judges expressly rejected the demand of the applicants to formulate a catalogue of non-transferable rights under the Czech eternity clause: In point 49 of the petition the petitioners ask the Constitutional Court to set ‘substantive limits to the transfer of powers’, and in points 51 to 56 they attempt to formulate these themselves, evidently inspired by the decision of the German Constitutional Court dated 30 June 2009 . . . However, the Constitutional Court does not consider it possible, in view of the position that it holds in the constitutional system of the Czech Republic, to create such [reference to the German Lisbon judgment, para 252] a catalogue of non-transferable powers and authoritatively determine ‘substantive limits to the transfer of powers’, as the petitioners request. It points out that . . . [t]hese limits should be left primarily to the legislature to specify, because this is a priori a political question, which provides the legislature wide discretion . . . Responsibility for these political decisions   Polish Constitutional Tribunal, n 70.   Czech Constitutional Court of the Czech Republic, n 75. 116   This cannot be dealt with in detail here. For a comprehensive analysis of the lines of reception see Wendel (n 69) Chapter 11. 117  French Constitutional Council, judgment of 23 September 1992, 92-313 DC, Maastricht III [1992] Recueil 94. See J Ziller, ‘Sovereignty in France: Getting Rid of the Mal de Bodin’ in N Walker (ed), Sovereignty in Transition (Hart, Oxford, 2003) 271. 118  M Cartabia, ‘The Legacy of Sovereignty in Italian Constitutional Debate’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 305, 316. 119   See J Iliopoulos-Strangas, ‘Offene Staatlichkeit in A von Bogdandy and PM Huber (eds), Ius Publicum Europaeum. Volume II (Heidelberg, CF Müller, 2008) § 16 Griechenland, para 43 ff. 120   M Poiares Maduro, ‘EU Law and National Constitutions: Portugal’, unpublished manuscript for XXth FIDE Congress London (London 2002). 121   V Duculescu and A Ruxandra, ‘Romania’ in AE Kellermann et al (eds), The Impact of EU Accession on the Legal Orders of New Member States and (Pre-)Candidate Countries (The Hague, TMC Asser Press, 2006) 118 ff. 122   For the legal situation in Cyprus see N Emiliou, ‘Cyprus’ in AE Kellermann et al (eds), The Impact of EU Accession on the Legal Orders of New Member States and (Pre-) Candidate Countries (The Hague, TMC Asser Press, 2006) 304 ff. 114 115

148  Franz C Mayer and Mattias Wendel cannot be transferred to the Constitutional Court; it can review them only at the point when they have actually been made on the political level . . . For the same reasons, the Constitutional Court does not feel authorised to formulate in advance, in an abstract context, what is the precise content of Article 1(1) of the Constitution, as requested by the petitioners, supported by the president, who welcomes the attempt ‘in a final list to define the elements of the ‘material core’ of the constitutional order, or more precisely, of a sovereign democratic state governed by the rule of law’, and states (in agreement with the petitioners) that this could limit future self-serving definition of these elements based on cases being adjudicated at the time.123

This example is a clear signal that the interaction and mutual influence of the jurisprudence of the Members States does not necessarily lead to the reception of a specific judgment in the jurisprudence of other courts, but may in fact evoke express rejection. Nevertheless, it should be acknowledged that the German Constitutional Court’s decisions have an impact outside Germany and that they do trigger reactions, dismissive as well as affirmative.

iii.  A Side Remark on Pluralist Elements in the Lisbon Decision In spite of the German Federal Constitutional Court’s approach of construing EU law as a legal order having derivative character, one may also identify pluralist elements in the Lisbon judgment. In the context of the protection of national identity, the Court refers to the ECJ’s Kadi judgment,124 stating that placing the assertion of one’s own identity as a legal community above the commitment that one otherwise respects was not only familiar in international legal relations but also corresponded, ‘at any rate if it is used in a constructive manner, to the idea of contexts of political order which are not structured according to a strict hierarchy’.125 The Court thus construes the protection of national identity under Article 4(2) TEU as the normative correlate or mirror of its own competence of protecting the German constitutional identity. According to the Federal Constitutional Court, the guarantee of national constitutional identity under constitutional and under EU law go hand in hand in the European legal area.126 Moreover, the newly claimed principle of openness towards European law (Europarechtsfreundlichkeit) may – despite its normative cloudiness – be read as the acknowledgment of a pluralist assumption by the court.127 123   Czech Constitutional Court, judgment of 3 November 2009, Pl. ÚS 29/09 Treaty of Lisbon II, English translation http://www.usoud.cz/view/726, paragraph 110 ff. 124  Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat International Foundation v Council and Commission [2008] ECR I-6351. 125   Lisbon, n 32, para 340 (emphasis added). 126   Ibid, para 240. 127  The concept of Europarechtsfreundlichkeit is developed further by the German Constitutional Court in Case 2 BvR 2661/06, judgment of 6 July 2010, Honeywell, http://www.bundesverfassungsgericht. de/en/decisions/rs20100706_2bvr266106en.html, paragraphs 58 ff.

Multilevel and Constitutional Pluralism  149 However, the understanding of Article 4(2) TEU as an authorisation for national courts to unilaterally decide on the applicability of EU law within the national legal order does not correspond to an appropriate conception of the ‘identity clause’ on the EU level. Indeed, EU law cannot determine the respective identity of its Member States. Article 4(2) leaves this determination to the Member States and national constitutional courts. However, it is to be decided by EU law – that is, by the ECJ – how far EU law limits its own claim of normative exclusivity and thus confers a normative relevance to the national identity claim within the realm of EU law. In other words, national constitutional courts decide on the extent of their national constitutional identity, while the ECJ decides on the extent of the normative relevance of the national identity claim, its permeation at supranational level. A concrete example may be found in the various margins of appreciation that the ECJ grants in the field of fundamental freedoms.128

C.  Querelle d’Allemand? If the German academic debate and the German Constitutional Court’s decisions reveal topics and concerns that are not typically German, the question remains whether these issues are really relevant, or whether they are a querelle d’Allemand, a quarrel about nothing. Does it really matter whether we use constitutionalism in order to understand and explain European integration? Or are we in a setting that resembles the antagonism between monism and dualism, which mainly exits in theory and in textbooks? And isn’t it true that even in Germany, suggesting the concept of constitutionalism for European integration is not considered a provocation any more? In this context it is worthy to note that the Judge rapporteur of the Lisbon case, Udo Di Fabio, at some point during oral proceedings in February 2009 said that nobody seriously challenges the assumption any more, that primary law can be considered constitutional law, a statement which may also be found in the Lisbon judgment itself.129 On the other hand, there are numerous examples that the public international law paradigm is quite alive in the European integration context. The approach of the German Constitutional Court in the European arrest warrant case is an example, but also the decisions of the Polish Constitutional Court concerning the EU are based on the underlying assumption that it is all about public international law treaties. And despite Udo Di Fabio’s remark, numerous elements of the Lisbon decision of the German Constitutional Court emphasise a public international law reading of European integration. A very short answer to justify the constitutional law approach to European integration points to the specific nature of European law – the density of the 128   See recently NN Shuibhne, ‘Margins of Appreciation: National Values, Fundamental Rights and EC Free Movement Law’ (2009) 34 European Law Review 230 ff. 129   Lisbon, n 32, para 231.

150  Franz C Mayer and Mattias Wendel public authority established by directly effective European law. The fact that the EU is the only non-statal entity that offers fundamental rights protection against its own acts is the flipside of this density. A constitutional law understanding of European integration is simply much better suited to grasp the relationship between European law and national law. More generally speaking, it is the developments at the supra-state level throughout the past 60 years with the emergence of public authority outside the traditional nation state that justify the call for a ‘rethinking of the concept of Constitution’.130 It is the changed circumstances of a ‘post-national constellation’ (Jürgen Habermas)131 that require a pragmatic concept of constitutionalism, emphasising that there is no state or public power beyond that established by the Constitution. And as far as the European Union is concerned, there are two observations that remain relevant in the present context: first, at least some Member State Constitutions clearly distinguish between European integration and other international activities.132 More importantly, EU public power already exists; it is out there and affects the individual directly in his or her legal status. In other words: if an individual in the EU wants to find out about his or her constitutional status, it is no longer sufficient to read the respective national Constitution, it requires a closer look at European law as well. We are not confronted with a question d’Allemand – on the contrary.

V.  Conclusion and Summary Looking back at how the concept of multilevel constitutionalism was invented and discussed, one can ask whether the critique and the challenges of the concept are specifically German and whether the concept and the debate around it matter at all. It turns out that the concerns behind the challenges are not exclusively German, and that the issues at stake in the debate are indeed relevant. It is therefore neither a querelle allemande nor just a querelle d’Allemand. It also turns out that multilevel constitutionalism and constitutional pluralism display significant similarities in terms of theoretical foundations, without merging into one single concept. While both concepts depart from similar pre130   P Badura, ‘Bewahrung und Veränderung demokratischer und rechtsstaatlicher Verfassungsstruktur in den internationalen Gemeinschaften’ (1966) 23 Veröffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer 34, 95. 131   J Habermas, Die postnationale Konstellation (Frankfurt-am-Main, Suhrkamp, 1998) 91 ff; see also M Zürn, ‘The State in the Post-National Constellation’ (1999) ARENA Working Paper, No 99/35; see also D Curtin, Postnational Democracy (The Hague, Kluwer, 1997) 5, 48 ff and 51 ff; J Shaw, ‘Postnational Constitutionalism in the European Union’ (1999) 18 Journal of European Public Policy 579, 586 ff. 132   The German Constitution, for example, points beyond itself by referring to the objective of a unified Europe (zur Verwirklichung eines vereinten Europas) in the Preamble and in Art 23(1).

Multilevel and Constitutional Pluralism  151 suppositions, it is the particular accentuation of coherence, mutual stabilisation and functional unity which renders multilevel constitutionalism conceptually distinct. The relevance of the debate on constitutionalism beyond the state is underlined by more recent developments elsewhere in the world, where constitutionalism has become an issue as well, in a theoretical and, considering the example of the Kadi case, a very practical context. The debate will go on.

7 The Fallacy of European Multilevel Constitutionalism RENÉ BARENTS*

Le juriste doit se garder de recourir au symbolique – le mot constitution et ses implications politiques – pour décrire l’objet qu’il tente d’analyser.1

The subject of this Chapter is a critical evaluation of the theory called ‘European multilevel constitutionalism’ (hereafter: ‘multilevel theory’). It starts with a short overview of the contents and background of this theory, followed by a critical examination of its main theses and underlying assumptions. The starting point for the discussion is the state of the law of the European Union (hereafter: ‘Union’) as laid down in the Treaty of Lisbon.2

I.  Multilevel Constitutionalism in a Nutshell A.  What’s in a Name? The theory under consideration bears a variety of names which do not all correspond to each other. In the Anglo-Saxon context it is known as ‘multilevel constitutionalism’,3 in Germany as ‘Verfassungsverbund’4 and in French, Italian *  The opinions expressed in this paper are strictly personal. 1   P-Y Monjal, ‘Le projet de traité établissant une Constitution pour l’Europe: quels fondements théoriques pour le droit constitutionnel de l’Union européenne?’ (2004) Revue trimestrielle de droit européen 465. 2   [2007] OJ C 306/1. 3   I Pernice, ‘Multilevel Constitutionalism and the Treaty of Amsterdam: European ConstitutionMaking Revisited?’ (1999) 36 Common Market Law Review 703; I Pernice, ‘Multilevel constitutionalism in the European Union’ (2002) 27 European Law Review 511; I Pernice, ‘Theorie und Praxis des Europäischen Verfassungsverbundes’ in C Callies (ed), Verfassungswandel im europäischen Staatenund Verfassungsverbund (Tübingen, Mohr Siebeck, 2007) 61 et seq. 4   I Pernice, ‘Bestandssicherung der Verfassungen: Verfassungsrechtliche Mechanismen zur Wahrung der Verfassungsordnung’ in R Bieber and P Widmer (eds), L’espace constitutionnel européen (Zurich, 1995); I Pernice, ‘Constitutional Law Implications for a State Participating in a Process of Regional Integration. German Constitution and ‘Multilevel Constitutionalism’ in E Riedel (ed), German Reports on

154  René Barents and Dutch literature as ‘constitution composée’,5 ‘costituzione integrata’6 and ‘samengestelde constitutie’7 respectively. The German denomination refers to a system of national constitutions in the context of the European treaties. From the word ‘Verbund’8 it follows that this system might be described as an ‘alliance’ or a ‘compound’, which seems to indicate a relation between independent elements. The French, Italian and Dutch denominations seem to imply a stronger relation between national Constitutions and the European Treaties (‘composed’, ‘integrated’). The English denomination differs considerably from the others as it employs the term ‘constitutionalism’, which is wider than ‘Constitution’. Moreover, the adjective ‘multilevel’ points to a system consisting of levels which, at least at first sight, supposes a certain hierarchy, since in such a system one level is necessarily higher or lower than the other.9

B.  The Concept of European Constitution and Multilevel Governance As appears from the literature on this subject, the multilevel theory concerns the relations between the Constitutions of the Member States (written or unwritten) and the European Treaties, which are also considered to be a Constitution. According to the multilevel theory, together these Constitutions constitute a unitary system in terms of substance, functions and institutions.10 This system bears the name ‘European Constitution’ (‘Verfassungsverbund’). Accordingly, this concept must be distinguished from the ‘European Constitution’ in a narrow sense (the European Treaties or the defunct Constitution for Europe). Furthermore, it must be emphasised that in the multilevel theory the connection between the European level and the national levels (in the framework of the ‘European Constitution’) is of a non-hierarchical nature, that is, not a vertical but a horizontal or parallel construction of Constitutions.11 It would thus seem that the German, French, Italian and Dutch denominations indicate the heterarchical nature of the Public Law (Baden-Baden, Nomos 1998) 40–66; I Pernice, ‘Europäisches und nationales Verfassungsrecht’ (2001) 60 Veröffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer; I Pernice, ‘Eine Rechtsgemeinschaft: Die neue Verfassungsgrundlage für die Europäische Union’ in O Beaud, A Lechevalier, I Pernice, S Strudel (eds), L’Europe en voie de Constitution. Pour un bilan critique des travaux de la Convention (Brussels, Bruylant, 2004) 140; I Pernice, ‘European v. National Constitutions’ (2005) 1 European Constitutional Law Review 99. 5   I Pernice and F Mayer, ‘De la constitution composée de l’Europe’ (2000) Revue trimestrielle de droit européen 623. 6   I Pernice and F Mayer, ‘La Costituzione integrata dell’Europa’ in G Zagrebelsky (ed), Diritti e Costituzione nell’Unione europea (Rome, Laterza, 2003) 43–68. 7   LFM Besselink, Een samengestelde Europese constitutie (Groningen, Europa Law Publishing, 2007). 8   Not to be confused with ‘Verband’ or ‘Bund’. 9   See in this sense J-L Quermonne, ‘L’approche de l’Union européenne par la science politique’ in Beaud et al (eds), n 4, 154–55. 10   F Mayer, ‘The European Constitution and the Courts. Adjudicating European Constitutional Law in a Multilevel System’ (2003) 9 Jean Monnet Working Paper 1, 50. 11   Pernice (2001), n 4 at 185.

The Fallacy of Multilevel Constitutionalism  155 system in a better way. On the other hand, the concept of ‘multilevel system’ makes clear that this theory is influenced by or derived from inputs of social sciences and organisational theory, in particular by the concept of ‘multilevel governance’.12 The latter concept is used in social and political science to overcome the difficulties in explaining the existence and functioning of the Union in terms of intergovernmentalism and state-centred concepts, the result of which is always more or less a denaturising, both in theory and in practice, of the Union as a legal and political phenomenon. According to the theory of multilevel governance the Union and its Member States represent a system in which regulatory competences are allocated over various actors situated at different levels.13 This view is based on the normative framework for and the empirical observation of the growing interdependence of Union and national policies, in other words the blurring of the distinction between domestic and European affairs.14 In that context, the state no longer occupies an exclusive position. Instead, it constitutes but one of many other actors which, although each is acting within its scope, mutually influence each others’ decision making (‘governance beyond and with but not above the state’). However, the theory of multilevel constitutionalism adds an important element to the model of multilevel governance by explaining the allocation of public powers to the European level and the national levels as a direct result of the common will of the citizens of the Member States.15 The concept of ‘Verfassungsverbund’ constitutes an alternative to the concept ‘Staatenverbund’ (‘alliance of states’), used by the Bundesverfassungsgericht in its notorious 1993 Maastricht judgment.16 Although the exact meaning of the latter term is subject to some controversy,17 it appears from the writings of the author of this judgment that ‘Staatenverbund’ relates to a decorated form of ‘Staatenbund’ (confederation).18 In any event, the ‘Staatenverbund’ concept implies that the   Mayer, n 10.   See, inter alia, L Hooghe and G Marks, Multilevel Governance and European Integration (Lanham, Rowman and Littlefield Publishers, 2001); N Bernard, Multilevel Governance in the European Union (The Hague, Kluwer International, 2002); N Bamforth and P Leyland (eds), Public Law in a MultiLayered Constitution (Oxford, Hart Publishing, 2003); I Bache and M Flinders (eds), Multilevel Governance (Oxford, Oxford University Press, 2004); A Follesdahl, R A Wessel, J Wouters (eds), Multilevel Regulation and the EU (Leiden, Martinus Nijhoff Publishers, 2008). See on the concept of ‘governance’ in the context of European law the study of S Frerichs, Judicial Governance in der europäischen Rechtsgemeinschaft. Integration durch Recht jenseits des Staates (Baden-Baden, Nomos, 2008). 14   Pernice (1998), n 4 at 48; see for an overview RA Wessels and J Wouters, ‘The phenomenon of multilevel regulation: interactions between global, EU and national regulatory spheres’ (2007) International Organizations Law Review 259. 15   Pernice (2007), n 3 at 66–67. 16   Case 2 BvR 2134, 2159/92 Maastricht [1993] BVerfGE 89, 155, published in English as [1994] 1 CMLR 57; Pernice (1998), n 4 at 43, Pernice (2007), n 3 at 67. 17   See B Kahl, ‘Europäische Union: Bundesstaat – Staatenbund – Staatenverbund’ (1994) 33 Der Staat 243. 18   According to Kirchhof, the term ‘Staatenverbund’ is intended to distinguish it from a ‘Staatenbund’ in the (old) sense of a military alliance; see P Kirchhof, ‘Der Staatenverbund der Europäische Union’ in P Hommelhof and P Kirchhof (eds), Der Staatenverbund der Europäische Union (Heidelberg, Müller, 1994) 11. See also HH Klein, ‘Die Zukunft der Europäischen Union und die Identität ihrer Mitgliedstaaten – Zehn Thesen’ in S Magiera, K M Meessen, H Meyer (eds), Politik und Recht, Gedächtnissymposion für Wilhelm A Kewenig (Baden-Baden, Nomos, 1996) 103; F Cromme, ‘Spezifische 12 13

156  René Barents Union is seen as a construction exclusively derived from and controlled by the Member States.19 The ‘Staatenverbund ’ theory is essentially a ‘Herren’ (‘masters’) theory, a rather unpleasant name for a doctrine according to which the relation between the Member States and the Union is presented in feudal terms (‘Herren der Verträge’ – ‘Masters of the Treaty’)20 and which, in other respects, displays a somewhat dubious ideological background.21 The concept of ‘Verfassungsverbund ’ is to be seen as a reaction22 to what has been called the ‘central complex of German constitutional ideology’,23 that is, the dominant position of the concepts of ‘Staat’ and ‘Staatlichkeit’, underlying the thinking of the Bundesverfassungsgericht and many German constitutional scholars.24 Later on, the ‘Verfassungsverbund’ concept was introduced in the Anglo-Saxon sphere under the denomination ‘multilevel constitutionalism’.25 At this background it is submitted that a full understanding of the multilevel theory requires the consultation of the numerous German language publications of its architects.

C.  Main Theses of the European Multilevel Constitutionalism Theory Taking account of this background, the principal elements of the multilevel theory may be presented as follows. Because of their unity in substance, Union law and national law together constitute a composite system in which public power is divided among several autonomous levels, in particular among national and European levels (unity in substance thesis). In this system (‘European Constitution’) the Union and the Member States are conceptualised as a creation by the citizens. Having the political status of national as well as European citizens, they are the single source of legitimacy for the exercise of public authority at both levels (citizenship thesis). In the multilevel system of the ‘European Constitution’ each level of government is autonomous vis-à-vis the others. However, in order to safeguard the Verfassungselemente des Staatenverbundes. Bausteine für die Europäische Union’ (1996) 5 Berliner Europa Studien; H Quaritsch, ‘La souveraineté de l’Etat dans la jurisprudence constitutionnelle allemande’ (2000) Les Cahiers du Conseil constitutionnel 49. 19   R Barents, The Autonomy of Community Law (The Hague, Kluwer International, 2004) Chapter 4. 20   Although it remains difficult to understand how a vassal can be loyal to more than one feudal lord. 21   In particular in view of the close correspondence of various parts of the Maastricht judgment with the constitutional ideology of the prominent Weimar (and later Nazi) scholar Carl Schmitt; see his Verfassungslehre, 8th edn (Berlin, Dunker & Humblot, 1993). 22   As becomes clear from the various writings of Pernice, see also his ‘Carl Schmitt, Rudolf Smend und die europäische Integration’ (1995) 120 Archiv des öffentlichen Rechts 100. 23   P Allott, ‘The crisis of European constitutionalism’ (1997) 34 Common Market Law Review 437, 442. 24   See extensively Barents (2004), n 19 at § 77 et seq. 25   In view of the implied hierarchy this is a misleading name; see also Besselink, n 7 at 6–7.

The Fallacy of Multilevel Constitutionalism  157 coherence of the system as a whole, the citizens have decided in favour of primacy of Union law in order to create an efficient tool for supranational action (auto­ nomy thesis). The multilevel system of the European Constitution is founded on the division of sovereignty between the Member States and the Union (divided sovereignty thesis). The advantages of this theory are manifold. They may be summarised as follows. By founding the existence of the Union exclusively on the will of the citizens, the problem of the position of Member States as ‘pouvoir constituant’ and as ‘Herren’ of the Union, which at least implicitly entails the possibility of an ‘actus contrarius’, is solved. If there are ‘masters’, this position is taken by the citizens as the sole source of legitimacy of the Member States and the Union in the overarching structure of the ‘European Constitution’. Accordingly, there is no need to quarrel about the true constitutional nature of the European Treaties26 since, like national constitutions, they derive their (original) legitimacy directly from the common will of the citizens. By presenting the system of national and European levels as a heterarchical model, it provides an answer to one of the most difficult questions of the relation between the Union and the Member States: who holds the ultimate power (‘Kompetenz-Kompetenz’)? Since the only source of ultimate power is the common will of the citizens to divide public power over various levels, in the multilevel model no entity can be considered to be subordinated to the other. As a result, there is no need to explain the nature of the Union and of the ‘European Constitution’ in terms of a state, federation or confederation. In spite of the autonomy of the various levels, the primacy of Union law over conflicting national law is upheld since this quality is directly based on the will of the citizens. Conflicts between diverging legal rules from different levels relating to the same situation are thus prevented or solved. It provides for an answer to the highly contested question whether concepts as ‘Constitution’, ‘democracy’ and ‘governance’ can be used as analytical tools to explain ‘post-national’ constellations. More generally, this theory provides an explanation of the concept of law or constitutional law ‘beyond the state’, that is, law flowing from a source independent from the state (‘transnational’ or ‘de-nationalised’ law).27 It offers an escape from the almost perpetual binary model of orthodox constitutional thinking according to which a polity can only be perceived as a (sovereign) state or as an ordinary international organisation. In the latter case, such an organisation can never obtain a constitutional or state-like status and is therefore always subordinated to the will of the founding states (tertium non datur).28 26   See on this question, in particular, P Eleftheriadis, ‘Aspects of European Constitutionalism’ (1996) 21 European Law Review 32 and ‘Begging the Constitutional Question’ (1998) 36 Journal of Common Market Studies 235. 27   See on this concept Barents (2004), n 19 at § 380 et seq. 28   Known in Germany as the doctrine of the ‘Staatenverbindungen’ (‘associations of states’), in the Anglo-Saxon context ‘(con)federations’; see for an overview of these theories Barents (2004), n 18 § at 62 et seq.

158  René Barents It follows from these observations that in the multilevel theory nearly all essential questions about the nature of the Union, Union law and its relation to national law are solved. This explains its wide acceptation and popularity.29 In any event, criticisms of this theory are scarce.30 Furthermore, these observations make clear that the period of the Declaration of Laeken, the Convention on the Future of Europe and the Constitution for Europe (2002–05) constituted the high noon of this theory since at the background of a presumed European ‘constitutional moment’, these events could be seen as a first crystallisation of democracy, gov­ ernance and constitutionalism beyond the state.31 It is not clear whether the subsequent events (the ‘de-constitutionalisation’ in the framework of the Lisbon Treaty) have resulted in any modification of the multilevel theory.

D.  Interdisciplinary Nature What is the nature of this theory? Does it belong to the realm of law, social science or political science? This question is far from easy to answer.32 For example, in one of the presentations of this theory, the ‘European Constitution’ (in the wide sense) is made up of more than two levels.33 It comprises a local, regional, national and a European level.34 When a ‘level’ is defined as platform for institutionalised decision making, this presentation does not seem to pose problems since as a matter of fact the daily life of a citizen is influenced by political and legal decisions taken at each of these levels. As a result, the lawyer is confronted with a system of hierarchical and heterarchical levels. In the framework of a national state the relation between the local, regional and national level is always of a hierarchical nature. This is even the case in a federal structure and in unitary states with autonomous regions since the powers of these levels find their origin in the national Constitution. On the other hand, the relation between the national and 29   In many contributions ‘European multilevel constitutionalism’ is used as a paradigm, although it is not always clear whether the authors accept all of its implications; see eg A d’Atena, Costituzionalismo multilivello e dinamiche istituzionali (Torino, Giappichelli, 2007) (prefazione). 30  See M Jestaedt, ‘Der Europäische Verfassungsverbund: verfassungstheoretischer Charme und rechtstheoretische Insuffizienz einer Unschärferelation’ in R Krause, W Veelken, K Vieweg (eds), Recht der Wirtschaft und der Arbeit in Europa. Gedächtnisschrift für W Blomeyer (Berlin, Duncker & Humblot, 2004), 637, also published in C Callies (ed), Verfassungswandel im europäischen Staaten- und Verfassungsverbund (Tübingen, Mohr Siebeck, 2007), 93; C Schönberger, ‘Die Europäische Union als Bund; zugleich ein Beitrag zur Verabschiedung des Staatenbund-Bundesstaat-Schemas’ (2004) 129 Archiv des öffentlichen Rechts 81; M Nettesheim, ‘Europäischer Verfassungsverbund? Zwischen Selbststand und Amalgamierung der Verfassungsrechtlichen Grundordnungen in Europa’ in O Depenheuer, M Heintzen, M Jestaedt (eds), Staat im Wort. Festschrift für Joseph Isensee (Baden-Baden, Nomos, 2007) 733. 31   See on this tendency R Barents, Het Verdrag van Lissabon. Achtergronden en commentaar (Deventer, Kluwer, 2008) § 79–81. 32   From Pernice (2004) n 4, it seems that the legal component of the multilevel theory is dominant. 33   Whether the ECHR is conceptualised as a part of the European level or a separate level is unclear. In I Pernice and R Kanitz, ‘Fundamental Rights and Multilevel Constitutionalism in Europe’ (2004) 7 WHI-Paper 1, 11 the position of the ECHR is described as ‘complementary’. 34   Pernice (2002), n 3 at 512.

The Fallacy of Multilevel Constitutionalism  159 the European level is ex hypothesi of a non-hierarchical nature. This example illustrates the complexity of the multilevel theory for lawyers. What may constitute an acceptable paradigm for social or political scientists – the distinction of various decision-making levels inside and outside the state – is not, at least not at first sight, self-evident for the lawyer who may argue that from the point of view of constitutional and international law a distinction has to be made between the levels within and outside the state.35 However, what matters for the purpose of the present paper is that the multilevel theory, whether or not of an interdisciplinary nature, pretends to offer ‘a theoretical approach to explaining how the Union can be conceptualised’.36 As a consequence, it must be possible with the aid of this theory to explain the contents, nature and evolution of the essential features, principles and rules of the Union as contained in the European Treaties and, in particular, whether the various theses of the multilevel theory referred to above are validated by the developments in Union law resulting from the Lisbon Treaty.

II.  The Unity in Substance Thesis A. Summary The first element of the multilevel theory is the unity in substance thesis. According to this thesis both Union law and national law must be understood as one part of one body of law applicable to the Member States and individuals in the Union.37 Despite their formal distinction, national constitutions and the European Treaties are perceived as a unity in substance and a coherent institutional system, within which competence for action, public authority or the power to exercise sovereign rights is divided among two or more levels.38 The unity in substance of Union law and national law forming one composite system is expressed by the judicial, administrative and legislative powers of the Union and the Member States, which together exercise European authority.39 This thesis is presented as a monist approach40 or a ‘Gesamtordnung’ (‘overall order’), representing a horizontal structure of national legal orders and the Union legal order.41 35   See, on the use of interdisciplinary concepts in European law, A Arnull, ‘The Americanization of EU Law Scholarship’ in A Arnull, P Eeckhout, T Tridimas (eds), Continuity and Change in EU Law. Essays in Honour of Sir Francis Jacobs (Oxford, Oxford University Press, 2008) 415. 36   Pernice (1999), n 3 at 707; Pernice (2002), n 3 at 512. 37   Pernice (1999), n 3 at 712. 38   Ibid, at 706. 39   Ibid, at 724. 40   Ibid, at 712; Pernice (1998B), n 4 at 43. A pluralist approach is explicitly denied; see I Pernice, Das Verhältnis europäischer zum nationalen Gerichten im europäischen Verfassungsverbund (Berlin, WHI, 2006) 54. 41   Pernice (2005), n 4 at 18.

160  René Barents

B.  Possible Meanings It must be kept in mind that the ‘European Constitution’, although presented as an overarching structure, is entirely a theoretical concept. As referred to above, this concept relates to the combination of national constitutions and the European Treaties, which together are perceived as a unity in substance. There is no document or act establishing this structure.42 Accordingly, the unity in substance thesis raises the question as to its subject-matter: unity of what? According to one presentation the composite legal system is constituted by Union law and national law. However, elsewhere this supposed unity is presented in more restrictive terms by referring to a unity in substance of national constitutions and the primary law of the Union.43 While in the latter presentation the unity seems to relate to a common system of constitutional law or constitutional principles (‘gemeinsames Verfassungsrecht’44), in the former it seems to concern an integrated or even a single legal system comprising primary and secondary EU law as well as 27 systems of national law. This lack of clarity is regrettable since, as multilevel constitutionalism pretends to be a theory, one might expect that its subject-matter is defined or at least delimited in a clear manner, in particular in view of a proper understanding of what is understood by ‘unity in substance’ (‘materielle Einheit’).45 The use of the expression ‘unity in substance’ in relation to concepts such as ‘legal’ system, ‘composite system’ or ‘Gesamtordnung’ seems to indicate that ‘substance’ refers to the scope of the ‘European Constitution’, that is, its personal (subjects), geographical (territory) or material (subject-matter) scope. Since the personal scope of the ‘European Constitution’ comprises the peoples of the Member States, unity in substance would then imply that each citizen is simultaneously subjected to the legal rules of 28 (different) systems: the European legal system and 27 national legal systems which, as referred to above, together constitute a composite system. In view of the absurdity of such a result, one may reasonably suppose that this is not what is meant by unity in substance. Next, unity in substance could refer to unity in a geographical sense. This raises the difficult question where the frontiers of Europe are drawn. However, the mere fact that legal systems each cover a part of a certain (European) territory cannot constitute 42   The question whether the TEU actually ‘allocates’ powers to the Union and to the Member States is not considered in this paper. According to Article 5(1) TEU the limits of the powers of the Union are governed by the principle of conferral. During the Convention Giscard in vain proposed to refer to the limits of the powers of the Union and of the Member States. See V Giscard d’Estaing, ‘The Convention and the Future of Europe: Issues and Goals’ (2003) 1 International Journal of Constitutional Law 346, 348. 43   Pernice (2005), n 4 at 17. However, this supposes that all primary Union law has a constitutional character. This seems difficult to accept, for example the substantive rules in the ex EC-Treaty on agriculture or transport. 44   See, inter alia, P Häberle, Europäische Verfassungslehre (Baden-Baden, Nomos, 2002). 45   What is not explored in this paper is the relation between the unity in substance thesis and the theories of Heller and Smend, which seem to have influenced the thinking of Pernice, see above n 23, since these theories are hardly known to non-German scholars. For an overview see B Grzeszick, ‘Staat, Verfassung und Einheit der Rechtsordnung’ in Depenheuer et al (eds), n 30 at 93 et seq.

The Fallacy of Multilevel Constitutionalism  161 a valid reason to treat these systems as a composite unity.46 Although it is possible to approach the Union (and some of its neighbours) as a ‘European legal area’ or even a ‘European constitutional area’, these qualifications are mainly descriptive. In this respect it is interesting to note that the multilevel theory has not attempted to substantiate the unity thesis by having recourse to the concepts of internal market and economic and monetary union, since on the basis of these concepts it can be argued that Union law constitutes a unity in substance and, as a consequence, an autonomous system of law.47 However, for the purpose of the multilevel theory this would not be sufficient since the unity of Union law does not automatically lead to the conclusion that there exists an intrinsic unity between Union law and the law of all Member States.48 It thus seems that ‘unity of substance’ relates to the subject-matter of the European Treaties and of national constitutions or national legal systems. A comparison between these constitutions and systems is certainly possible as far as it concerns the constitutional and institutional aspects (structure, organisation and powers of the public entities involved, legal protection, human rights, etc.), certain values (Article 2 TEU) and, more generally, common convictions on the relation between the citizens and public power. On the other hand, a striking difference between the European Treaties and national constitutions can be noted with respect to the substantive law provisions, for example the rules on freedom of movement and competition. However, from a theoretical perspective, the mere fact that two or more legal systems provide for rules with respect to an identical or similar subject-matter or express similar values does not justify the conclusion that these legal systems constitute one entity. Neither is this the case if ‘unity in substance’ is interpreted as the existence of common constitutional principles or ‘ius commune europeum’. Although these and other factors enable or even require a ‘communication’ between the (ex hypothesi) autonomous systems,49 these common legal and meta-legal elements are not sufficient to consider them as automatically constituting one composite legal system. The convergence or even merger of national economic and social systems does not necessarily lead to intrinsic legal unity.50 Moreover, is it not the very function of national constitutions to express and to maintain the peculiar features inherent to a national identity?51 46   See C Harding, ‘The identity of European law: mapping out the European legal space’ (2000) 6 European Law Journal 128. 47   See Barents (2004), n 19 at 220 et seq. 48   It is also astonishing that the multilevel theory does not refer to Art 220 EC (now Art 19 TEU), according to which the Community judicature has ‘to ensure the observance of the law in the EC’. According to the court’s case law, the concept of ‘the law in the EC’ is wider than Community law. See Barents (2004), n 19 § 357 et seq. 49   Leaving aside the question how it is possible for two or more autonomous legal orders to constitute one single legal order (see below). 50   Unless of course one takes the view that law is exclusively a product of economic and social conditions. However, one may safely assume that such an essentially Marxist approach is not followed by the multilevel theory. Moreover, such an approach would render the theory superfluous. 51   See, inter alia, A von Bogdandy (ed), Handbuch Ius Publicum Europeum, Bd. I Grundlagen und Gründzüge staatlichen Verfassungsrechts (Heidelberg, CF Müller, 2007) 772 et seq. This point is forcefully made by F Mélin-Soucramanien, ‘La Constitution, le juge et le “droit venu d’ailleurs”’ in Mélanges

162  René Barents In more practical terms one may wonder how the unity thesis can provide an explanation for the growing tendency in the primary law of the Union to differentiate its scope between various Member States.52 Since certain parts of Union law, such as the provisions on the economic and monetary union, the Schengen acquis and the common foreign policy, do not apply to all Member States, it is not possible to argue that Union law and national law constitute a unity in substance.53 But even if unity in substance is taken to mean a unity of certain fundamental principles (human rights, for example), the multilevel theory cannot explain why, by virtue of a protocol attached to the Lisbon Treaty, the Charter of Fundamental Rights of the Union is exempted from the jurisdiction of the Court of Justice and national courts in the UK and Poland.54 Admittedly, it can be argued that there is a kind of political unity in the Union, as expressed by the (empty) slogan ‘unity in diversity’. However, if that approach is taken, it still has to be explained what the unity in substance is about. If a number of different elements are considered to be a system, the qualification ‘system’ has only a formal or descriptive meaning.55 Moreover, a political slogan cannot, as such, constitute an appropriate basis for a legal theory.

C.  The Concept of Integrated Legal Order The question thus remains what is meant by the proclaimed unity in substance of (constitutional) Union law and national (constitutional) law as the composite parts of the ‘European Constitution’. Since the expressions ‘unity in substance’, ‘one composite system’ and ‘one body of law’ are elaborated by a reference to (a form of) monism, it seems likely that the subject-matter of the multilevel theory relates to or is based on the hypothesis of the existence of an integrated legal order,56 traces of which can be found in the case law of the ECJ.57 The paradigm of the integrated legal order was developed by the ECJ in order to overcome the difficulties encountered in its early case law by explaining the relation between Community law and national law in terms of pluralism, i.e. the existence of two separate legal orders each of which is autonomous within its own scope.58 To eliminate these difficulties, the court argued that the ‘new’ legal order en l’honneur de Slobodan Milaccic. Démocratie et liberté: tension, dialogue, confrontation (Brussels, Bruylant, 2007) 177, 182. 52   This seems to be recognised implicitly in Pernice (1998), n 4 at 48. 53   The same problem arises with respect to non-Member States which have taken over large parts of the EU acquis, such as Norway, Iceland and Liechtenstein (EEA) and Switzerland. 54   Moreover, the attempt to ‘hide’ the Charter through a mere reference to its existence in Art 6 TEU is difficult to explain in terms of unity in substance created by human rights. 55   Eg, apples, pears and bananas are fruits. 56   Pernice (2007), n 4 at 62. 57   Case 6/64 Costa v ENEL [1964] ECR 614; Case 14/68 Walt Wilhelm [1969] ECR I, para 6. 58   Case 1/58 Stork v High Authority [1958] ECR 63; Case 6/60 Humblet v Belgium [1960] ECR 1167, 1173; Case 13/61 De Geus [1962] ECR 92, 103 ECR 1167, 1173. The problems raised by this approach are outlined in CF Ophüls, ‘Grundzüge europäischer Wirtschaftsverfassung’ (1962) 124 Zeitschrift für das gesamte Handelsrecht und Wirtschaftsrecht 136.

The Fallacy of Multilevel Constitutionalism  163 established by the EC Treaty has been ‘integrated’ into the national legal orders, in other words a kind of fusion of legal orders.59 Attractive as this paradigm may seem at first sight, it nevertheless poses serious theoretical problems.60 First of all, ‘integration’ may be perceived, as in the multilevel theory, as the Constitution of one single legal order composed of European law and national law (‘melting pot’).61 However, this would mean that the territory of each Member State is also subject to the law of all other national legal orders. Such a situation is not even the case in a federal state. On the contrary, ‘integration’ in this sense amounts to nothing less than one, single European order for all Member States and their citizens, which in practical terms means a European state. Such a nightmare is difficult to imagine. Secondly, integration could be interpreted as a partial fusion of legal orders. Union law is then integrated into that part of the national legal order that falls within the scope of the Treaties. As a result, the national legal order would be divided into two parts: one truly national part and one Union part or one independent and one dependent part. However, a completely divided national legal order is a contradiction in terms. A legal order is one or it does not exist. Moreover, this interpretation has been rejected by the ECJ.62 Thirdly, integration could mean that the European legal order and each national legal order are transformed into one (mixed) ‘national-European’ legal order. In view of the differences between national legal orders and their territorial limitations, this would amount to 27 separate ‘mixed’ legal orders. Union law and thus the Union legal order would then be partitioned in a geographical sense, a result which runs counter to the very concept of a substantive EU legal order as developed in the ECJ case law.63 These brief observations demonstrate that the court’s doctrine on the integrated legal order cannot solve the theoretical problem of the relation between Union law and national law. As such this is not surprising, since – but this is often overlooked – the concept of ‘legal order’ is typically state oriented and therefore closely linked to the presumed formal or substantive unity of its legal system.64 Since the sole intention of the paradigm of the integrated legal order is to provide for a uniform application of Union law in the national legal orders,65 ‘integration’ in reality means ‘penetration’ or ‘invasion’ of Union law into these orders by 59   Elsewhere I have argued that monist approaches cannot provide for a satisfactory explanation of the relation between Union law and national law; see Barents (2004), n 19 at § 204 et seq. 60   Ibid, at 181 et seq. 61   Pernice (1998), n 4 at 64. 62   Case 33/67 Kurrer [1968] ECR 179, 193; Opinion 1/78 Rubber [1979] ECR 287, para 62. 63   See Case 166/73 Rheinmühlen [1974] ECR 1. 64   Barents (2004), n 19 at § 213. 65   That this was indeed the intention of the ‘integration’ formula of the ECJ is explained in the various writings of the members of the ECJ; see P Pescatore, ‘Droit communautaire et droit national selon la jurisprudence de la Cour de justice des Communautés européennes’ (1969) Recueil Dalloz Sirey 179, 180 and N Catalano and R Monaco, ‘Le problème de l’applicabilité directe et immédiate des normes des traités instituant les Communautés européennes’, in 2ième Colloque international sur le droit européen (Zwolle, Tjeenk Willink, 1968) 9, 10.

164  René Barents virtue of its own (original) authority, whether that authority is formally recognised by the Member States or not.66 This ‘unilateral’ effect of Union law (its autonomy) is exactly the opposite of integration; indeed it points to a model of pluralism. Admittedly, for the authority and effectiveness of Union law this paradigm of the ECJ has proved to be highly successful. However, while paradigms, artificial or not, may play a useful role in case law – ‘the life of the law has not been logic, it has been experience’, as OW Holmes once expressed it67 – for the development of an academic theory such a way of reasoning is not sufficient.68 According to the multilevel theory, the unity in substance of the various levels of the ‘European Constitution’ is confirmed by the principle of mutual recognition which, according to the ECJ’s case law, underlies the concept of free movement. But does this effect of the prohibitions to restrict free movement automatically result in a unity in substance of national legal systems to the effect that they constitute one body of law? The effect of the principle of mutual recognition (country of origin principle) is twofold. On the one hand the legal effect of national rules and decisions is extended from the national territory to the territory of all other Member States, while on the other the application of national law on the national territory is precluded or restricted. In other words, the principle of mutual recognition embodied in the provisions on free movement amounts to a shift in the territorial scope of national law. Its effect is not that the rules and decisions of one system of law become an integral part of the legal system of all other Member States. Accordingly, the principle of mutual recognition and its effects can also be explained in terms of pluralism: one territory is subjected to the law flowing from different and independent sources.69

D.  The Exercise of European Authority According to the multilevel theory the existence of a composite system or unity in substance of Union law and national law is reflected by the system of decentralised implementation of Union law. By implementing Union law, national admin66  See, inter alia, J Mertens de Wilmars, ‘De rechtsbescherming in de EG’ (1963) Rechtskundig Weekblad 1418; HP Ipsen, ‘Über Supranationalität’ in H Ehmke, U Scheuner, J H Kaiser, W A Kewening (eds), Festschrift für Ulrich Scheuner zum 70. Geburtstag (Berlin, Duncker & Humblot, 1974) 211, 222; R Kovar, ‘La contribution de la Cour de justice à l’édification de l’ordre juridique communautaire’ (1995) 4 Collected Courses of the Academy of European Law 15, 47 et seq; L Dubouis, ‘Le juge français et le conflit entre norme constitutionnelle et norme européenne’, in L’Europe et le droit; mélanges en hommage à Jean Boulouis (Paris, Dalloz, 1991) 205–19; D Simon, ‘Les fondements de l’autonomie du droit communautaire’, in Droit international et droit communautaire, perspectives actuelles, (Paris, Pedone, 2000) 207, 247. 67   OW Holmes Jr, The Common Law (1881) 1. 68   It is therefore not surprising that the court has not elaborated the concept of ‘integrated legal order’ any further by indicating the legal basis of the ‘integration’ (the Treaty or national constitutions). The court’s doctrine on the ‘new’ or ‘integrated’ legal order is essentially an ideological postulate, based on what in the judgment in Case 26/62 Van Gend en Loos [1963] ECR 1 is described as a ‘view’; see Barents (2004), n 19 at § 213. 69   Ie national law, Union law and by virtue of Union law also the law of other Member States.

The Fallacy of Multilevel Constitutionalism  165 istrations, courts and legislatures exercise European authority and are acting as ‘agents’ of the Union.70 Obviously, in factual terms this state of affairs can be described as an ‘integrated system’.71 However, the mere existence of multiple legal links between national and EU administrations does not necessarily imply ‘unity in substance’ between national legal systems among themselves and between those systems and that of the Union. Such a situation can also be explained in terms of ‘dédoublement fonctionnel’, that is, one body is acting on the account of different sources of authority.72 In any event, the concept of one integrated European authority for the implementation of Union law cannot be deduced from the ECJ’s case law. On the contrary, the court has explicitly stated that by implementing their Treaty obligations (ex-Article 10 EC), national authorities do not obtain the status of Union organs and that the implementing acts do not belong to Union law.73 Particularly inconsistent is the argument on the existence of a unity between the European and national administration of justice. In this respect it is argued that while according to the case law of the ECJ national courts are considered to be ‘EU courts’, as far as the German context is concerned, according to the Bundesverfassungsgericht, the ECJ takes the position of a ‘geseztlichen Richter’ (‘statutory court’) within the meaning of Article 101 of the German Basic Law.74 Undoubtedly national courts act as EU courts when they apply EU law. However, they remain a part of the national judiciary.75 The court has stressed over and again that the relation between the EU judicature and national courts in the framework of Article 267 TFEU (ex-Article 234 EC) is based on the cooperation of courts, each of which acts according to their own responsibilities. Accordingly, while Article 267 TFEU provides for a direct relation between the national and the EU judicature, the Union has no unitary court system for the administration of justice. The reference to Article 101 Grundgesetz even more striking. The message behind this qualification is that German courts by not fulfilling their obligations under Article 267 TFEU also violate national constitutional law. Undoubtedly this qualification reinforces the practical effect of the preliminary ruling procedure in Germany. However, whether or not the ECJ has the status of a court within the meaning of the German Basic Law is completely irrelevant for compliance with Article 267 TFEU.76 Instead, at the background of old conceptions of dualism, the qualification based on Article 101 Grundgesetz   Pernice (1999), n 3 at 724.   References of this kind can be found in the ECJ’s case law, see, inter alia, Case 155/79 AM&S [1982] ECR 1575, § 18; Case C-446/04 Test Claimants in the FII Group Litigation [2006] ECR I-11753, para 170. 72   See, in particular, Case 158/80 Rewe-Markt Steffen [1981] ECR 1805, para 44 and Case C-432/05 Unibet [2007] ECR I-2271, para 44. 73   See, inter alia, Case C-97/91 Oleifici Borelli [1992] ECR I-6313; C-6/99 Greenpeace France [2000] ECR I-1651; Case C-366/00 Huber [2002] ECR I-7736. 74   Case 2 BvR 197/83 Solange II [1986] BVerfGE 73, 339. 75   See extensively M Claes, The National Court’s Mandate in the European Constitution (Oxford, Hart Publishing, 2006). 76   As is illustrated by the ECJ’s case law on what constitutes a ‘court’ or ‘tribunal’ in the sense of Art 267 TFEU. 70 71

166  René Barents constitutes but a unilateral measure by the German legal order to ‘accept’ the interaction with the EU legal order.77 It thus appears that whatever ‘unity in substance’ may be, in any event it does not relate to unity of law of the European and national levels.78 The mere fact that Union law and national law demonstrate considerable converging tendencies and that between these orders a multitude of factual and legal links exist does not necessarily lead to the conclusion that these systems constitute a unity in substance. The same tendencies can also be explained in terms of pluralism. It might be possible that unity in substance relates to other than legal factors (economic, political or social unity), but even if this can be demonstrated, it does not justify the conclusion as to the existence of unity in law between both levels. Accordingly, the conclusion is that, from a theoretical point of view, one of the pillars of the multilevel theory, the thesis of unity in substance of national and Union law or the single, integrated legal order, is false.

III.  The European Citizenship Thesis A. Summary One of the cornerstones of the multilevel theory relates to the foundation of the existence and functioning of the Union. Between the Union and the citizens a direct connection exists in the sense that its origin is derived exclusively from the common will of these subjects.79 This argument goes at follows. First, Member States are or must be democracies.80 Second, being democratic, the state is supposed to act always on behalf of the people.81 Third, therefore, the establishment of the Union by the Member States reflects the common will of the people(s). As a consequence, the existence and functioning of the Union directly reflects the will of the peoples or citizens as much as the Member States are a matter and creature of their respective citizens. Another consequence is that each citizen is deemed to have more than one political identity, at least a ‘national’ and a ‘European’ identity.82 This reasoning is elaborated by a reference to the social contract theory of

77   Therefore one could even argue that the reasoning of the Bundesverfassungsgericht is contrary to EU law. 78  See also M Nettesheim, ‘Europäischer Verfassungsverbund? Zwischen Selbststand und Amalgamierung der Verfassungsrechtlichen Grundordnungen in Europa’ in Depenheuer et al (eds), n 30 at 733, 738 et seq. 79   Pernice (1999), n 3 at 710; Pernice (2005), n 4 at 99. 80   The multilevel theory does not indicate the consequence if a Member State does not fulfil this requirement. This problem is not solved by pointing to the sanction procedure provided for in Art 7 TEU. 81   Pernice (2005), n 4 at 102. 82   Pernice (1998), n 4 at 44, 47, Pernice (2005), n 4 at 707, Pernice (2007), n 4 at 85 et seq.

The Fallacy of Multilevel Constitutionalism  167 Rousseau and to the European citizenship.83 Following the contract theory of public power, it is argued that while on the basis of a national social contract citizens have constituted ‘their’ state, on the basis of an existing or emerging European social contract these citizens have decided to manifest their general will or a part of it through the Member States to establish ‘their’ Union.84 The existence of a European social contract is expressed by the notion of European citizenship in the sense that the citizens by proclaiming that status have declared themselves to be subjects of Union authority.85 The final result of this construction is that the citizens directly legitimise the exercise of public power at two levels.

B. Fiction The citizenship thesis raises the question whether it concerns a hypothesis, to be validated by legal or other arguments or a political and/or legal fiction. The preceding presentation strongly points in the direction of a fiction. The fiction seems to be that at one point in time the citizens have acted as a ‘pouvoir constituant’ by constituting ‘their’ state and, at another point in time, by constituting ‘their’ Union. The parallel between the ‘national’ constitutional moment (or process) and the ‘European’ one demonstrates that the citizenship thesis is essentially a matter of democratic ideology. Although the result is entirely different, this line of reasoning is comparable to the fictional doctrine of ‘Volk-Staat-Verfassung’ (people, state, Constitution) which in most of the Member States constitutes the meta-legal foundation of the modern constitutional state. For the Union the reasoning seems to be ‘European citizens – Union – European Treaties’.86 It is striking that while the multi­level theory claims to offer an alternative to the ‘Volk-Staat-Verfassung’ doctrine, this alternative is based on a similar fiction-based methodology. The first problem, by introducing fictions and doctrines in what should be a rational debate, is that it renders any theory developed with the aid of such elements impervious from criticism. For example, even the most fervent defender of the European cause may raise his eyebrows by reading that in 1993 the citizens of the Member States have decided to award themselves (through their Member States) the political status of European citizenship. If that decision finds its origin exclusively in the will of the citizens, the obvious question is why they have contended themselves with a largely formal status of their citizenship. Indeed, in spite of some positive developments in the case law, the concept of European citizenship in its present form is far removed from its conceptual essence which, as Advocate General Jacobs has stated, is that all Union citizens should be entitled to say ‘civis europeum sum’.87 Of course, it can be argued that it was (or is) the will of the   Pernice (1999), n 3 at 709.   Pernice (1998), n 4 at 44, 47, Pernice (1999), n 4 at 715, Pernice (2007), n 4 at 66.   Pernice (1999), above n 4 at 207. 86   See also T Schmitz, ‘Das europäische Volk und seine Rolle bei einer Verfassungsgebung in der Europäischen Union’ (2000) Europarecht 217, 219–20. 87   § 46 of his opinion in Case C-168/91 Konstantinidis [1993] ECR I-1191. 83 84 85

168  René Barents citizens to start with an embryonic form of European citizenship, to be fleshed out at a later stage by subsequent decisions or expressions of will by the same citizens. However, possible future developments, whether likely or not, cannot constitute a sound basis for the formulation of a theory. Moreover, even the ‘constitutional moment’ of Europe, the adoption of the draft European Constitution by the European Convention, did not result in any effort to substantiate the concept of European citizenship. In the end, therefore, the position of European citizenship in the multilevel theory is a matter of belief or even wishful thinking.

C.  Ex Post Facto Explanations The next problem raised by the fiction of European citizenship as the source or the confirmation of a European political identity88 is that it explains all possible outcomes of the various political processes at the European level. Indeed, since the result of any intergovernmental conference is perceived (ex hypothesi) as the expression by the Member States of the political will of the citizens acting in their European identity, any result whatsoever is always legitimate. This fiction poses no problem when an amendment to the European Treaties or a new European Treaty is ratified by the national parliaments or approved by the population through a referendum. However, if a national parliament refuses to ratify such a proposal or when it is rejected by a referendum, the conclusion can only be that the Member State(s) have not expressed the will of the citizens and, as a consequence, that they have acted undemocratically. In other words, at any given moment the situation at the European level always reflects the will of the citizens. However, a theory that explains everything explains nothing. Indeed, the situation would be not different if instead of referring to the will of the citizens a reference would be made to divine will! Accordingly, the fiction of direct democratic legitimation at the European level has no added value since it only explains the status quo. As such, this is not surprising. The explanation of a political situation as the result of the general will is always an explanation ex post facto. The preceding observation also shows that the fiction of the (partial) European identity of the citizens is inherently flawed. By its very nature, since it presupposes the existence of a European identity, the citizenship thesis cannot explain its own rejection. The various referenda on the amendments of the European Treaties since 1986 and in particular the negative votes of the French, Dutch (2005) and Irish (2007) population indicate that when given the opportunity to express themselves, the citizens do not accept their subordination to Union governance. Instead, by expressing themselves directly in a democratic manner, they have exercised an essential democratic right: the right to self-determination in the framework of a national social contract.89 Accordingly, following the direct legit­   It is not clear what constitutes the cause or the effect.  See, inter alia, C Shore, ‘Government without Statehood’? Anthropological Perspectives on Governance and Sovereignty in the European Union’ (2006) 12 European Law Journal 709, 718. 88 89

The Fallacy of Multilevel Constitutionalism  169 imation argument, the process to ‘deconstitutionalise’ the Constitution for Europe by, inter alia, the deletion of the primacy clause and the symbols identifying the direct legitimation of the Union by its citizens (flag, hymn, etc.) must also be seen as the reflection of the will of the citizens. In other words, the will of the European citizens is not to have a common European will. This seems to be a paradox, to say the least. These events as well as the desperate attempts of Member States to heal the breaches caused by the consultations of their citizens indicate that assumptions about ‘post-national democracy’, ‘post-national citizenship’ or ‘post-national sovereignty’ are (still) extremely fragile and that, accordingly, utmost care should be taken by developing (constitutional) theories on such bases.90 Indeed, there is still no definite answer to one of the essential questions involved in post-national constitutionalism: whether a citizen can actually have two or more political identities or, to repeat old wisdom, whether he can serve two masters.91 The violent dissolution of Yugoslavia, the near end of the Kingdom of Belgium, Scottish ‘independence’ as well as the delicate position of minorities in some of the new Member States all seems to indicate the contrary. These unpleasant observations are supported by the life and times of the European Convention and its draft Constitution for Europe. In numerous publications this body, in particular because of its composition, was considered to be a kind of ‘(semi) European pouvoir constituant’ or a manifestation of deliberative democracy, securing an independent place for the European citizen in the construction of a European constitutional entity.92 However, its functioning as well as its result showed a dramatic absence of any substantial democratic influence and public awareness.93 These as well as many other developments on the European level raise serious questions on whether constitutional theories about the present and future structure of the Union can still be based on concepts of democracy, multiple political identities, ‘constitutional tolerance’, ‘multiple demoi’, etc.94 All these theories are based on the fragile assumption that given the existence of post-national constitutional phenomena such as the Union, it must, as a matter of principle, be possible to explain these phenomena with the aid of ‘democratic’ tools and concepts. However, this is but a normative assumption and not necessarily a validated 90   See R Barents, De Grondwet voor Europa. Achtergronden en commentaar (Deventer, Kluwer, 2005), § 147. 91   See, inter alia, M Nettesheim, ‘Decision-making in the EU: identity, efficiency and democratic legitimacy’ (2004) European Review of Public Law 197, 201 et seq. 92  See, inter alia, M Beyer, Der Konvent zur Zukunft Europas. Deliberativ-demokratische Verfassungsgebung für die Europäische Union – Zugleich ein Beitrag zum Begriff der Verfassungsgebung (Baden-Baden, Nomos, 2007). 93   See, inter alia, I Ward, ‘Bill and the Fall of the Constitutional Treaty’ (2007) 13 European Public Law 461. 94   The intention of these arguments is not at all to deny the hypothesis that the European Treaties may have an original foundation, ie that their existence does not necessarily derive exclusively from the will of the Member States. However, this does not require a normative democratic foundation. See R Barents, ‘Hoe constitutioneel is de Grondwet voor Europa’ [How constitutional is the European Constitution] (2004) Sociaal-Economische Wetgeving 1.

170  René Barents hypothesis. In other words, the emergence of post-national constellations might require us to think the unthinkable, that is, that both in practice and in theory actual democratic legitimation of constitutional developments beyond the state is not possible.95 These politically incorrect observations demonstrate what Loughlin and Walker have aptly called ‘the paradox of constitutionalism’: whether, as Hobbes has put it, the minds of the people ‘are like clean paper, fit to receive whatsoever by public authority shall be imprinted in them’ or if they can be considered to be ‘agents of active (constitutional) change’.96

D.  The Function of Member States Finally, in the fiction that present and future developments at the level of the Union can all be explained in terms of the will of the citizens, the function of Member States becomes purely formal. Since the citizens are able to express their will to constitute a post-national structure only through the Member States, the latter are reduced to a mere means of transmission. In fact, state and national citizens become identical in the same way as the Union and European citizens are identical, since they constitute but a product of the will of the citizens to organise society. Apart from the fact that this presentation amounts to a gross distortion of the political and legal reality with respect to the standing of states in the Union,97 it fails to explain the past and present institutional structure of the Union. In particular, the multilevel theory cannot explain why the organisation of public power at the Union level is substantially different from that at the national level. If the Union is seen as an exclusive manifestation of the will of the citizens, the European Parliament must necessarily be considered to be the supreme organ of the Union. There would be no place for institutions like the European Council or the Council, whose function (according to the Treaties) is to represent the governments of the Member States. At most, national interests would be represented in a ‘chamber’ of the European Parliament, comparable to the Senate, the Bundestag or the House of Lords. If there is one common will of the citizens at the European level, there is no objective reason why this will has to be represented by three institutions which, as the case may be, are competing. If the existence and functioning of the Union are derived entirely from the will of the citizens, the Union would, in general terms, be featured by a kind of ‘trias’ structure. However, since its inception through the European Steel and Coal Community, the Union is featured by its

95   See, inter alia, A Hurrelmann, ‘European Democracy, the ‘permissive’ consensus and the collapse of the EU Constitution’ (2007) European Law Journal 343. 96   M Loughlin and N Walker, The Paradox of Constitutionalism. Constituant Power and Constitutional Reform, (Oxford, Oxford University Press, 2007) 2. 97   See, in particular, P Eleftheriadis, ‘The standing of states in the European Union’ in N Tsagourias (ed), Transnational Constitutionalism. International and European Models (Cambridge, Cambridge University Press, 2007) 44.

The Fallacy of Multilevel Constitutionalism  171 quadri-partite institutional structure, intended to represent the will of the Member States as well as of their peoples.98 When the citizenship thesis is confronted with the details of the institutional structure of the Union, numerous questions cannot be answered. For example, how is it possible to explain the requirement of double qualified majorities for the decision making in the Council (a percentage of the Union population and a percentage of the number of Member States)? It is difficult to perceive the codecision procedure as an exclusive expression of two common wills, one national (through the Council) and one European (through the European Parliament), while in the third lecture the conciliation committee, as an expression of both wills, has to find a balance between the two in case of a conflict.99 According to the multilevel theory the adoption of a Union directive expresses the will of the European citizens. If so, why is it necessary that a directive has to be implemented by a parliamentary act expressing the will of the citizens at the national level? Moreover, without elaborating this any further, the view that the function of states for the purpose of building a post-national community with constitutional features is exclusively restricted to the conclusion of Treaties conflicts with insights on the existence and nature of international constitutional law. In this perspective Habermas has cogently argued that states, together with their citizens, should consider themselves as the constituent part of an international constitutional community.100

E.  The Social Contract Theory The multilevel theory is based on or inspired by the social contract ideology of Rousseau.101 In view of the totalitarian nature of this ideology this is surprising. According to Rousseau the social compact implies that individuals must be ‘forced to be free’.102 Only by being made to conform to the general will is it possible for man to be liberated from the original sin: the invention of property, which constitutes the ‘fall from grace’ out of the ideal state of nature. For Rousseau the social compact refers to the covenant between individuals to constitute a collective body, the ‘corps moral et collectif’, representing the perpetual and intangible nature of the general will of the people or nation.103 Therefore, according to Rousseau, the social contract unites individuals into a body politic or the sovereign, from which there is no escape. Indeed, as this philosopher argued, even the slightest modification of the contract would render it ‘vain and ineffective’. The body politic thus constituted can never bind itself, even to an outsider, for 98  See P Pescatore, ‘L’exécutif communautaire: justification du quadripartisme institué par les traités de Paris et de Rome’ (1978) Cahiers de droit européen 387. 99   One may even identify a ‘third common will’ in the framework of the control to be exercised by national parliaments on compliance with the subsidiarity principle (yellow and orange carts). 100   See J Habermas, The Divided West (Cambridge, Polity, 2006) 159. 101   See for an English version J-J Rousseau, The Social Contract (eBooks@Adelaide, 2005). 102   Ibid, Book I at § 9. 103   See M Albertini et al, ‘L’idée de nation’ (1969) 8 Annales de philosophie politique.

172  René Barents example by submitting to another sovereign, since that means giving up its independence and capacity to express its general will.104 To argue that citizens are bound at the same time by two social contracts would make Rousseau turn in his grave! ‘There is only one contract in the state, and that is the act of association, which in itself excludes the existence of a second. It is impossible to conceive of any public contract that would not be a violation of the first.’105 These (too) short observations show that the explanation of European citizenship in terms of a new (second) social contract, establishing the Union as a new source of authority, constitutes an outright distortion of Rousseau’s philosophy.106 The citizenship thesis raises the question of what the European citizens have exactly in common, since the creation of the Union as a result of their common political will presupposes the existence of the will to live together within one (partial) system. Ex hypothesi, this common element cannot relate to the existence of a ‘demos’ based on social and political homogeneity since, according to orthodox constitutional ideology, this would lead to the Constitution of a state. Instead, the multilevel theory identifies the common element as ‘common history and experience’,107 in particular the desire for peace and the will of common action to protect and to promote common values, such as those expressed in Article 2 TEU. However, this is a political postulate, a matter of conviction that one can accept or not.108 It seems to suppose, for example, that the history and values of the American people are entirely different from those of the European peoples, in spite of the fact that their political and constitutional philosophies have a common origin and that their societies are highly interdependent. What is the common element in the history of Spain and Estonia? If common history is that important, one may wonder why it proved to be impossible to include in the Constitution for Europe a reference to religion and Christianity, one of the most important common elements of European history.109 Is the desire for peace not a universal desire? Moreover, do common values not express and even presuppose the existence of a certain political and social homogeneity? Finally, what about the Protocol attached to the Lisbon Treaty to exempt the UK and Poland from the jurisdiction of the ECJ and national courts in matters concerning the Charter of Fundamental Rights. As one author has rightly observed, this seems more to be antidote to the promotion of a European identity.110   Rousseau, above n 101, Book II at § 1 and § 6.   Ibid, Book IV at § 16. 106   This is not the place to dwell on the variety of social contract theories, going from Plato to Rawls. However, one may wonder why the multilevel theory did not try to find support in the theories of Locke; see, inter alia, O Zetterquist, ‘A European Social Contract?’ in P Wahlgren (ed), Constitutional Law – Constitutions (Stockholm, Scandinavian Studies in Law, 2005) 322 et seq. 107   Pernice (1999), n 3 at 720. 108   The issue of the uneasy relation between values on the one hand and democracy, fundamental rights and freedoms on the other is not considered here. 109   Barents (2005), n 90 at § 373 et seq; C Joerges, ‘On the Disregard for History in the Convention Process’ (2006) 12 European Law Journal 2. 110   J Baquero Cruz, ‘What’s left of the Charter? Reflections on law and political mythology’ (2008) 15 Maastricht Journal of European and Comparative Law 65, 74. 104 105

The Fallacy of Multilevel Constitutionalism  173

F.  The Union as Creation of Its Citizens However, since the multilevel theory pretends to offer an explanation for what the process of European integration in general and the development of the Union in particular is about, the fictions referred to above must be considered to be hypothe­ses which, if they are valid, must at least be reflected in the text of the Treaties as amended by the Treaty of Lisbon. In other words, the text and system of the Treaties should confirm that the existence and functioning of the Union can be explained exclusively in terms of the will of the citizens while the role of Member States is basically that of transmitting and executing that will. According to the citizenship thesis, the Union level of authority draws its legit­ imacy from the citizens, acting through their Member States. From a formal point of view, this hypothesis is undoubtedly correct since the last consideration of the preamble of the TEU makes clear that the Heads of State have designated their plenipotentiaries to reach agreement on the text. However, since this pronouncement applies to all international Treaties, whatever their form or contents (for example, the Charter of the UN), it has no distinctive value with respect to the foundation of the Union. The most that can be said is that all Treaties concluded by democratic states represent the will of the citizens. As far as the substance of this hypothesis is concerned, it is contradicted by Article 1 TEU. According to this provision, the High Contracting Parties establish among themselves a European Union. Any reference to the states acting on behalf of the citizens is absent. It is true that in Article I-1 of the European Constitution, the foundation of the Union is formulated differently: ‘Reflecting the will of the citizens and States of Europe to build a common future, this Constitution establishes the European Union.’ However, even the latter formula does not confirm the hypothesis that the legitimacy of the Union is exclusively derived from the citizens. On the contrary, citizens and states are presented as two independent actors featured by a common will. It is obvious that in this formula ‘states’ cannot be interpreted as ‘states acting on behalf of their citizens’, since this would render this phrase to a tautology. In this respect it is also worth noting that the introductory quotation in the Constitution for Europe of Thucydides about democracy was deleted by the intergovernmental conference because it did not refer to the states! Seen at this background, it is clear that the draftsmen of the Treaty of Lisbon deliberately omitted any reference to the will of the citizens to establish the Union. Far more important, however, is the fact that both in Article I-1 of the European Constitution and in Article 1 TEU, the competences of the Union are conferred on it ‘by the Member States’.111 If there is any conclusion to be drawn from this in terms of legitimacy, it cannot be anything else than that the Member States in

  And not ‘by the citizens’ or ‘by this Treaty’.

111

174  René Barents their capacity as sovereign states have acted as ‘pouvoir constituant originaire’.112 One could, of course, argue that even the omission of any reference to the will of the citizens is ultimately an expression of will by the same citizens, but this amounts to nothing else than a reductio ad absurdum of the very idea of direct legitimacy of the Union.113 That according to Article I TEU the establishment of the Union is not, even partially, founded on the will of the citizens is confirmed by the provisions in the TEU on the functioning of the Union. According to Article 10(1) TEU it is but the functioning of the Union which is founded on the principle of representative democracy.114 Moreover, the same provision contradicts any idea of the Union drawing its legitimacy exclusively of the citizens. Instead, as far as the functioning of the Union is concerned, it introduces the concept of dual legitimacy: the citizens and the states.115 The first express their will through European political parties. In this respect it is striking that the same formula is not used with respect to the states represented in the European Council and the Council. As observed above, in the perspective of direct legitimacy, there would have been no objective reason to refer to these two institutions. Finally, the citizenship thesis fails to offer an explanation of the procedure for Treaty revision (Article 48 TEU), both in its present form and as laid down in the Lisbon Treaty. Even if it is argued that the ratification by each Member State constitutes a direct expression of the will of the citizens, this procedure is such that, in theory, less than 0.1 per cent of the Union’s citizens (Malta) is able to prevent the will of more than 99.9 per cent of these citizens from being materialised in an amendment of the Treaties. This simple fact as well as the refusal of the Member States to lay down any other procedure refutes the thesis of democratic legitimacy of Treaty amendments in terms of the will of the citizens. Instead, it demonstrates the relevance of the states as constitutional actors.116

G. Conclusion It results from the preceding observations that even if the fiction that the Union can be conceptualised as a creature of its citizens acting through the Member 112   See, inter alia, A Berramdane, ‘Le traité de Lisbonne et le retour des États’ (2008) La Semaine Juridique 122. 113   See on the inconsistencies of the hypothesis of the citizens as the ‘pouvoir constituant’ of the Union E Maulin, ‘Le pouvoir constituent dans l’Union européenne’ (2005) 45 Droits. Revue française de théorie, de philosophie et de culture juridique 73. 114   Moreover, the reference to representative democracy confirms that any legitimation of the Union in terms of a social contract à la Rousseau is excluded, since in his philosophy there was no place for representatives of the people. 115   See, inter alia, A von Bogdandy, ‘Zur Übertragbarkeit staatsrechtlicher Figuren auf die Europäische Union: vom Nutzen der Gestaltidee supranationaler Föderalismus anhand des Demokratieprinzips’ in P Huber, Peter M. Huber, M Brenner, M Möstl (eds), Der Staat des Grundgesetzes - Kontinuität und Wandel: Festschrift für Peter Badura zum siebzigsten Geburtstag (Tübingen, Mohr Siebeck, 2004) 1033, 1046. 116   See also R Hirschl, ‘Preserving hegemony? Assessing the political origins of the EU Constitution’ (2005) 3 International Journal of Constitutional Law 269.

The Fallacy of Multilevel Constitutionalism  175 States is perceived as a hypothesis, its validity is not confirmed by the text and system of the Union as laid down in the Constitution for Europe and, subsequently, in the Treaty of Lisbon. Moreover, as demonstrated, this thesis suffers from inherent contradictions.

IV.  The Autonomy Thesis A. Summary In the multilevel model each level of public power is autonomous.117 ‘Autonomy’ in this context has to be understood as ‘original’ autonomy: the origin and functioning of each level of authority is not dependent on any other level. The absence of hierarchy between the Union and national level of authority results from the common will of the citizens to divide public power over two levels of government. The autonomy thesis constitutes an essential element of the multilevel theory in order to prevent a system of multilevel government from being categorised as a federation and hence a state. Indeed, in most federations the federal Constitution prescribes the priority of federal law over the law of the federated states.118 However, notwithstanding their autonomy, the various levels constitute a single system (‘European Constitution’) or, as set out above, an integrated legal order. This assumption is necessary since, in order to be qualified as a legal system, the simultaneous application of two contradictory legal norms, each flowing from an autonomous source, has to be excluded. It is the overarching system of the European Constitution ‘which must produce ultimately one legal answer to each case’.119 Accordingly, although there is no hierarchy of norms, in case of a conflict Union law nevertheless enjoys priority over national law. This feature results also from the common will of the citizens. If it were otherwise, the equality of the citizens before the law would not be guaranteed and, as a consequence, the ‘European’ social contract between the citizens would be infringed.

B.  Monism or Pluralism? This presentation of the relationship between Union law and national law raises serious conceptual problems. How is it possible for (ex hypothesi) two autonomous legal systems (pluralist model) to constitute, by reason of their presumed unity in substance, a single legal system (monist model) in which the law of one system has priority over the law of the other? In order to explain this problem, it   Pernice (2002), n 3 at 520.   See, inter alia, LM Diez-Picazo, ‘What does it mean to be a State within the European Union’ (2002) XII Revista italiana di diritto publico comunitario 651. 119   Pernice (2002), n 3 at 520. 117 118

176  René Barents is necessary to elaborate briefly the consequences of the approach of law as a system or order, to which it must immediately be added that for this purpose there is need to have recourse to Kelsenian or other ‘Stufenbau’ theories. In terms of law, the autonomy of a level, system or order means that only principles and rules produced or accepted by that system are considered to be law. On other words, such a system is self-referential.120 In an autonomous legal system rules of another system have no legal effect as such. For example, a judgment by a court of system A has no legal effect in system B. This is only the case if system B accepts the legal authority of such a judgment, either unilaterally or by virtue of an agreement with system A. In both cases legal system B remains autonomous since the unilateral acceptance or the agreement can be terminated. In other words, the autonomy of legal system B means that a rule or decision of system A is considered at most as a legally relevant fact. If, on the other hand, a rule or decision of system A would have legal effect in system B as such, that is, because of the mere fact that it is adopted by system A, system B is no longer autonomous, but subordinated to system A. At most it enjoys relative autonomy vis-à-vis system A. If that is the case, system B is no longer the master over the law applicable in its system. Admittedly, two or more autonomous systems of law may be linked by agreements between these systems or they may have other features in common, such as natural factors, history, culture, ethnicity, religion, etc. However, these links do not entail the existence of one legal system characterised by a unity in substance.121 Accordingly, to argue that two or more autonomous systems of law may constitute one single system in which the law of one system has priority over the other is, at least from a legal point of view, a contradiction.

C.  Priority of Union Law If, ex hypothesi, two legal systems are autonomous (self-referential), the law of one system can never enjoy autonomous priority over the law of the other system. Even if the law of system X lays down its priority over the law of system Y, for the latter system this is at most a legally relevant fact. System Y could of course recognise the priority of the law of system X, but in that case the priority of the law of system X finds its ultimate origin in the law of system Y. As observed above, system Y could revoke this priority, for example by adopting an act excluding the legal effect of certain rules of system X or as the result of a ruling by a constitutional court.

120   See, inter alia, N MacCormick, ‘On the very idea of a European Constitution; jurisprudential reflections from the European Parliament’ (2001–02) 13 Juridisk Tidskrift vid Stockholms Universitet 529, 531–32. 121   At most is it possible to use the qualification ‘system’ or ‘order’ in a purely formal sense. However, this qualification applies to all autonomous legal orders mutually linked by whatever kind of agreements. As a result, such a qualification has no distinctive value, it constitutes but a description of an existing situation.

The Fallacy of Multilevel Constitutionalism  177 In order to cut this Gordian knot of pluralism and monism, the multilevel theory argues that it is the common will of the citizens that the law of the auto­ nomous level of the Union has priority over the law of the autonomous national level. However, this ‘democratic’ fiction on the foundation of priority finds no support in the ECJ’s case law. According to well-established case law the priority of Union law over national law follows from the principle of effectiveness, that is, to ensure that Union law is applied in the whole of the Union under the same conditions. Furthermore, this reasoning is contradictory. If, ex hypothesi, both levels are considered to be autonomous, the legal basis of Union law for its priority over national law is totally irrelevant for the national level. Whether the priority of Union law is recognised by the ECJ’s case law, is laid down in an explicit treaty provision (such as Article I-6 of the Constitution for Europe), or finds its origin in a fictional decision of the citizens, for the autonomous national level this makes no difference at all. Accordingly, for the purpose of explaining the relation between Union law and national law, the supposed ‘democratic’ foundation of the priority of Union law over national law has no added value. At most, the supposed ‘democratic’ origin may strengthen the practical effect of Union law at the national level. However, even this ‘democratic’ argument is doubtful. The priority of Union law over national law as laid down in Article I-6 of the Constitution for Europe was one of the major reasons for the Dutch population to reject this Treaty122 and, subsequently, for the deletion of that provision in the Lisbon Treaty. Finally, the argument of the multilevel theory that the priority of Union law finds its origin in the common will of the citizens has another important, albeit not intended effect. This origin of the priority of Union law means that national constitutional courts, by denying the priority of Union law or by postulating the supreme position of the national Constitution for the national legal system, are acting contrary to the democratic will of the European citizens and thus violating the ‘new’ social contract. In Rousseau’s thinking a breach of the social contract amounts to nothing less than a crime against the volonté générale! In order to explain the priority of Union law over national law in spite of the fact that both orders are autonomous, the multilevel theory argues that the priority of Union law is only a matter of ‘Anwendungsvorrang’ (‘primacy in application’) and not of ‘Geltungsvorrang’ (‘primacy in validity’).123 In other words, in case of a conflict between national law and Union law, national law remains valid; the only effect of Union law is that the conflicting national rule can no longer be applied. Both systems are thus autonomous since the validity of each rule depends only on the system to which it belongs. However, even supposing that this artificial distinction between application and validity, which finds its origin in old 122  See T Toonen, B Steunenberg, W Voermans, ‘Saying no to a European Constitution: Dutch Revolt, Enigma or Pragmatism’ (2005) Zeitschrift für Staats- und Europawissenschaften 594, 609. 123   Pernice (2007), n 4 at 87.

178  René Barents dualistic conceptions, actually corresponds to the priority doctrine of the ECJ,124 it contradicts the unity in substance thesis. Unity in substance cannot imply the existence of two valid but contradictory rules. If there is a unity in substance (monist model), the priority of Union law would, by virtue of its own authority, automatically lead to the nullity of the national rule or decision in question. However, in the ECJ’s case law this monist effect has been denied in categorical terms.125 There is also another contradiction. In order to support the thesis that the autonomous national and European systems nevertheless constitute a unity in substance, the multilevel theory invokes the duty of national courts to interpret national law in conformity with Union law.126 Leaving aside the question whether this argument actually supports the monist view, it contradicts the thesis referred to above that the primacy of Union law is limited to ‘Anwendungsvorrang’. According to the case law of the ECJ the duty of conform interpretation of national law finds its origin in the primacy of Union law.127 In other words, the argument of the multilevel theory that the primacy of Union law is restricted to ‘Anwendungsvorrang’ is false, since the duty of conform interpretation does not involve the application of Union law and, as a consequence, the disapplication of national law. What is applied is but national law interpreted in a different manner.128

D. Conclusion The multilevel theory thus appears to suffer from a serious conceptual weakness: the hypothesis of two or more autonomous levels is incompatible with the hypothesis of a unity in substance of Union law and national law (integrated legal order). By arguing that the autonomous Union level and the autonomous national levels constitute one single system in which the law of the former level has priority over the law of the latter level, the multilevel theory tries to square a circle or, in other words, tries to mix a monist and a pluralist approach although conceptually both approaches are mutually exclusive.129 If the starting point is that both levels are autonomous, there can only be a pluralist model and not a monist one. Since the law of these autonomous systems relates to the same territory, the only conclusion is that Union law and national law are ‘competing’.130 The ‘competitive’ 124   See on this R Barents, ‘De voorrang van Unierecht in het perspectief van constitutioneel pluralism’ [The primacy of Union law in the perspective of constitutional pluralism] (2009) SociaalEconomische Wetgeving 67. 125   Joined Cases C-10/97 to C-22/97 IN.CO.GE [1998] ECR I-6307. 126   Pernice (1998), n 4 at 46. 127   See, inter alia, Case C-106/89 Marleasing [1980] ECR I-4135, para 8; Joined Cases C-397/01 to C-403/01 Pfeiffer [2004] ECR I-8835, para 114. 128  This demonstrates that the distinction between ‘Anwendungsvorrang’ und ‘Geltungsvorrang’ cannot explain the priority of Union law over national law. 129   See, inter alia, L Azoulai, ‘La Constitution et l’intégration: les deux sources de l’Union européenne en formation’ (2003) Revue française de droit administratif 859, 865 et seq. 130   See extensively Barents (2004), n 19 at § 334 et seq.

The Fallacy of Multilevel Constitutionalism  179 position of Union law is based on its self-proclaimed priority as the necessary condition to ensure its effectiveness and the equal treatment of citizens on the territory of all Member States. The ‘competitive’ position of national law follows from the fact that the implementation and enforcement of Union law depends on the authorities of the Member States.131 A practical illustration of this pluralist model of ‘competitive’ orders is provided by the divergences between the case law of the ECJ on the priority of Union law and of national constitutional courts on the supreme position of national constitutions.132 In the pluralist setting this does not amount to a unity in substance but to a peaceful co-existence of two legal systems, to which it must be added that when reading the publications of various past and present members of the Bundesverfassungsgericht, one may wonder if a ‘war’ is that far away.133

V.  The Divided Sovereignty Thesis A. Summary According to orthodox constitutional theory the general will of the citizens is tantamount to the sovereignty of the people, a feature reflected in the preambles and first articles of numerous national Constitutions. This raises the question of how the existence of two autonomous levels of public power can be reconciled with the concept of sovereignty. The present paper is obviously not the place to treat this concept, which is neither exclusively legal nor political in nature. To cut a long story short, it may be stated that irrespective of time, place and circumstances, all perceptions of and theories on sovereignty always have one thing in common: a single (supreme) source of political and/or legal authority. For this reason, sovereignty has been compared to a singularity or a geometrical point; it cannot be divided. Indeed, as Rousseau134 and later Calhoun put it: ‘To divide it is to destroy 131   See on this process in particular K Alter, Establishing the Supremacy of European Law (Oxford, Oxford University Press, 2001). 132   See T Georgopoulos, ‘La neutralisation de la justice constitutionnelle en matière d’intégration européenne: transformation et limites de la fonction de contre-pouvoir de juge constitutionnel en France et en Allemagne’ (2005) 17 Revue européenne de droit public 1481; O Dörr, ‘Rechtsprechungskonkurrenz zwischen nationalen und europäischen Verfassungsgerichten’ (2006) Deutsches Verwaltungsblatt 1088; LS Rossi, ‘Supremazie incrociate: Trattato costituzionale europe e Costituzioni nazionali’, in L Daniele (ed), La dimensione internazionale ed europea de dirrito nell’esperienza della Corte costituzionale (Napoli, Edizioni ESI, 2006) 399–414; J-P Jacqué, ‘Droit constitutionnel et droit communautaire’ in Diritto comunitario e diritto interno (Milano, Giuffré, 2008) 3 et seq. 133   S Bross, ‘Bundesverfassungsgericht – Europäischer Gerichtshof – Europäischer Gerichtshof für Kompetenzkonflikte’ (2001) Verwaltungsarchiv 425; H-J Papier, ‘Das Subsidiaritätsprinzip – Bremse des europäischen Zentralismus’ in Depenheuer (ed), n 30 at 691, 704. See also the threats in HJ Papier, ‘Das muss sich ändern!’ (24.7. 2007) Frankfurter Allgemeine Zeitung. 134   Rousseau, n 101, Book IV, at § 13.

180  René Barents it.’135 However, for the multilevel theory this existential feature of sovereignty does not seem to pose a problem. Indeed, according to this theory sovereignty must be considered as being divided or shared since the general will of the citizens is expressed in two social contracts.136

B.  The Fallacy of Divided Sovereignty The theory (or ideology) that sovereignty can be divided over two levels of government dates from the end of the eighteenth century. The US Constitution does not answer the question whether the Union or the states are sovereign. In order to make the proposed Constitution acceptable for the states, the founding fathers argued that sovereignty of the people should be seen as being divided between the federation and the states. In the Federalist Papers, which are also quoted as support by the multilevel theory, many references can be found to sovereignty as being divided between the states and the federation.137 However, in the context of the doctrine of ‘limited government’, the founding fathers regarded sovereignty first and foremost as a matter of supreme competences. The advantage of this approach was to circumvent the question of who holds the ultimate power, the Union or the states.138 The same approach was taken by de Tocqueville, who argued that sovereignty relates to ‘le droit de faire les lois’.139 This line of argument also became popular in Germany, where the Bismarck Bundesverfassung of 1867 and the Reichsverfassung of 1871 carefully refrained from answering the question of whether the German Empire or the states were sovereign.140 However, in the abundant literature the question on the ultimate source of sovereignty was never solved.141 This short sketch indicates that sovereignty can only be considered as being divided if it is seen as a bundle of competences (‘sovereign rights’) allocated over two levels of authority.142 However, this is not what sovereignty is about, since this 135   PJ van Löben Sels, Beschouwingen over den Noord-Amerikanschen staten-oorlog van 1861–1865 (Utrecht, 1878) 36. 136   However, in Pernice (1999), n 3 at 726 it is argued that due to the European constraints on national constitutional autonomy, ‘national sovereignty is overcome’. 137  See The Federalist Papers (commentaries on the proposed US Constitution by Alexander Hamilton, James Madison and John Jay), 46 (New York, 1961), 295; see also no 9 at 76; no 33 at 201, no 40 at 250 and no 82 at 491. 138   See on this, inter alia, JN Rakove, ‘American Federalism: Was There an Original Understanding?’ in MR Killenbeck (ed), The Tenth Amendment and State Sovereignty (Lanham, Rowman and Littlefield Publishers, 2002) 107–29. 139   A de Tocqueville, De la démocratie en Amérique, Tome I (Paris, Calman Lévy, 1886) 136. 140   See, inter alia, G Oltmann, ‘The American constitutional model and German constitutional politics’ in RC Simmons (ed), The United States Constitution. The first 200 years (Manchester, Manchester University Press, 1989) 90. 141  See on this in particular S Oeter, ‘Souveränität und Demokratie als Probleme in der “Verfassungsentwicklung” der Europäischen Union’ (1995) 55 Zeitschrift für ausländisches öffentliches Recht und Völkerrecht 659. 142   See also D Friggieri, The EU and Sovereignty – Towards Multilevel Governance (Malta, Minima, 2002) 61.

The Fallacy of Multilevel Constitutionalism  181 concept relates to the ultimate source of power in a polity which, by its very nature, remains in the dark until one of the levels actually exercises its power.143 In this respect no one has ever given a better definition of sovereignty than Carl Schmitt: ‘Souverän ist, wer über den Ausnahmezustand entscheidet’144 (‘Sovereign is he who decides on the exception’145). This essential feature of sovereignty was demonstrated by the American Civil War and the events in Germany after 1933. Moreover, it is difficult to understand why a theory aimed at explaining constitutionalism beyond the state employs a concept which is more than anything else connected to the state. Arguing that sovereignty can be divided means in fact arguing that sovereignty has become obsolete146 or at least inappropriate for theoretical purposes,147 which is a more acceptable starting point for the explanation of a supranational construction aimed at the continuing integration of states and peoples in a context of globalisation. Moreover, the theory of divided sovereignty finds no support in the case law of the ECJ. In its various judgments on the nature of the EC, the ECJ has consistently and carefully argued that by the Treaties a new legal order is created the institutions of which are ‘endowed with sovereign rights’,148 for the benefit of which the ‘Member States have limited their sovereign rights’.149 The fallacy of divided sovereignty is illustrated by the provision in the Constitution for Europe and in the TEU, according to which a Member State can withdraw from the Union. This provision either has a declaratory or a constitutive value. Considered to be of declaratory value, the inevitable conclusion is that it recognises a right to withdraw which the state has by virtue of its sovereignty and/or by international law. In either case the state thus remains fully sovereign during the period of its membership of the Union. If that provision has a constitutive value, the conclusion is inevitable that by becoming a member of the Union, the state has lost its sovereignty since it is only on the basis of the said provision that it may withdraw. In that case sovereignty lost becomes sovereignty regained. In any event, in both cases sovereignty cannot be considered to be divided over the national and European level.

143   Accordingly, to explain the phenomenon of a federation as being different from a unitary state or a confederation is only possible as long as the question of who holds sovereignty remains unanswered. If the federation is based on political homogeneity, the question becomes obsolete. See Schmitt, n 21 at 371. 144  C Schmitt, Politische Theologie. Vier Kapitel zur Lehre von der Souveränität 3rd edn (Berlin, Duncker & Humblot, 1985) 9. 145   C Schmitt, Political Theology: Four Chapters on the Concept of Sovereignty (Chicago, University of Chicago Press, 2005) 5. 146   F Luchaire, ‘Un concept inutile: la souveraineté’ in Renouveau du droit constitutionnel. Mélanges en l’honneur de Louis Favoreu (Paris, Dalloz, 2007) 789, 792. 147   N MacCormick, Questioning Sovereignty (Oxford, Oxford University Press, 1999) 126. 148   Case 26/62 Van Gend & Loos [1963] ECR 1. 149   Case 6/64 Costa v ENEL [1964] ECR 1149 (‘albeit in limited fields’) and Opinion 1/91 EEA I [1991] I-6079, para 21 (‘in ever wider fields’).

182  René Barents

VI.  A Sad but Inevitable Conclusion The multilevel theory can thus be qualified as a fallacy.150 It is based on dubious assumptions (unity in substance, divided sovereignty), applies a questionable methodology (citizenship and democratic legitimacy), and suffers from serious contradictions (autonomy and monist priority). In view of the laudable ambitions of the multilevel theory, this conclusion is more than regrettable, in particular having regard to the impressive intellectual efforts of its architects and their intentions to build on the thoughts of Walter Hallstein.151 As observed at the beginning of this paper, this theory was developed in order to provide for an alternative to the explanation of the Union in terms of orthodox constitutional ideology, in which the state is considered to be the ultimate form of ‘the political’ and the sole guarantee for democracy. In particular, the multilevel theory attempts to escape from the consequences of the near religious people-state-Constitution doctrine professed in its ultimate form by the Bundesverfassungsgericht152 and followed by several other constitutional courts, the essence of that doctrine being that, as an inherent result of the postulate of national sovereignty, Union law can never be perceived as anything else but ‘alien law’. Ambitions and intentions notwithstanding, the multilevel theory is in any case incapable of providing a theoretical explanation of what the Union is or might be according to the Treaties and the case law of the Court. This theory is essentially based on a fiction about democratic legitimacy: since the Member States draw their legitimacy from the will of the citizens, it must therefore be possible to explain the existence and functioning of the Union on the same basis. In spite of the fact that the ‘Verfassungsverbund’ theory is aimed to reach the opposite result of the ‘Staatenverbund’ theory of the Bundesverfassungsgericht, both theories have in common that they are based on (democratic) constitutional ideology. Where the ‘Staatenverbund’ theory is based on the will of the democratic states as the supreme source of politics and law, in the ‘Verfassungsverbund’ theory this foundation is exchanged for the hypothetical common will of the citizens. The ideological basis is the weakness of both theories. Neither the will of the democratic Member States nor the democratic will of the citizens can provide for a sound theoretical explanation of what the Union is. The multilevel theory also teaches the important lesson that concepts of social and political science cannot be employed to develop a legal theory on the Union

150   This conclusion is not new; see, inter alia, M Jestaedt, ‘Der Europäische Verfassungsverbund: verfassungstheoretischer Charme und rechtstheoretische Insuffizienz einer Unschärferelation’ in Krause et al (ed), n 30 at 637. 151   See, inter alia, I Pernice, ‘Zur Finalität Europas’ in GF Schuppert (ed), Europawissenschaften (Baden-Baden, Nomos, 2005). 152   See, inter alia, RC van Ooyen, Die Staatstheorie des Bundesverfassungsgerichts und Europa. Von Solange über Maastricht zum EU-Haftbefehl (Baden-Baden, Nomos, 2006) 55 et seq.

The Fallacy of Multilevel Constitutionalism  183 without incurring the risk of serious inconsistencies and contradictions.153 One cannot to fail to conclude that what is considered to be situated beyond the state (post-national constellations), is explained with the aid of distorted tools from state theory (sovereignty, legal order, legal unity, social contract, etc.) or tools from state origin (democracy, citizenship) with respect to which, in spite of the fact that they belong to the obligatory grammar of the Treaties, it still has to be demonstrated that are actually useful for this purpose.154 As Jo Shaw has accurately observed, in the domain of European constitutionalism it is difficult to get rid of the ‘touch of stateness’.155 The conclusion that the multilevel theory is a fallacy confirms what has been called the continuing ‘crisis’ of European constitutionalism.156 Although EU governance demonstrates that public power, law and politics beyond the state are a (constitutional) reality, European constitutionalism still fails to offer a comprehensive theory for the explanation of these post-national phenomena. At this background and in view of the short lived inspirations from the Constitution for Europe, one might wonder whether lawyers should not be more modest: by going back to the original source, the European Treaties.157

153  See, inter alia, E Rumler-Korinek, ‘Governance’ und ‘Accountability’. Reine Modeworte oder Schlüsselbegriffe einer Demokratie auf EU-Ebene?’ (2004) Journal für Rechtspolitik 227. 154  See on this methodological problem A Dyèvre, ‘The constitutionalisation of the European Union: discourse, present, future and facts’ (2005) 2 European Law Review 165; K Thomalla, ‘Der Begriff der europäischen Verfassung als verfassungstheoretisches und sozialphilosophisches Problem’ (2007) Archiv für Rechts- und Sozialphilosophie 178; P Dann, ‘Thoughts on a methodology of European constitutional law’ (2005) European Review of Public Law 1413, 1420. See also J-C Piris, ‘L’Union européenne: vers une nouvelle forme de fédéralisme’ (2005) Revue trimestrielle de droit européen 243. 155   J Shaw and A Wiener, ‘The Paradox of the `European Polity’ in MG Cowles and M Smith (eds), State of the European Union 5: Risks, Reform, Resistance, and Revival (Oxford, Oxford University Press, 2000) 65. 156   Allot, n 23 at 439, 476. 157   HM Heinig, ‘Europäisches Verfassungsrecht ohne Verfassung(svertrag)?’ (2007) 62 Juristen Zeitung 905, 909.

8 Federalism as Constitutional Pluralism: ‘Letter from America’ ROBERT SCHÜTZE

I.  Introduction: Divided Sovereignty – An American ‘Invention’ The aim of the 1787 Constitution had been ‘to divide the sovereign authority into two parts’.1 ‘This idea of the divisibility of sovereignty had emerged at the time of the Federal Convention, and was a distinctly American theory’.2 Each State would give up part of its sovereignty;3 the ‘nation’, on the other hand, would remain ‘incomplete’.4 While sovereignty lay ultimately in ‘the people’, the Constitution had delegated and divided it between two levels of government.5 Who, then, was the ‘sovereign’? The ‘We the people’ formula had left the identity of the popular sovereign ‘indeterminate’. The formula simply acknowledged that the Americans had rejected the British concept of governmental sovereignty. Sovereignty meant popular   A de Tocqueville (P Bradley ed), Democracy in America (New York, Vintage, 1954) 151.   EK Bauer, Commentaries on the Constitution: 1790–1860 (New York, Columbia University Press, 1952) 214. On this point, see also AC McLaughlin, The Foundations of American Constitutionalism (New York, New York University Press, 1932) 78. 3  In The Federalist No 42, Madison ridiculed the theory according to which absolute sovereignty had remained in the States. ‘[T]he articles of Confederation have inconsiderately endeavored to accomplish impossibilities; to reconcile a partial sovereignty in the Union, with complete sovereignty in the States; to subvert a mathematical axiom, by taking away a part, and letting the whole remain.’ See A Hamilton, J Madison and J Jay (T Ball ed), The Federalist (Cambridge, Cambridge University Press, 2003) 206. 4   On the idea of a ‘partial’ nation, see The Federalist No 39 (ibid 181–7). In the words of the Supreme Court in McCulloch v Maryland 17 U.S. 316 (1819), 410: ‘In America, the powers of sovereignty are divided between the Government of the Union and those of the States. They are each sovereign with respect to the objects committed to it, and neither sovereign with respect to the objects committed to the other’. Thus, even ‘Marshall recognized the divided sovereignty of American federalism and acknowledged a sphere in which the states were supreme’. AH Kelly, WA Harbison and H Belz, The American Constitution: Its Origins and Development. Volume I (New York, Norton, 1991), 185. 5   The import of this is clarified in The Federalist No 51 (n 3, 253–54): ‘In the compound republic of America, the power surrendered by the people, is first divided between two distinct governments, and then the portion allotted to each, subdivided among distinct and separate departments. Hence a double security arises to the rights of the people. The different governments will control each other; at the same time that each will be controuled by itself.’ 1 2

186  Robert Schütze sovereignty. But who were ‘the people’: the peoples of the United States, or the people of the United States? This was a question the Constitution had left unanswered. For while it had been ‘authorized’ by the State peoples acting separately, the amendment power signalled a willingness to renounce their – unilateral – sovereignty. The locus of sovereignty in the compound republic was thus left in suspense. What would happen where the visions of the State people(s) and the ‘American people’ diverged? Who would be the ultimate arbiter in the event of constitutional conflicts? Constitutional conflicts soon emerged.6 How did the United States resolve them? Which authority would resolve them? We shall try to answer these questions by looking at constitutional conflicts over federal supremacy in American constitutional history.7 Federal supremacy was expressly provided for in the 1787 Constitution.8 However, this supremacy was confined to the ‘national’ sphere, that is: those competences that the Constitution delegated to the federal government. What would happen when the Union trespassed the constitutional boundary? The first part of this Chapter investigates the emergence of theories of ‘interposition’ and ‘nullification’ as constitutional remedies in the antebellum period. Interposition still accords with the doctrine of divided sovereignty. By contrast, (unilateral) nullification was based on the idea of State sovereignty. This second view was opposed by the American nation and led to the ‘Civil War’. The Civil War was an (extra)constitutional watershed. The ‘re-construction’ that followed offered the Union an unprecedented constitutional reformation. But while it accelerated national ‘consolidation’, postbellum constitutionalism returned to the idea of divided sovereignty. And while the theory of State sovereignty was 6   For excellent surveys over the – numerous – American constitutional controversies see C Warren, ‘Legislative and Judicial Attacks on the Supreme Court of the United States: A History of the TwentyFifth Section of the Judiciary Act (Part I and II)’ (1913) 47 American Law Review 1 and 161; Anonymous, ‘Interposition vs. Judicial Power’ (1956) 1 Race Relations Law Reporter 465; as well as F McDonald, States’ Rights and the Union: Imperium in Imperio, 1776–1876 (Lawrence, University Press of Kansas, 2000); and for an outstanding overview of the ‘literary’ constitutional conflicts prior to the Civil War, see EK Bauer, n 2. 7   My argument concentrates on legislative (or political) protests. However, we must not forget that in the history of constitutional conflicts State courts would equally loudly protest against nationalist tendencies. One of the most famous judicial ‘rebellions’ is the Martin v Hunter’s Lessee saga. The case concerned the confiscation of loyalist property by Virginia after independence. The land was sold to Hunter, whose right was challenged by Martin – the loyalist owner’s devisee. The latter claimed restitution on the basis of the Jay Treaty. The Supreme Court had found in favour of Martin in Fairfax’s Devisee v Hunter’s Lessee, 11 U.S. 603 (1813). However, the Virginia Court of Appeals refused to obey. It justified its judicial disobedience by arguing that the Supreme Court had no monopoly in interpreting the Constitution (cf Hunter v Martin, 18 Va. 1 (1815)). The Supreme Court critically evaluated this ‘pluralist’ position in Martin v Hunter’s Lessee, 14 U.S. 304 (1816). For a detailed analysis of this case, see CG Haines, The Role of the Supreme Court in American Government and Politics: 1789–1835 (Berkeley, University of California Press, 1944), 340–51. For a comparison of this case with the Maastricht Decision of the German Constitutional Court in the European Union, see SJ Boom, ‘The European Union after the Maastricht Decision: Will Germany be the “Virginia of Europe?”’ (1995) 43 American Journal of Comparative Law 177. 8   Article VI – Section 2 states: ‘This Constitution, and the Laws of the United States which shall be made in Pursuance thereof; and all Treaties made, or which shall be made, under the Authority of the United States, shall be the supreme Law of the Land; and the Judges in every State shall be bound thereby, any Thing in the Constitution or Laws of any State to the contrary notwithstanding.’

Federalism as Constitutional Pluralism  187 discredited, the pluralist tradition re-entered the constitutional scene in the second half of the twentieth century. The conclusion, consequently, argues that constitutional pluralism is an inherent characteristic of American federalism.

II.  Constitutional Conflicts before the Civil War The ingenious pragmatism of The Federalist had sacrificed a metaphysical disputation of abstract sovereignty to a physical analysis of concrete constitutional structures. But this did not mean that the potential for constitutional conflicts had not been seen. The fathers of American constitutionalism were alert to divergent interpretations of the federal Constitution: ‘If there is in each State, a court of final jurisdiction, there may be as many different final determinations on the same point, as there are courts.’9 And this was particularly true for federal points.10 But accepting that a central authority was required to interpret the federal Constitution, did this mean that a central court was to judge all constitutional controversies? What would happen where the federal government had overreached its powers, say: under the ‘Necessary and Proper Clause’? This was the ‘Hamiltonian’ answer to this question: Who is to judge of the necessity and propriety of the laws to be passed for executing the powers of the Union? I answer, first, that this question arises as well and as fully upon the simple grant of those powers as upon the declaratory clause: And I answer, in the second place, that the national government, like every other, must judge, in the first instance, of the proper exercise of its powers, and its constituents in the last. If the federal government should overpass the just bounds of its authority and make a tyrannical use of its powers, the people, whose creature it is, must appeal to the standard they have formed, and take such measures to redress the injury done to the Constitution as the exigency may suggest and prudence justify . . . If it be said that the legislative body are themselves the constitutional judges of their own powers, and that the construction they put upon them is conclusive upon the other departments, it may be answered, that this cannot be the natural presumption, where it is not to be collected from any particular provisions in the Constitution. It is not otherwise to be supposed, that the Constitution could intend to enable the representatives of the people to substitute their will to that of their constituents. It is far more rational to suppose, that the courts were designed to be an intermediate body between the people and the legislature, in order, among other things, to keep the latter within the limits assigned to their authority.11

  The Federalist No 22 (Hamilton), n 3, 104.   Ibid. ‘In this case, if the particular tribunals are invested with a right of ultimate jurisdiction, besides the contradictions to be expected from differences of opinion, there will be much to fear from the bias of local views and prejudices, and from the interference of local regulations.’ 11   The Federalist No 33 (Hamilton) and The Federalist No 78 (Hamilton), n 3, 150 and 379 (emphasis added). 9

10

188  Robert Schütze The view was shared by another voice within The Federalist : It is true that in controversies relating to the boundary between the two jurisdictions, the tribunal which is ultimately to decide, is to be established under the general government. But this does not change the principle of the case. The decision is to be impartially made, according to the rules of the Constitution; and all the usual and most effectual precautions are taken to secure this impartiality.12

But what if this impartiality would fail? Conceding that ‘[i]n the first instance, the success of the [legislative] usurpation will depend on the executive and judicial departments, which are to expound and give effect to the legislative acts’; still, ‘in the last resort a remedy must be obtained from the people, who can by the election of more faithful representatives, annul the acts of the usurpers’. National elections would be the constitutional remedy of ‘the people’. What, on the other hand, was the role of the State governments? The State legislatures ‘will be ever ready to mark the innovation, to sound the alarm to the people, and to exert their local influence in effecting a change of federal representatives’.13 Would these constitutional remedies be sufficient to resolve conflicts over the federal boundary? It did not take too long before constitutional words became constitutional deeds. The founders’ views would be tested in 1798, when the Union faced a major constitutional crisis over two federal laws that were claimed to abridge fundamental human rights.

A.  The Theory of Interposition: The Alien and Sedition Crisis In 1798, the federal government enacted the Alien and Sedition Acts. The Acts allowed the Union, through its president, to adopt emergency measures. These could lead to extreme limitations of personal liberties.14 Two States regarded the federal laws as violations of the Bill of Rights.15 Their constitutional discontent expressed itself through resolutions of their respective legislatures. The Virginia and Kentucky Resolutions reignited the final arbiter question. The Virginia protest – secretly drafted by James Madison – elaborated the theory of interposition and (re)stated the view of the Constitution as compact between the States: That this Assembly doth explicitly and peremptorily declare, that it views the powers of the federal government, as resulting from the compact, to which the states are parties; as limited by the plain sense and intention of the instrument constituting the compact; as no further valid that they are authorized by the grants enumerated in that compact; and   The Federalist No 39 (Madison), n 3, 186.   The Federalist No 44 (Madison), n 3, 221. 14  Extracts of the Alien and Sedition Acts can be found in HS Commager (ed), Documents of American History. Volume 1 (New York, Appleton-Century-Crofts, 1949) 175–78. 15   The Bill of Rights had become part of the Constitution through the first 10 amendments. On the origin and status of the Bill of Rights, see AR Amar, ‘The Bill of Rights as a Constitution’ (1990–91) 100 Yale Law Review 1131 and, by the same author, The Bill of Rights: Creation and Reconstruction (New Haven, Yale University Press, 2000). 12 13

Federalism as Constitutional Pluralism  189 that in case of a deliberate, palpable, and dangerous exercise of other powers, not granted by the said compact, the states who are parties thereto, have the right, and are in duty bound, to interpose for arresting the progress of the evil, and for maintaining within their respective limits, the authorities, rights and liberties appertaining to them. That the General Assembly doth also express its deep regret, that a spirit has in sundry instances, been manifested by the federal government, to enlarge its powers by forced constructions of the constitutional charter which defines them; and that implications have appeared of a design to expound certain general phrases (which having been copied from the very limited grant of power, in the former articles of confederation were the less liable to be misconstrued) so as to destroy the meaning and effect, of the particular enumeration which necessarily explains and limits the general phrases; and so as to consolidate the states by degrees, into one sovereignty.16

The Virginia legislature considered the Alien and Sedition Acts ‘alarming infractions of the Constitution’. The Acts would violate the federal boundary as well as fundamental human rights. The Virginia Parliament thus appealed ‘to the like dispositions of the other states, in confidence that they will concur with this commonwealth in declaring, as it does hereby declare, that the acts aforesaid, are unconstitutional’.17 This constitutional protest was joined by the Kentucky legislature. But the tone of the Kentucky Resolutions was stronger. They bore the imprint of Thomas Jefferson – then Vice President of the United States.18 A first Resolution confid­ ently asserted [t]hat the several states composing the United States of America, are not united on the principle of unlimited submission to their general government; but that by compact under the style and title of a Constitution for the United States . . . [T]he government created by this compact was not made the exclusive or final judge of the extent of the powers delegated to itself; since that would have made its discretion, and not the Constitution, the measure of its powers; but that as in all other cases of compact among powers having no common Judge, each party has an equal right to judge for itself, as well of infractions as of the mode and measure of redress.19

A second Resolution even invoked the novel constitutional remedy of nullification. As the States were ‘sovereign and independent’, as the authors of the Constitution, they had the ‘unquestionable right to judge of the infraction; and, [t]hat a nullification of those sovereignties, of all unauthorized acts done under color of that instrument is the rightful remedy’. And while ordinarily ‘disposed to surrender its opinion to a majority of its sister states’, in the case of the Alien and Sedition Acts ‘it would consider a silent acquiescence as highly criminal’. Kentucky would therefore not cease ‘to oppose in a constitutional manner’ every attempt by the federal government to violate the constitutional compact.20   Virginia Resolutions (24 December 1798), Commager, n 14, 182 (emphasis added).   Ibid 183 (emphasis added).   Jefferson had originally drafted an even stronger worded version; see T Jefferson (PL Ford ed), The Writings of Thomas Jefferson. Volume VII (New York, Putnam’s Sons, 1896) 289–309. 19   Kentucky Resolutions (16 November 1798), Commager, n 14, 178–9. 20   Kentucky Resolutions (22 February 1799), Commager, n 14, 184. 16 17 18

190  Robert Schütze None of these Resolutions gained wide support. While all the States (but one) accepted the compact theory,21 the sister States were critical.22 The use of the nullification concept in the 1799 Kentucky Resolution even inspired internal criticism in the form of a rejoinder from Virginia. The ‘Report of 1800’ – written by Madison and adopted by the legislature of his State – expressly rejected the concept of nullification as a unilateral right of each State. Because the Constitution was a compact between the States, only the latter – in the collective plural – could be its final interpreters: The Constitution of the United States was formed by the sanction of the states, given by each in its sovereign capacity. It adds to the stability and dignity, as well as to the authority of the Constitution, that it rests on this legitimate and solid foundation. The states then being the parties to the constitutional compact, and in their sovereign capacity, it follows of necessity, that there can be no tribunal above their authority, to decide in the last resort, whether the compact made by them be violated; and consequently that as the parties to it, they must themselves decide in the last resort, such questions as may be of sufficient magnitude to require their interposition. It does not follow, however, that because the states as sovereign parties to their constitutional compact, must ultimately decide whether it has been violated, that such a decision ought to be interposed, either in a hasty manner, or on doubtful and inferior occasions. Even in the case of ordinary conventions between different nations, where, by the strict rule of interpretation, a breach of a part may be deemed a breach of the whole; every part being deemed a condition of every other part and of the whole, it is always laid down that the breach must be both willful and material to justify an application of the rule. But in the case of an intimate and constitutional union, like that of the United States, it is evident that the interposition of the parties, in their sovereign capacity, can be called for by occasions only, deeply and essentially affecting the vital principles of their political system.23

The view that the Supreme Court was to be the sole and final judge of constitutional conflicts neglected that ‘dangerous powers not delegated, may not only be usurped and executed by the other departments, but that the Judicial Department also may exercise or sanction dangerous powers beyond the grant of the Constitution’.24 But what remedy was there to fight off a federal usurpation? The Report characterized the Resolutions as ‘expressions of opinion’ with no legal effect. They were to ‘excite reflection’. The constitutional protest of the States was an appeal to 21   AH Kelly et al, n 4, 140. Only Vermont argued that the Union had been formed by ‘the’ people of the United States. 22   For an overview of the replies by the sister States, see the appendix to FM Anderson, ‘Contemporary Opinion of the Virginia and Kentucky Resolutions (Part II)’ (1899) 5 American Historical Review 225, 244 et seq. 23   J Madison, ‘The Report of 1800’, in DB Mattern et al (eds), The Papers of James Madison. Volume 17 (Chicago, University Press of Virginia, 1991) 309–10 (emphasis added). 24   ‘However true therefore it may be, that the Judicial Department, is, in all questions submitted to it by the forms of the Constitution, to decide in the last resort, this resort must necessarily be deemed the last in relation to the authorities of the other departments of the government; not in relation to the rights of the parties to the constitutional compact, from which the judicial as well as the other departments hold their delegated trusts’. Ibid 311.

Federalism as Constitutional Pluralism  191 the people in the sister states to limit the federal government.25 Madison’s constitutional ‘contractualism’ thus did not acknowledge unilateral rights to the States.26 States’ rights were collective rights: only the States – in the collective plural – were masters of the Constitution.27 By contrast, interposition was a constitutional remedy that belonged to the State governments. The State governments could place themselves ‘between’ their people and the national government.28 They would signal a (potential) breach of the federal contract by means of – to borrow a constitutional concept from European constitutionalism29 – a yellow card mechanism.30 Yet, ultimately only ‘the people’ could decide to amend the Constitution or decide to change the federal government. The former had successfully happened in 1795.31 The latter would ultimately resolve the present constitutional crisis. The alarm sounded by Virginia and Kentucky was heard. The changed public opinion led to the ‘Revolution of 1800’. The election of Thomas Jefferson to the presid­ ency was a confirmation of the ‘principles of ‘98’.32

  Ibid 348.   HJ Powell, ‘The Principles of ‘98: An Essay in Historical Retrieval’ (1994) 80 Virginia Law Review 689, 718. This contrasted with Jefferson’s vision on the nature of the federal union. The latter saw the Constitution as a bilateral compact of each State with its co-States. On the constitutional differences between Jefferson and Madison, see A Koch and H Ammon, ‘The Virginia and Kentucky Resolutions: An Episode in Jefferson’s and Madison’s Defense of Civil Liberties’ (1948) 5 William and Mary Quarterly 145. 27   For the same point in the context of the European legal order, see R Schütze, ‘On “Federal” Ground: The European Union as an (Inter)national Phenomenon’ (2009) 46 Common Market Law Review 1069, 1082. 28   What Madison and his contemporaries meant by ‘interposition’ derived from an astronomical context. It described the movement of something between two other things in a relationship that would – temporarily – interrupt this relationship. For the historical connotations of this notions, see CG Fritz, American Sovereigns: The People and America’s Constitutional Tradition before the Civil War (Cambridge, Cambridge University Press, 2008) 193 (referring to Johnson’s Dictionary). 29   On the role of national Parliaments and the yellow card mechanism in controlling the European legislature, see R Schütze, ‘Subsidiarity after Lisbon: Reinforcing the Safeguards of Federalism?’ (2009) 68 Cambridge Law Journal 525. 30  In The Federalist No 46, n 3, 231–2, we read: ‘But ambitious encroachments of the federal government, on the authority of the State governments, would not excite the opposition of a single State, or of a few States only. They would be signals of general alarm. Every government would espouse the common cause. A correspondence would be opened. Plans of resistance would be concerted. One spirit would animate and conduct the whole. The same combinations, in short, would result from an apprehension of the federal, as was produced by the dread of a foreign, yoke; and unless the projected innovations should be voluntarily renounced, the same appeal to a trial of force would be made in the one case as was made in the other. But what degree of madness could ever drive the federal government to such an extremity’. And the ‘Report of 1800’ also appealed to the ‘vigilance’ of State governments to ‘sound the alarm to the public’ (Madison, n 23, 350). 31   In 1795, the Eleventh Amendment overturned the Supreme Court’s interpretation of Article III – Section 2 in Chisholm v State of Georgia, 2 U.S. 419 (1793). 32   On this point, see HJ Powell, n 26, 694 fn 16 and CG Fritz, n 28, 210: ‘Consistent with the theory of interposition, in 1800 American voters went to the polls and chose between candidates who took opposing positions on those acts. Jefferson made “violations of the true principles” of the constitution a central campaign issue for the Republican party. His election to the presidency and that of his followers to Congress reflected public opinion about the constitutionality of the Acts’. 25 26

192  Robert Schütze

B.  The Theory of Nullification: The ‘Tariff of Abominations’ While the idea of the Constitution as a compact between the States had settled by 1800, a theory of State sovereignty only emerged a few decades later in the wake of another constitutional controversy: the ‘Tariff of Abominations’. The agricultural South protested against a federal tariff that was claimed to favour the industrial North. South Carolina ‘exposed’ its anger. The South Carolina ‘Exposition and Protest’ (1828) was a brainchild of John Calhoun – then Vice President of the United Sates.33 It challenged the competence of the federation to adopt the federal tariff and claimed the ‘discriminatory tariff’ was unconstitutional.34 The Exposition hereby returned to the quintessential federal question: who was the ultimate arbiter of the Union’s competences? What constitutional safeguards would protect the States against a federal usurpation? The Exposition’s answer was the theory of nullification. While Calhoun tried to portray nullification as a logical extension of the (Madisonian) theory of interposition, the former was a – radical – step beyond the latter. The theory of nullification derived from the (European) theory of – undivided – state sovereignty. The constitutional philosophy behind the Exposition was this: There is no difficulty in understanding how powers, appertaining to sovereignty, may be divided; and the exercise of one portion delegated to one set of agents, and another portion to another. But how sovereignty itself – the supreme power – can be divided – how the people of the several States can be partly sovereign, and partly not sovereign – partly supreme, and partly not supreme, it is impossible to conceive. Sovereignty is an entire thing – to divide, is – to destroy it. 35 33   For a reprint of the ‘Exposition and Protest’ see JC Calhoun (RM Lence), Union and Liberty (Indianapolis, Liberty Fund, 1992), 311. 34   The Southern States claimed that the ‘manufacturing States’ would bear no share of the burden of the tariff: ‘We cultivate certain great staples for the supply of the general market of the world: They manufacture almost exclusively for the home market. Their object in the Tariff is to keep down foreign competition, in order to obtain a monopoly of the domestic market. The effect on us is, to compel us to purchase at higher price, both what we can obtain from them and from others, without receiving a correspondent increase in the price of what we sell’ (ibid 321). The point was elaborated later on: ‘The question, then, is not, whether those States should or should not manufacture – for necessity, and the policy of other nations had decided that question – but whether they should, with or without a bounty. It was our interest that they should without. It would compel them to contend with the rest of the world in our market, in free and open competition; the effects of which would have been, a reduction of prices to the lowest point; thereby enabling us to exchange the products of our labor most advantageously – giving little, and receiving much; while, on the other hand, in order to meet European competition, they would have been compelled to work at the lowest wages and profits. To avoid this, it was their interest to manufacture with a bounty; by which our situation was completely reversed. They were relieved by our depression. Thus, through our political connection, by a perversion of the powers of the Constitution, which was intended to protect the States of the Union in the enjoyment of their natural advantages, they have stripped us of the blessings bestowed by nature, and converted them to their own advantage. Restore our advantages, by giving us free trade with the world, and we would become, what they now are by our means, the most flourishing people on the globe. But these are withheld from us under the fear that, with their restoration, they would become, what we are by their loss, among the most depressed’ (ibid 328–29). 35   JC Calhoun, ‘A Discourse on the Constitution and Government of the United States’ in Calhoun, n 33, 105.

Federalism as Constitutional Pluralism  193 The insistence on – absolute – sovereignty ‘marked a significant turning point in nineteenth century constitutionalism’.36 Dialectically, as a result of Calhoun’s insistence on state sovereignty (1), defenders of federal authority would henceforth claim that the Union was organic and thus permanent (2).

1.  The Sovereign States View: The South Carolina Exposition In its protest against the ‘Tariff of Abominations’, South Carolina had proposed the theory of nullification. This idea, voiced in the Kentucky Resolution, would receive its first theoretical justification in the South Carolina Exposition. The Exposition started from first principles, in particular the concept of sovereignty and the orthodox distinction between ‘Government’ and ‘Sovereignty’. In the American Union, neither the federal nor the State governments were sovereign. Sovereignty lay in the people. But which people – the American people or the State peoples? The Exposition’s answer was that ‘sovereignty resides in the people of the States respectively’.37 But if that was the first principle behind the Constitution, what constitutional remedy was there to enforce it? ‘If we look to the history and practical operation of the system, we shall find, on the side of the States, no means resorted to in order to protect their reserved rights against the encroachments of the General Government’; yet, by contrast, ‘the latter has, from the beginning, adopted the most efficient [sic] to prevent the States from encroaching on those delegated to them.38 Alluding to The Federalist’s position on this point, it was held to be a ‘strange misconception’ to believe that the Supreme Court would be on the side of the States.39 It was equally mistaken to rely on the amendment power.40 The only right position was this: 36  Kelly et al, n 4, 208: ‘Calhoun’s argument for nullification marked a significant turning point in nineteenth century constitutionalism. Up to this time, the debate over federalism had tended to focus on the division and allocation of powers between the federal government and the states. People could argue about the locus of specific powers without having to challenge the Union. Calhoun’s insentience on a unitary conception of sovereignty, however, had had the effect of forcing a more precise definition of the federal-state relationship’. 37   Calhoun, n 33, 344. 38   Ibid 344–45. 39   ‘But, by a strange misconception of the nature of our system – and, in fact, of the nature of government – it has been regarded as the ultimate power, not only of protecting the General Government against the encroachments of the governments of the States, but also of the encroachments of the former on the latter – and as being, in fact, the only means provided by the Constitution of confining all the powers of the system to their proper constitutional spheres; and, consequently, of determining the limits assigned to each. Such a construction of its powers would, in fact, raise one of the departments of the General Government above the parties who created the constitutional compact, and virtually invest it with the authority to alter, at its pleasure, the relative powers of the General and State Governments, on the distribution of which, as established by the Constitution, our whole system rests – and which, by an express provision of the instrument, can only be altered by three-fourths of the States, as has already been shown’ (Calhoun, n 33, 345). 40  ‘The disease is, that a majority of the States, through the General Government, by construction, usurp powers not delegated, and by their exercise, increase their wealth and authority at the expense of the minority. How absurd, then, to expect the injured States to attempt a remedy by proposing an amendment to be ratified by three-fourths of the States, when, by supposition, there is a majority opposed to them? Nor would it be less absurd to expect the General Government to propose amendments, unless compelled to that course by the acts of a State. The Government can have no inducement. It has a more

194  Robert Schütze If it be conceded, as it must be by every one who is the least conversant with our institutions, that the sovereign powers delegated are divided between the General and State Governments, and that the latter hold their portion by the same tenure as the former, it would seem impossible to deny to the States the right of deciding on the infractions of their powers, and the proper remedy to be applied for their correction. The right of judging, in such cases, is an essential attribute of sovereignty – of which the States cannot be divested without losing their sovereignty itself – and being reduced to a subordinate corporate condition. In fact, to divide power, and to give to one of the parties the exclusive right of judging of the portion allotted to each, is, in reality, not to divide it at all; and to reserve such exclusive right to the General Government (it matters not by what department to be exercised), is to convert it, in fact, into a great consolidated government, with unlimited powers, and to divest the States, in reality, of all their rights. It is impossible to understand the force of terms, and to deny so plain a conclusion. The opposite opinion can be embraced only on hasty and imperfect views of the relation existing between the States and the General Government. But the existence of the right of judging of their powers, so clearly established from the sovereignty of States, as clearly implies a veto or control, within its limits, on the action of the General Government, on contested points of authority; and this very control is the remedy which the Constitution has provided to prevent the encroachments of the General Government on the reserved rights of the States; and by which the distribution of power, between the General and State Governments, may be preserved forever inviolable, on the basis established by the Constitution. It is thus effectual protection is afforded to the minority, against the oppression of the majority.41

Since each State was sovereign it would have the power to ‘veto or control, within limits’ the actions of the federal government ‘on contested points of authority’. This constitutional right to nullify national laws followed from the nature of the Union as a federal government. How was the remedy to be used? It would be confined to ‘a violation so deliberate, palpable, and dangerous, as to justify the interposition of the State to protect its rights’.42 Who can use it: the State government (legislature) or the State people? Since the former may not perfectly represent the people, Calhoun’s federal philosophy prefers the latter option. The people of each State may express their wish to nullify a federal law through a summary mode – the assumption of power by construction. The consequence is clear – neither would resort to the amending power – the one, because it would be useless – and the other, because it could effect its purpose without it – and thus the highest power known to the Constitution – on the salutary influence of which, on the operations of our political institutions, so much was calculated, would become, in practice, obsolete, as stated; and in lieu of it, the will of the majority, under the agency of construction, would be substituted, with unlimited and supreme power’ (Calhoun, n 33, 356). 41   Calhoun, n 33, 348–49 (emphasis added). 42   Ibid 351. Calhoun would insist on this limitation in Calhoun, n 35, 197–98: ‘To avoid this, prudence and propriety require that they should abstain from interposing their authority, to arrest an act of their common government, unless the case, in their opinion, involve a clear and palpable infraction of the instrument. They are bound to go further – and to forbear from interposing, even when it is clear and palpable, unless it be, at the same time, highly dangerous in its character, and apparently admitting of no other remedy; and for the plain reason, that prudence and propriety require, that a right so high and delicate should be called into exercise, only in cases of great magnitude and extreme urgency.’

Federalism as Constitutional Pluralism  195 State Convention.43 The (State) people here speak with the same authority and in the same ‘sovereign’ capacity as when ratifying the original 1787 Constitution.44 Constitutional theory led to constitutional practice. Four years after the Exposition, South Carolina proceeded to adopt ‘[a]n ordinance to nullify certain acts of the Congress of the United States, purporting to be laws laying duties and imposts on the importation of foreign commodities’.45 In line with Calhoun’s theory, a State Convention had found that the federal Union had ‘exceeded its just powers under the Constitution’. The Convention ordained thus the following: We, therefore, the people of the State of South Carolina, in Convention assembled, do declare and ordain . . . [t]hat the several acts and parts of acts of the Congress of the United States, purporting to be laws for the imposing of duties and imposts on the importation of foreign commodities, and now having actual operation and effect within the United States, and, more especially . . . [the tariff acts of 1828 and 1832] . . . are unauthorized by the Constitution of the United States, and violate the true meaning and intent thereof, and are null, void, and no law, nor binding upon this State, its officers or citizens . . . And it be further ordained, [t]hat it shall not be lawful for any of the constitutional authorities, whether of this State or of the United States, to enforce the payment of the duties imposed by the said acts within the limits of the State; but it shall be the duty of the Legislature to adopt such measures and pass such acts as may be necessary to give full effect to this Ordinance, and to prevent the enforcement and arrest the operation of the said acts and parts of acts of the Congress of the United States within the limits of this State.46

The people of South Carolina insisted that this was their constitutional right. And, in order to show their determination, ‘the people of South Carolina’ declared that it would not submit to the application of federal force. Were the federal government to employ military force against the State, such action would be deemed ‘as inconsistent with the longer continuance of South Carolina in the Union’.47 In 43   On the concept of ‘convention’ in American constitutionalism, see GS Wood, The Creation of the American Republic, 1776–1787 (Chapel Hill, University of North Carolina Press, 1998), Chapter 8. 44   How could nullification be reconciled with Article V of the U.S. Constitution? According to Calhoun, the general government could insist on its construction of a federal power, but it would have to change the Constitution (Calhoun, n 33, 356–60). For that, it would have to marshal the support of three-fourths of the States. Only were three-fourths of the States backed the federal construction, would the State nullification be overturned. Put the other way: nullification was to be final if supported by one quarter of the States. This ‘inversion’ of the amendment power was admitted to ‘have the effect of placing the majority under the control of the minority’, but this was justified as a necessary constitutional ‘check’ on the tyranny of the numerical majority. (This point would become central in Calhoun’s later constitutional writings as the ‘doctrine of concurrent majority’.) However, the doctrine of State sovereignty set substantive limits even to the amendment power. Thus, a State would not be bound by an Article V amendment ‘if it transcends the limits of the amending power – be inconsistent with the character of the constitution and the ends for which it was established – or with the nature of the system’. ‘In such a case, the State is not bound to acquiesce. It may choose whether it will, or whether it will not secede from the Union. One or the other course it must take. To refuse acquiescence, would be tantamount to secession; and place it as entirely in the relation of a foreign State to the other States, as would a positive act of secession’. (Calhoun, n 35, 212.) 45   The text of the Ordinance can be found in Commager, n 14, 261. 46   Ibid 261–62. 47   Ibid 262.

196  Robert Schütze such a situation, the people of South Carolina would ‘hold themselves absolved from all further obligation to maintain or preserve their political connection with the people of the other States, and will forthwith proceed to organize a separate government, and do all other acts and things which sovereign and independent States may of right do’.48 This was a threat to use the ultimate remedy: secession.

2.  The Nationalist View: ‘The Constitution not a Contract’ The ‘national’ reaction to this constitutional challenge was swift. Two weeks after the South Carolina Ordinance, President Jackson issued the ‘Proclamation regarding Nullification’.49 The latter was categorical: The ordinance is founded, not on the indefeasible right of resisting acts which are plainly unconstitutional and too oppressive to be endured, but on the strange position that any one State may not only declare an act of Congress void, but prohibit its execution; that they may do this consistently with the Constitution; that the true construction of that instrument permits a State to retain its place in the Union and yet be bound by no other of its laws than those it may choose to consider as constitutional. It is true, they add, that to justify this abrogation of a law it must be palpably contrary to the Constitution; but it is evident that to give the right of resisting laws of that description, coupled with the uncontrolled right to decide what laws deserve that character, it to give the power of resisting all laws.50

This position was incompatible with the nature of the Union. But what was the President’s view on the nature of the Union? The Constitution of the United States, then, forms a government, not a league, and whether it be formed by compact between the States or in any other manner, its character is the same. It is a government in which all the people are represented, which operates directly on the people individually, not upon the States; they retained all the power they did not grant. But each State having expressly parted with so many powers as to constitute, jointly with the other States, a single nation, can not, from that period, possess any right to secede, because such secession does not break a league, but destroys the unity of a nation; and any injury to that unity is not only a breach which would result from the contravention of a compact, but it is an offense against the whole Union. To say that any State may at pleasure secede from the Union is to say that the United States are not a nation, because it would be a solecism to contend that any part of a nation might dissolve its connection with the other parts, to their injury or ruin, without committing any offense.51

The Proclamation underlined that the ‘States severally have not retained their entire sovereignty’ as ‘they surrendered many of their essential parts of sovereignty’.52 The States had founded an (incomplete) nation and could not unilaterally secede. Despite its nationalist tone, the Jacksonian response was still based on the idea of   Ibid.   The text of Jackson’s Proclamation to the People of South Carolina (10 December 1832) can be found in Commager, n 14, 262 et seq. 50   Ibid 263. 51   Ibid 266 (emphasis added). 52   Ibid 267. 48 49

Federalism as Constitutional Pluralism  197 divided sovereignty and tolerant to the idea of the Constitution as a compact between the States.53 This would gradually change. For in trying to counter the ‘sovereign State’ view, ‘nationalist’ commentary gradually abandoned the contract theory of the union.54 The famous illustration of this strategic shift was Webster’s ‘The Constitution not a Compact between Sovereign States’.55 How did Webster reply to the ‘South Carolina doctrine’? Webster criticised the idea of a ‘constitutional compact’. ‘When applied to compacts between sovereign States, the term constitutional affixes to the word compact no definite idea’.56 And if one admits ‘our instrument of government to be a constitution, then, for that reason, it is not a compact between sovereigns; a constitution of government and a compact between sovereign powers being essentially unlike in their very natures, and incapable of being the same’.57 This theme was subsequently developed: What is a constitution? Certainly not a league, compact, or confederacy, but a fundamental law. The fundamental regulation which determines the manner in which the public authority is to be executed, is what forms the constitution of a State . . . The Constitution, Sir, is not a contract, but the result of a contract . . . The people have agreed to make a constitution; but when made, that constitution becomes what its name imports. It is no longer a mere agreement . . . The people of the several States had their separate State government . . . [and] with this condition of things the people were not satisfied . . . It was proposed, therefore, to erect a new, common government . . . This proposal was assented to, and the instrument was presented to the people of the several States for their consideration. They approved it, and agreed to adopt it, as a constitution, and henceforth it must stand as a constitution until it shall be altogether destroyed.58

Could the States secede from the Union? Secession was no constitutional remedy, but revolution.59 ‘[T]here can be no such thing as secession without revolution’. While admitting that the people in every State lived under two governments, each 53  KE Whittington, ‘The Political Constitution of Federalism in Antebellum America: The Nullification Debate as an Illustration of Informal Mechanisms of Constitutional Change’ (1996) 26 Publius 1, 14–17. 54   D Tipton, Nullification and Interposition in American Political Thought (Albuquerque, University of New Mexico, 1969) 31. 55   D Webster, The Constitution not a Compact between Sovereign States (London, Woodfall & Kinder, 1862). This famous speech, given in 1833, was a reply to John Calhoun’s speech ‘Introducing Resolutions Declaratory of the Nature and Power of the Federal Government’ of 22 January 1833. Calhoun had argued that the United States had been formed on the basis of a ‘constitutional compact’ to which the people of each State acceded as a separate sovereign community. The federal government was ‘not made the final judge of the powers delegated to it, since that would make its discretion, and not the Constitution, the measure of its powers’. ‘[A]s in all other cases of compact, among sovereign parties, without any common judge, each [state] has an equal right to judge for itself, as well of the infraction as of the mode and measure of redress’ (ibid 1–2). 56   Ibid 6. 57   Ibid 7. 58   Ibid 26–30. 59   Ibid 12. And a little later (ibid 18): ‘Sir, as soon as this ordinance [of nullification] shall be carried into effect, a revolution will have commenced in South Carolina. She will have thrown off the authority to which her citizens have heretofore been subject. She will have declared her own opinions and her own will to be above the laws and above the power of those intrusted [sic] with their administration. If she makes good these declarations, she is revolutionized.’

198  Robert Schütze having its separate sphere, the State peoples did not have a unilateral right to secede: ‘The State constitutions are established by the people of the States. Th[e] [federal] constitution is established by the people of all the States. How, then, can a State secede? How can a State undo what the whole people have done? How can she absolve her citizens from their obedience to the laws of the United States?’60 From this view also followed the illegitimacy of the theory of nullification. In the case of nullification a single State interposes his veto. The result would be that ‘no act of Congress can bind all the States, the constitutionality of which is not admitted by all; or, in other words, that no single State is bound, against its own dissent’.61 The theory of nullification thus violated the supremacy clause. Yet, Webster admitted that federal laws were only supreme within their sphere. The central question still was: Who is to construe finally the Constitution of the United States? We all agree that the Constitution is the supreme law; but who shall interpret that law? In our system of the division of powers between different governments, controversies will necessarily sometimes arise, respecting the extent of the powers of each. Who shall decide these controversies? Does it rest with the general government, in all or any of its departments, to exercise the office of final interpreter? Or may each of the States, as well as the general government, claim this right of ultimate decision? The practical result of the whole debate turns on this point. The gentleman contends that each State may judge for itself of any alleged violation of the Constitution, and may finally decide for itself, and may execute its own decisions by its own power. All the recent proceedings in South Carolina are founded on this claim of right. In my opinion, Sir, even if the Constitution of the United States had made no express provision for such cases, it would yet be difficult to maintain, that, in a Constitution existing over four-and-twenty States, with equal authority over all, one could claim a right of construing it for the whole. This would seem a manifest impropriety; indeed, an absurdity . . . Congress must judge of the extent of its own powers so often as it is called on to exercise them, or it cannot act at all; and it must also act independent of State control, or it cannot act at all. The right of State interposition strikes at the very foundation of the legislative power of Congress. It possesses no effective legislative power, if such right of State interposition exists; because it can pass no law subject to abrogation. It cannot make laws for the Union, if any part of the Union may pronounce its enactments void and of no effect.62

For the ‘nationalist’ Webster, the competence to rule on competence thus belongs to the federal government, and to the courts of the United States. ‘Congress may judge of the extent of its own powers’; and what was true for the legislature was ‘still more express and emphatic’ for the judiciary.63 This followed from ‘a settled axiom in politics that every government must have a judicial power co-extensive with its legislative power’.64 In decisions relating to the federal   Ibid 44.   Ibid 20. 62   Ibid 45–49 (emphasis added). 63   Ibid 49–50. 64   Ibid 50 and 52. This had also been the view of A Hamilton in The Federalist No 80 (n 3, 387): ‘If there are such things as political axioms, the propriety of the judicial power of a government being co-extensive with its legislative, may be ranked among the number.’ In the European Union legal order, this is – wrongly – contested by Professor JHH Weiler. For a critical analysis of this mistaken view, see 60 61

Federalism as Constitutional Pluralism  199 boundary between the States and the Union, the tribunal that was thus ultimately to decide was the Supreme Court. ‘If the general legislature should, at any time, overleap their limits, the judicial department is a constitutional check’.65 Nullification was thus ‘as distinctly revolutionary as secession’. And, it was even less ‘respectable’ than secession, since it did not avoid the contradiction of wishing to belong to the Union while rejecting its authority.66

C.  Excursus: The ‘Civil War’ as (Extra)Constitutional Conflict Despite its constitutional victory in 1833,67 the federal government had lost a philosophical battle in the South. The doctrine of State sovereignty emerged strengthened from the tariff crisis.68 It would, in the following decades, gain respectability in constitutional politics. With the slavery question hardening into a divisive constitutional conflict, it was a matter of time until the manufacturing North and the agricultural South would fatally clash. In 1852, South Carolina declared that the continuing encroachments upon the reserved powers of the States would justify its withdrawal from the American Union. (It only declined to exercise her right ‘in deference to the opinions and wishes of the other slaveholding States’.69) Considering these encroachments to have become unbearable, it took the final step in 1860 and adopted the ‘Ordinance to dissolve the union between the State of South Carolina and the other States united with her under the compact entitled “The Constitution of the United States of America” ’.70 Ten States would follow its lead. For the North this was a ‘rebellion’. The contrasting constitutional visions between the North and the South on the question of sovereignty led to the ‘Civil War’.71 R Schütze, From Dual to Cooperative Federalism: The Changing Structure of European Law (Oxford, Oxford University Press, 2009) 155.  Webster, The Constitution not a Compact, n 55, 55.   Ibid 62.   In 1833, Congress adopted the Force Bill authorizing the President to enforce the tariff, but also the ‘Compromise Tariff ’ that became law on the same day as the Force Bill. The latter induced South Carolina to repeal its nullification ordinance by the same convention that had passed it. 68   The nullification crisis, in the opinion of an American commentator, ‘created the concepts and some of the political conditions that eventually led to the Civil War’. RE Ellis, The Union at Risk: Jacksonian Democracy, States’ Rights and the Nullification Crisis (Oxford, Oxford University Press, 1987) 198. 69  Cf ‘Declaration of the Immediate Causes which induce and justify the Secession of South Carolina from the Federal Union’, in Commager, n 14, 373. 70   For the text of the Ordinance see Commager, n 14, 372. It read: ‘We, the people of the State of South Carolina, in convention assembled, do declare and ordain, and it is hereby declared and ordained, that the ordinance adopted by us in Convention on the 23d day of May, in the year of our Lord 1788, whereby the Constitution of the United States of America was ratified, and also all Acts and parts of Acts of the General Assembly of this State ratifying amendments of the said Constitution, are hereby repealed; and that the union now subsisting between South Carolina and other States under the name of the United States of America is hereby dissolved.’ 71  On the various theories of the ‘civil war’, see Kelly et al, n 4, 298–301. According to President Buchanan – Lincoln’s predecessor – secession should not have been countered by force. ‘The question fairly stated is, Has the Constitution delegated the Congress the power to coerce a State into submission which is attempting to withdraw or has actually withdrawn from the Confederacy? If answered in the affirmative, it must be on the principle that the power has been conferred upon Congress to declare and 65 66 67

200  Robert Schütze What was the constitutional philosophy of the North during the Civil War? We find a powerful ‘nationalist’ interpretation of the American Union in President Lincoln’s address on 4 July 1861.72 For the President, it was unthinkable that ‘any state of the Union may, consistently with the national Constitution, and therefore lawfully, and peacefully, withdraw from the Union, without the consent of the Union, or of any other state’.73 A unilateral right to secede had not only unjust practical consequences.74 From a (national) theoretical perspective it was wrong: This sophism derives much – perhaps the whole – of its currency, from the assumption, that there is some omnipotent, and sacred supremacy, pertaining to a State – to each State of our Federal Union. Our States have neither more, nor less power, than that reserved to them, in the Union, by the Constitution – no one of them ever having been a State out of the Union. The original ones passed into the Union even before they cast off their British colonial dependence . . . [In the Declaration of Independence] the ‘United Colonies’ were declared to be ‘Free and Independent States’; but, even then, the object plainly was not to declare their independence of one another, or of the Union; but directly the contrary, as their mutual pledge, and their mutual action, before, at the time, and afterwards, abundantly show. Having never been States, either in substance, or in name, outside of the Union, whence this magical omnipotence of ‘State rights’, asserting a claim of power to lawfully destroy the Union itself? Much is said about the ‘sovereignty’ of the States; but the word, even, is not in the national Constitution; nor, as is believed, in any of the State constitutions . . . The Union is older than any of the States; and, in fact, it created them as States. Originally, some dependent colonies made the Union; and, in turn, the Union threw off their old dependence, for them, and made them States, such as they are. Not one of them ever had a State constitution, independent of the Union.75

Lincoln’s North won the Civil War. This extra-constitutional ‘fact’ henceforth discredited the theory of State sovereignty.76 The constitutional ‘reconstruction’ that followed was the greatest federal reform in the history of the American to make war against a State. After much serious reflection I have arrived at the conclusion that no such power has been delegated to Congress or to any other department of the Federal Government . . . The fact is that our Union rests upon public opinion, and can never be cemented by the blood of its citizens shed in civil war.’ (Cf J Buchanan, ‘States and Withdrawal from the Union’ (Fourth Annual Message, 3 December 1869), as quoted in Commager, n 14, 368–69.) 72   A Lincoln, ‘Message to Congress in Special Session’ (4 July 1861), in RP Basle (ed), The Collected Works of Abraham Lincoln. Volume IV (New Brunswick, Rutgers University Press, 1953) 421 et seq. 73   Ibid 433. 74  ‘The nation purchased, with money, the countries out of which several of these States were formed. Is it just that they shall go off without leave, and without refunding? The nation paid very large sums, (in the aggregate, I believe, nearly a hundred millions) to relieve Florida of the aboriginal tribes. Is it just that she shall now be off without consent, or without making any return? The nation is now in debt for money applied to the benefit of these so-called seceding States, in common with the rest. Is it just, either that creditors shall go unpaid, or the remaining States pay the whole? A part of the present national debt was contracted to pay the old debts of Texas. Is it just that she shall leave, and pay no part of this herself? Again, if one State may secede, so may another; and when all shall have seceded, none is left to pay the debts. Is this quite just to creditors? Did we notify them of this sage view of ours, when we borrowed their money?’ ibid 435–36). 75   Ibid 433–35. 76   ‘Secession failed the test of arms. The Civil War did in fact repudiate the principle of secession as a constitutional resort of sovereign states.’ Tripton, n 54, 50.

Federalism as Constitutional Pluralism  201 Union.77 The Civil War Amendments caused an unprecedented shift towards the nationalist end of the federal spectrum. Emblematic for this ‘consolidation’ of the Union was the Fourteenth Amendment.78 It not only inverted the relationship between State and federal citizenship. It subjected the States – for the first time – to the homogenizing forces of national fundamental rights. The United States had moved towards a more United State(s). Yet, the American doctrine of divided sovereignty had survived the Civil War, and with it the constitutional debate on the nature of the Union. The meaning of the ‘We the people’ formula would continue to attract a plurality of constitutional visions in the postbellum era. They will be discussed in the next section.

III.  Constitutional Conflicts after the Civil War While the Civil War was a constitutional break, it caused no total revision of the American Constitution. There were continuities as well as discontinuities with the constitutional status quo ante. Most importantly: constitutional theory returned to the idea of divided sovereignty.79 The Union had split the atom of sovereignty and was, therefore, neither ‘Staatenbund’ nor ‘Bundesstaat’.80 Constitutional 77   B Ackerman goes as far as to claim that reconstruction was a ‘re-founding’. However, it may be too strong a claim to insist that ‘the First Reconstruction Act is functionally equivalent to Article Seven of the 1787 Constitution’ (B Ackerman, We the People: Transformations (Harvard University Press, Cambridge, Mass. 2000), 198–99). 78   The first section of the Fourteenth Amendment states: ‘All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside. No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws.’ For an analysis of the (various) constitutional positions on the Fourteenth Amendment, see AR Amar, ‘The Bill of Rights and the Fourteenth Amendment’ (1991–2) 101 Yale Law Journal 1193. 79   WH Bennett, American Theories of Federalism (Tuscaloosa, University of Alabama Press, 1967), 179. See also Haines, n 7, 107: ‘Not even the civil war fought largely on the issue of state versus national sovereignty could destroy the belief in the divisibility of sovereignty in a federal system such as that established by the Constitution.’ 80   German federal thought was never widely adopted in the United States. However, we can see the dichotomy between ‘Bundesstaat’ and ‘Staatenbund’ enter into the constitutional discourse through two outstanding individuals: Alexander Stephens and Woodrow Wilson. Stephens – ‘the’ Southern theorist of nineteenth-century constitutional law – continued to insist that the American Union was part of the ‘genus’ of ‘Staatenbund’: ‘Our system, taken altogether, we have seen, is a peculiar one. The world never saw its like before. It has no prototype in any of all the previous Confederations, or Federal Republics, of which we have any account. It is neither a ‘Staaten-bund’ exactly, nor a ‘Bundesstaat’, according to the classification of Federal Republics by the German Publicists. It differs from their ‘Staaten-bund’ in this, that the powers to be exercised by the Federal Head are divided into three departments, the Legislative, Judicial, and Executive, with a perfectly organized machinery for the execution of these powers within its limited sphere, and for the specific objects named, upon citizens of the several States without the intermediate act or sanction of the several States. In the ‘Staatenbund’, or ‘States’ Confederation’, according to their classification, the Federal Government can enact no laws which will operate upon the citizens of the several States composing it, until the States severally

202  Robert Schütze conflicts gradually ebbed down: ‘Interest in the nature of sovereignty and in pinpointing its location in the union waned as the country got farther and farther away from the War.’81 But this constitutional peace was not to last. The balance between ‘national’ and ‘international’ elements within the American Constitution would be sig­ nificantly challenged in the twentieth century. The ‘new nationalism’ expressed itself in two – interrelated – ways. First, the ‘New Deal’ radically shifted the federal boundary in favour of the ‘nation’ as the Union would be allowed to legislate in areas that were previously viewed as exclusive police powers of the States.82 Federal laws increasingly posed an external limit to State powers. Secondly, through the doctrine of ‘incorporation’ of federal rights,83 the States would encounter ever more internal limits to their powers: national individual rights. These two developments – eventually – broke the constitutional peace between the States and the Union. New constitutional conflicts emerged and with them the old language of interposition and State sovereignty.

give them their sanction. Such was our Federal Union under the first Articles. But our present system, as we have seen, went a step further, and introduced a new principle in Confederations. While, therefore, our system differs specifically in this particular from their ‘Staaten-bund’, or ‘States’ Confederation’, yet it agrees entirely with it in its essential Generic difference from their Bundesstaat, in this, that the States collectively constitute an international unit as regards third parties, but do not cease to be international units as regards each other. It differs further Generically from their ‘Bundesstaat’, or ‘Federative State’, or what may properly be called ‘an incorporate Union’, in this, that no Sovereign Power whatever, under our system, is surrendered or alienated by the several States; it is only delegated. The difference between our system and their ‘Staaten-bund’, is, however, only specific, as we see. It is not Generic. They are both essentially the same. Ours is a newly developed species of Government of their Genus ‘Staaten-bund’. This specific difference is what struck De Tocqueville as ‘a wholly novel theory, which may be considered as a great discovery in modern political science’, and for which there was as yet no specific name’. Cf A Stephens, A Constitutional View of the Late War between the States. Volume II (Philadelphia, National Publishing Company, 1870) 18–19. By contrast, Wilson argued that the American Union was a ‘Federal State’. Cf W Wilson, An Old Master And Other Political Essays (Whitefish, Kessinger, 2006). The Union was a sovereign state that had ‘competence-competence’ (Ibid 93–94): ‘In the federal state self-determination with respect to their law as a whole has been lost by the member states. They cannot extend, they cannot even determine, their own powers conclusively without appeal to the federal authorities. They are unquestionably subject to a political superior. They are fused, subordinated, dominated.’ While the States have ‘dominion’, the Union has ‘sovereignty’: ‘For with the federal state lie the highest powers of originative [sic] legal determination, the ultimate authority to warrant change and sanction jurisdiction’. On the ‘German’ concept of ‘competence-competence’ see Schütze, n 64, 34–36. 81   Bennett, n 79, 195. While this was true for the internal sphere, sovereignty as a concept would still be employed in the twentieth century in the context of the international powers of the United States. Cf United States v Curtiss-Wright Export Corp, 299 U.S. 304 (1936), 315–18: ‘[T]he investment of the federal government with the powers of external sovereignty did not depend upon the affirmative grants of the Constitution’. ‘[They are] vested in the federal government as necessary concomitants of nationality.’ For a discussion of the ‘exceptional’ nature of (American) foreign affairs federalism see Schütze, n 64, 108–22. 82   For an analysis see Schütze, n 64, 80 et seq. 83   On the meaning and development of the ‘incorporation’ doctrine see Amar, n 78.

Federalism as Constitutional Pluralism  203

A.  Pluralist Traditions Revived: Brown v Board of Education The unitary potential of – nationally defined – individual rights, and their potential conflict with States’ rights, can be seen in Brown v Board of Education.84 Brown was a parent from Topeka in Kansas. His daughter was forced to attend a segregated school, because of a Kansas law that permitted school districts to establish separate elementary schools for black and white pupils. Together with 12 other parents, Brown brought proceedings against the Board of Education of his city on the basis of the Fourteenth Amendment. This was innovative. The Supreme Court had previously held that racial segregation was constitutional if it conformed to the ‘separate but equal’ formula;85 and 17 States had racial segregation legislation, when the question came once more to the Supreme Court. To the surprise of the Southern States, the Court concluded ‘that in the field of public education the doctrine of “separate but equal” has no place’. Separate educational facilities were ‘inherently unequal’ and constituted, as such, a violation of the Fourteenth Amendment.86 The State attack on the Supreme Court was swift and aggressive, with the judgment being vehemently decried in the South. ‘The Supreme Court ha[d] run into a storm of protest, as severe as it has ever encountered.’87 Feelings ran so high that after decades of – relative – constitutional peace, five Southern States revived the ideas of interposition and nullification.88 In February 1956, the Virginia legislature thus adopted a Resolution entitled ‘Interposing the sovereignty of Virginia against encroachment upon the reserved powers of this State, and appealing to sister states to resolve a question of contested power’.89 It read as follows: That the General Assembly of Virginia expresses its firm resolution to maintain and to defend the Constitution of the United States, and the Constitution of this State, against every attempt, whether foreign or domestic to undermine the dual structure of this Union, and to destroy those fundamental principles embodied in our basic law, by which the delegated powers of the Federal Government, and the reserved powers of the respective States have long been protected and assured. That this Assembly explicitly declares that the powers of the Federal Government result solely from the compact to which the Sates are parties, and that the powers of the Federal Government, in all its branches and agencies, are limited by the terms of the instrument creating the compact, and by the plain sense and intention of its provisions.

  Brown v Board of Education of Topeka, 347 U.S. 483 (1954).   Plessy v Ferguson, 163 U.S. 537 (1896). 86   Brown v Board of Education of Topeka (n 81), 495. The Segregation cases led to ‘the most extensive presentation of historical materials ever made to the Court’. AM Bickel, ‘Original Understanding and the Segregation Decision’ (1955–6) 69 Harvard Law Review 1, 6. Famously, Chief Justice Warren held the historical evidence ‘inconclusive’. Cf Brown, n 84, 489. 87   C Fairman, ‘The Supreme Court 1955 Term. Foreword: The Attack on the Segregation Cases’ (1956) 70 Harvard Law Review 83. 88   Various State resolutions (extracts) can be found in AS Miller and RF Howell, ‘Interposition, Nullification and the Delicate Division of Power in a Federal System’ (1956) 5 Journal of Public Law 2. 89   The text can be found in Anonymous, n 6, 445 et seq. 84 85

204  Robert Schütze That by its decision of May 17, 1954, the school cases, the Supreme Court of the United States placed upon the Constitution an interpretation, having the effect of an amendment thereto, which interpretation Virginia emphatically disapproved. That the State of Virginia did not agree, in ratifying the Fourteenth Amendment, nor did other States ratifying the Fourteenth Amendment agree, that the power to operate racially separate schools was to be prohibited to them thereby; and as evidence of such understanding of the terms of the amendment, and its plain sense and intention, the General Assembly of Virginia notes that the very Congress which proposed the Fourteenth Amendment for ratification established separate schools in the District of Columbia. [The] declaration upon the part of the Supreme Court of the United States constitutes a deliberate, palpable, and dangerous attempt by the court itself to usurp the amendatory power that lies solely with no fewer than three-fourths of the States. That the General Assembly of Virginia, mindful of the resolution it adopted on December 21, 1798, and cognizant of similar resolutions adopted on like occasions in other States, both North and South, again asserts this fundamental principle: That whenever the Federal Government attempts a deliberate, palpable and dangerous exercise of powers not granted it, the States who are parties to the compact have the right, and are in duty bound, to interpose for arresting the progress of the evil, and for reserving the authorities, rights and liberties appertaining to them . . . That failure on the part of this State thus to assert her clearly reserved powers would be construed as tacit consent to the surrender thereof and that such submissive acquiescence to palpable, deliberate and dangerous encroachments upon one power would in the end lead to the surrender of all powers, and inevitably to the obliteration of the sovereignty of the States contrary to the sacred compact by which this Union of States was created.90

The Virginia Resolution reinstated the idea of the Constitution as a compact concluded by ‘sovereign’ States. The powers of the Federal Government were delegated and enumerated powers; thus, the Supreme Court – an agent of the Federal government – could not be the exclusive and ultimate arbiter of constitutional conflicts. It followed from the ‘contractual’ nature of the Constitution, that the States had the ‘right to interpose’. The 1956 Virginia Resolution here borrowed from the antebellum tradition and its famous 1798 predecessor. To underline the constitutional continuity with the ‘Principles of ‘98’, it even adopted a ‘Report’ on the ‘Doctrine of Interposition: Its History and Application’.91 In ‘following humbly the example of Mr Madison in 1799’,92 the Report offered a detailed commentary of the constitutional philosophy underlying the legislative protest.93 The Union ‘was created by the people of the separate States, acting   Ibid, paras 1–2, 5–6, 8 and 10 (emphasis added).   Senate of Virginia, The Doctrine of Interposition: Its History and Application – A Report on Senate Joint Resolution 3 (Commonwealth of Virginia, Division of Purchase and Printing, Richmond 1957). 92   Ibid 5. 93   This is how the Report itself summed up its elementary assumptions: ‘The right of a State to interpose its sovereign powers against encroachment by the Federal Government rests upon certain assumptions in history and law. These are: First, that when the colonies dissolved the political bands that had connected them with Great Britain, they became precisely what they declared themselves to be: Free and Independent States. Second, that in uniting under the Articles of Confederation, and later under the Constitution of 1787, the States acted as separate, individual States. Third, that the people of the States, in agreeing to the constitutional compact have delegated only certain enumerated powers to 90 91

Federalism as Constitutional Pluralism  205 separ­ately as States’.94 ‘[T]he States formed the constitutional compact as an agreement among themselves as individual sovereign political entities’.95 On the basis of this constitutional compact, powers were delegated to the Federal Government. Conceding ‘that the Supreme Court had authority to interpret the Constitution’, what was important was ‘that boundary line – and there must be some boundary line capable of being fixed and defined by an agency other than the court itself – where mere interpretation ceases and substantive amendment begins’.96 It followed ‘that the States, who under Article V alone have the power to amend the Constitution, also must have the power to contest effectively an attempted amendment by judicial construction’.97 Where did this line between ‘interpretation’ and ‘amendment’ lie? According to Virginia, the Court could not strike ‘at the most intimate social and political institutions of the States’ and ignore the original understanding or settled constitutional practice.98 What constitutional remedy did the States enjoy, when the federal government had acted ultra vires? ‘[E]very State has a clear constitutional right as a party to the compact to allege an infraction of the compact. To allege is not to prove; to allege is not to nullify; to allege is not to threaten armed resistance. It is simply to charge an infraction.’99 The 1956 Virginia Resolution thus would not recommend positive defiance, but asserted the right and duty of interposition against federal usurpation for one purpose only – ‘for arresting the progress of evil’. ‘The effort is to bring enforcement of the court’s challenged mandates to a pause, to an intermediate position where matters may be held for a time in statu quo until opportunity may be had for careful consideration.’ ‘Our appeal is explicitly directed to our sister States: it is generally directed to public opinion.’100 In essence: the constitutional remedy was ‘interposition’ and not (unilateral) nullification.101 the general government, and have reserved all other powers to their States or to themselves. Fourth, that when the general government usurps powers not delegated, the States have an inalienable right to interpose their sovereign powers so as to arrest the progress of the evil. Fifth, that the question of such encroachment cannot properly be decided by an agent of the general government itself, but can only be decided by the States themselves as parties to the compact’ (ibid 5). 94   Ibid 6. 95   Ibid 10. 96   Ibid 11. 97   Ibid 12. 98   Ibid 14. 99   Ibid 15. 100   Ibid 20. One expression of (Southern) public opinion was the ‘Southern Manifesto’ (1956), in which Southern members of Congress denounced the federal abuse of power. The Manifesto read: ‘We regard the decision of the Supreme Court in the school cases as clear abuse of judicial power. It climaxes a trend in which the Federal judiciary under-taking to legislate, in derogation of the authority of Congress, and to encroach upon the reserved rights of the states and the people’. ‘Even though we constitute a minority in the present Congress, we have full faith that a majority of the American people believe in the dual system of government which has enabled us to achieve our greatness and will in time demand that the reserved rights of the states and of the people be made secure against judicial usurpation’. Extracts of the ‘Southern Manifesto’ can be found in FD Drake and LR Nelson (eds), States’ Rights and American Federalism (Westport, Greenwood Press, 1999), 203–05. 101   By contrast, Georgia’s ‘Interposition Resolution’ of 9 March 1956 declared ‘[t]hat said decisions and orders of the Supreme Court of the United States relating to separation of the races in the public institutions of a State as announced and promulgated by said court on May 17, 1954, and May 31,

206  Robert Schütze This was, undoubtedly, the constitutional theory of 1798. But had not the constitutional context changed after the Civil War? ‘We have inquired whether the melancholy events of 1861–65, and the subsequent changes made to the Constitution, have in any wise affected the fundamental principles asserted to be still embodied in our basic law’.102 Virginia did not think so. Its resolutions expounded ‘in 1956 as they did in 1798, the true nature of the Federal Government and the relationship of the States to it’.103 Yet, Virginia’s 1956 Resolution was soon to encounter the national reaction. After continued defiance in the Southern States, the constitutional conflict over Brown was ‘resolved’ by the threat of arms. President Eisenhower sent federal troops to enforce the judicial order. Southern protest would thereafter be channelled into proposals to amend the Constitution.104 The most daring proposal in this respect was the suggestion to create a new ‘Court of the Union’, which would have jurisdiction on ‘federal questions’ and empowered to overrule the Supreme Court.105 The proposal – unsurprisingly – failed.106

B.  ‘Split’ Opinions: Constitutional Pluralism in the Supreme Court Constitutional pluralism has not been confined to – incommensurable – positions between the States and the Union. Different constitutional visions may even be found within the federal government. We find an illustration of competing visions over the question of sovereignty in the Supreme Court’s decision in U.S. Term Limits v Thornton.107 The Court had been asked to rule on a State constitutional amendment that barred candidates for Congress from appearing on the general election ballot if the candidate had already served three terms in the House of Representatives or two terms in the Senate. Did the federal Constitution allow individual States to adopt their own qualifications for congressional service; or was this an exclusively federal reserve? The Constitution did not expressly solve this question.108 The Court decided that neither Congress nor the States could add 1955, are null, void and of no force or effect’. (Extracts of the Georgia Resolution can be found in: HO Reid, ‘The Supreme Court Decision and Interposition’ (1956) 25 Journal of Negro Education 109, 114.) 102   The Doctrine of Interposition, n 91, 6. 103   Ibid. 104   PL Hanes, ‘The Proposed Constitutional Amendments: A New Definition of Federalism’ (1963) 12 Journal of Public Law 448. 105   Section 1 of the proposed amendment read: ‘Upon demand of the legislatures of five states, no two of which shall share a common boundary, made within two years after the rendition of any judgment of the Supreme Court relating to the rights reserved to the states or to the people by this Constitution, such judgment shall be reviewed by a Court composed of the several states to be known as the Court of the Union. The sole issue before the Court of the Union shall be whether the power or jurisdiction sought to be exercised on the part of the United States is a power granted to it under this Constitution’. 106   For a similar (if not borrowed) proposal in the European Union, see JHH Weiler, The Constitution of Europe (Cambridge, Cambridge University Press, 1999) 322. 107   U.S. Term Limits, Inc. v Thornton, 514 U.S. 779 (1995). 108   Article I, Section 2, Clause 2 of the Constitution states that ‘[n]o person shall be a Representative who shall not have attained to the Age of twenty five Years, and been seven Years a Citizen of the United

Federalism as Constitutional Pluralism  207 conditions to the ‘Qualification Clauses’ in the Constitution. However, the tight five to four decision confirmed the constitutional presence of ‘duelling sovereignties’ as part and parcel of the American Union. The case was indeed a ‘showdown’ over constitutional federalism.109 Judge Stevens spoke for the majority. Referring to past jurisprudence, he invoked the framers’ understanding that the qualifications for members of Congress had been fixed by the Constitution.110 First, we conclude that the power to add qualifications is not within the ‘original powers’ of the States, and thus is not reserved to the States by the Tenth Amendment. Second, even if the States possessed some original power in this area, we conclude that the Framers intended the Constitution to be the exclusive source of qualifications for Members of Congress, and that the Framers thereby ‘divested’ States of any power to add qualifications.111

Justice Kennedy concurred. The selection of the national legislature was a national matter, and as such belonged exclusively to the Nation: Federalism was our Nation’s own discovery. The Framers split the atom of sovereignty. It was the genius of their idea that our citizens would have two political capacities, one state and one federal, each protected from incursion by the other . . . The political ident­ity of the entire people of the Union is reinforced by the proposition, which I take to be beyond dispute, that, though limited as to its objects, the National Government is, and must be, controlled by the people without collateral interference by the States.112

Judge Thomas spoke for the dissenters.113 Claiming that ‘the majority fundamentally misunderstands the notion of “reserved” powers’, the dissent went back to ‘some first principles’.114 What were the fundamental principles underlying the 1787 Constitution? The constitutional vision of the dissenters was this: Our system of government rests on one overriding principle: all power stems from the consent of the people. To phrase the principle in this way, however, is to be imprecise about something important to the notion of ‘reserved’ powers. The ultimate source of the Constitution’s authority is the consent of the people of each individual State, not the consent of the undifferentiated people of the Nation as a whole. The ratification procedure erected by Article VII makes this point clear. The Constitution took effect once it had been ratified by the people gathered in conventions in nine different States. But the States, and who shall not, when elected, be an Inhabitant of that State in which he shall be chosen’. However, Article I, Section 4, Clause 1 gives States the power to regulate the ‘Times, Places and Manner of holding Elections’, but Congress is given the power to ‘make or alter such Regulations’. Finally, Article I, Section 5, Clause 1 concedes that ‘[e]ach House shall be the Judge of the Elections, Returns and Qualifications of its own members’. 109   KM Sullivan, ‘Duelling Sovereignties: U.S. Term Limits, Inc. v Thornton’ (1995) 109 Harvard Law Review 79. 110   U.S. Term Limits, n 107, 792. 111   Ibid 800–01. 112   Ibid 838 and 841. 113   They were Justice Thomas, Chief Justice Rehnquist, Justice O’Connor and Justice Scalia. Their dissent exceeds the court’s opinion by 25 pages. 114   U.S. Term Limits, n 107, 856.

208  Robert Schütze Constitution went into effect only ‘between the States so ratifying the same’, Art. VII; it did not bind the people of North Carolina until they had accepted it. In Madison’s words, the popular consent upon which the Constitution’s authority rests was ‘given by the people, not as individuals composing one entire nation, but as composing the distinct and independent States to which they respectively belong.115

What did these ‘first principles’ mean in the present case? [T]he Constitution calls for Members of Congress to be chosen State by State, rather than in nationwide elections . . . In short, the notion of popular sovereignty that undergirds the Constitution does not erase state boundaries, but rather tracks them . . . Political scientists can debate about who commands the ‘primary allegiance’ of the Members of Congress once they reach Washington. From the framing to the present, however, the selection of the Representatives and Senators from each State has been left entirely to the people of that State or to their State legislature . . . The very name ‘congress’ suggests a coming together of representatives from distinct entities . . . Although the United States obviously is a Nation, and although it obviously has citizens, the Constitution does not call for Members of Congress to be elected by the undifferentiated national citizenry; indeed, it does not recognize any mechanism at all (such as a national referendum) for action by the undifferentiated people of the Nation as a whole.116

Thus, ‘when it comes to the selection of Members of Congress, the people of each State have retained their independent political identity’.117 This meant that each State people itself could decide on the qualifications ‘its’ congressional members should have.118 Some have seen the dissent in Term Limits as a ‘manifesto: a declaration of the primordial sovereignty of the states’.119 This goes – perhaps – too far, unless ‘primordial’ is understood as ‘residual’. Indeed, the dissenters in Term Limits clarified their position in Aden v Maine,120 but now speaking as the majority for the Court.121 The case concerned the status of the States as litigants. Could an individual bring proceedings against its State, without the latter’s consent, in the State’s own courts? The textual limits to the sovereign immunity expressed in the Eleventh Amendment were ambiguous on this point.122 The Court held that   Ibid 846, referring to The Federalist No 39, emphasis added.   Ibid 849 (referring to Marshall in McCulloch v Maryland, 17 US 316 (1819), 403: ‘No political dreamer was ever wild enough to think of breaking down the lines which separate the States, and of compounding the American people into one common mass.’), 857 and 859. 117   Ibid 860. 118   Ibid 869. 119   Sullivan, n 109, 109. 120   Alden et al v Maine, 527 U.S. 706 (1998). For a (very) critical appraisal of the judgment, see L Weinberg, ‘Of Sovereignty and Union: The Legends of Alden’ (2001) 76 Notre Dame Law Review 1113. 121   The majority consisted of Justice Kennedy, Chief Justice Rehnquist, Justice O’Connor, Justice Scalia and Justice Thomas. 122   The Eleventh Amendment, introduced to overrule Chisholm v Georgia, n 31, stated: ‘The Judicial power of the United States shall not be construed to extend to any suit in law or equity, commenced or prosecuted against one of the United States.’ While confirming the sovereign immunity of States in federal courts, it did – arguably – not provide a conclusive answer in the context of a State’s own courts. 115 116

Federalism as Constitutional Pluralism  209 ‘sovereign immunity’ need not derive from the Eleventh Amendment, but rooted in the original structure of the 1787 Constitution. ‘[T]he States’ immunity from suit is a fundamental aspect of the sovereignty which the States enjoyed before the ratification of the Constitution, and which they retain today’.123 But the Court did not postulate that the States had retained their complete sovereignty after ratification. It affirmed the principle of divided sovereignty on which the American Union was founded. Under the 1787 Constitution the States had only retained a ‘residuary’ sovereignty.124 The States had accepted a limitation of their sovereignty on the basis of their ‘constitutional compact ’.125 This vision had important consequences for the nature of the Union and the powers of the federal government. ‘Although the Constitution begins with the principle that sovereignty rests with the people, it does not follow that the National Government becomes the ultimate, preferred mechanism for expressing the people’s will’. ‘We the people’ was thus not the (unitary) ‘We, the American people’. In establishing the 1787 Constitution, ‘the people insisted upon a federal structure for the very purpose of rejecting the idea that the will of the people in all instances is expressed by the central power, the one most remote from their control’.126 The court thus accepted the idea that in federal republics, there are two democratic constituencies: the federal people and the State people. The national people – even less the national government – could not always be the ultimate interpreter of the Constitution. And since federal and State loyalties co-exist and compete within ‘the people’, constitutional conflicts are bound to occur – even in mature federal orders. The question of sovereignty thus remained indeterminate.127

IV.  Conclusion: Federalism as Constitutional Pluralism The (metaphysical) question of sovereignty has never been a serious obsession of American constitutionalism. Concentrating on (physical) constitutional structures, this federal tradition invented the idea of ‘divided’ or ‘dual’ sovereignty. In the words of an early commentator of the 1787 Constitution, the latter was based on the idea that ‘[t]wo sovereignties are necessarily in presence of each other’.128 But who were these two sovereigns? According to the idea of popular sovereignty,

  Alden v Maine, n 120, 713 (emphasis added).   Ibid 715 (referring to The Federalist No 39). The dissenters equally embraced the concept of divided sovereignty. In the words of Justice Souter (ibid 799): ‘The National Constitution formally and finally repudiated the received political wisdom that a system of multiple sovereignties constituted the ‘great solecism of an imperium in imperio’. 125   The court expressly uses this term: ibid 708 (emphasis added). 126   Ibid 759. 127   D Farber, Lincoln’s Constitution (Chicago, University of Chicago Press, 2003) 28: ‘Apparently, the sovereignty issues remains unsettled, even today.’ 128   A de Tocqueville, n 1, 172. 123 124

210  Robert Schütze they were the State peoples and the – constitutionally posited – American people.129 Each citizen belonged to two political bodies. The existence of two political ‘sovereigns’ would give rise to constitutional conflicts. In federal orders, constitutional conflicts are no anomaly. They are federal normality: the evidence of a living federation.130 This Chapter looked at four constitutional conflicts in the history of the United States – two before and two after the Civil War. Each historical episode invoked the ‘ultimate arbiter’ question. ‘Who is to decide constitutional conflicts?’ While the crisis over the Alien and Sedition Acts elaborated on the theory that the State governments could ‘interpose’, the ‘Principles of ‘98’ were in conformity with the idea of divided sovereignty. Three decades later, the tariff crisis generated the idea of State sovereignty. As sovereigns, each State was said to be entitled to ultimately interpret the Constitution and to ‘nullify’ federal laws that were viewed as unconstitutional. Conflict over the – sublimated – question of sovereignty would lead to the Civil War. The latter discredited the constitutional philosophy of State sovereignty (and the right of secession). However, as the second part of this article has shown, postbellum constitutionalism has not led to the (European) idea of national sovereignty. American constitutionalism continues to subscribe to the idea of divided sovereignty. Constitutional pluralism thus continues to exist. We saw this in the form of the revived the theory of interposition in Brown and the ‘plural’ voices within the Supreme Court in Term Limits. What can this ‘Letter from America’ offer to European constitutionalism? European constitutionalism has much to learn from American constitutional theory (and vice versa). Europe’s constitutional tradition continues to insist on the idea of undivided sovereignty.131 This poses – unsolvable – problems for an analysis of the political and constitutional dualism that characterizes the European Union. For a tradition that (tacitly or expressly) relies on the – unitary – concept 129   On the ‘invention’ of the American people, see ES Morgan, Inventing the People: The Rise of Popular Sovereignty in England and America (New York, Norton, 1989) esp Ch 11. For the nineteenthcentury (Southern) view that there was no ‘American people’, see HS Tucker, Lectures on Constitutional Law (Richmond, Shepherd & Colin, 1843) 95: ‘For as there is no people of the United States, considered aggregately, the sovereignty must be in the people of each State.’ See also AP Upshur, A Brief Enquiry into the True Nature and Character of Our Federal Government: Being a Review of Judge Story’s Commentaries on the Constitution of the United States (Philadelphia, Campbell, 1863) 92: ‘In the States the sovereign power is in the people of the States respectively; and the sovereign power of the United States would, for the same reason, be in “the people of the United States”, if there were any such people, known as a single nation, and the framers of the Federal Government. We have already seen, however, that there are no such people, in a strict political sense, and that no such people had any agency in the formation of our Constitution, but that it was formed by the States, emphatically as such.’ 130   For Carl Schmitt, a living federal organism is one in which multiple polities are coordinated and in which multiple perspectives on the questions of sovereignty and supremacy are – relatively – valid. Wherever the sovereignty question is definitively answered in favour of one government, the federal substance of the body politic – the co-existence of two levels of constitutional politics – disappears. Depending on the answer, the federation either disintegrates into an international organization or it consolidates into a decentralized (unitary) State. Federalism, by contrast, represents an existential equilibrium – a perpetuated constitutional dualism. On C Schmitt’s theory of federalism see Schütze, n 64, 38–40. 131   Ibid 30–36 and 58 et seq.

Federalism as Constitutional Pluralism  211 of sovereignty, constitutional pluralism must be seen as a ‘novelty’ or ‘aberration’. The absence of an ‘Archimedean point’ from which all legal authority can be explain is thus – wrongly – hailed as a sui generis quality of the European Union.132 How introverted and unhistorical! Why not see the normative ambivalence surrounding the supremacy principle in the European Union as part and parcel of Europe’s federal nature?133 The theory of constitutional pluralism speaks federal prose,134 without – as Molière’s Monsieur Jourdain – being aware of it. Let us hope that Europe’s constitutional theorists will soon open their eyes (and ears) to the treasures of comparative constitutional law.

132   Cf N Walker, ‘The Idea of Constitutional Pluralism’ (2002) 65 Modern Law Review 317, 338. This is how Walker, a leading figure of the ‘constitutional pluralists’, describes the origin of this ‘new’ constitutional philosophy: ‘It is no coincidence that this literature has emerged out of the study of the constitutional dimension of EU law, for it is EU law which poses the most pressing paradigm-challenging test to what we might call constitutional monism. Constitutional monism merely grants a label to the defining assumption of constitutionalism in the Westphalian age . . . namely the idea that the sole centres or units of constitutional authorities are states. Constitutional pluralism, by contrast, recognizes that the European legal order inaugurated by the Treaty of Rome has developed beyond the traditional confines of inter-national law and now makes its own independent constitutional claims exist alongside the continuing claims of states’ (ibid 337). This – ‘Eurocentric’ – view strikingly ignores the American experience, in which the Union and the States were seen to have ‘constitutional’ claims and in which the ‘Union’ was – traditionally – not (!) conceived in statist terms (cf E Zoeller, ‘Aspects internationaux du droit constitutionnel. Contribution à la théorie de la féderation d’états’ (2002) 194 Recueil des Cours de l’Académie de la Haye 43). 133   On the two perspectives on the supremacy principle in the European Union, see the discussion in P Craig and G de Búrca, EU Law: Text, Cases, and Materials 4th edn (Oxford, Oxford University Press, 2008) Ch 10. 134   The family resemblance between federalism and constitutional pluralism was identified by a former President of the European Commission. When Europe ‘turn[ed] to the principles of federalism in a bid to find workable solutions, it is precisely because they provide all the necessary guarantees on pluralism and the efficiency of the emergent institutional machinery’ (address by Jacques Delores at the College of Europe (17 October), Bull. EC 10-1989, 110, 114–15, emphasis added).

9 Out with the New, in with the Old – Neo-Roman Constitutional Thought and the Enigma of Constitutional Pluralism in the EU OLA ZETTERQUIST

I. Introduction Constitutional pluralism within the EU primarily concerns the question of whether the EU can be considered to be a separate and independent legal order in co-existence with the legal orders of the Member States. The notion of constitutional pluralism is controversial mainly because it challenges both the notion of sovereignty (and the sovereign state) and the concept of the unity of the legal order. Taken seriously, constitutional pluralism is incompatible with the traditional and immensely influential models of formal norm-hierarchy (formal legit­ imacy), developed by Hans Kelsen, and sovereignty, developed originally by Thomas Hobbes. It will be argued that these problems arise as a consequence of a) a formal account of constitutionality and b) an institutional understanding of the notion of freedom as non-interference. I will claim that another approach based on substantive constitutionality (res publica) and freedom understood as nondomination, may be more helpful in (normatively) explaining the idea of constitutional pluralism in the EU. In doing so, this Chapter will analyse the enigma of constitutional pluralism within the EU seen through the lens of a neo-Roman republican constitutional theory drawing inspiration from both classical authors, such as Cicero and Machiavelli, and modern ones such as Skinner, Pettit, and Dworkin. The focus of this Chapter will be the issues of sovereignty and the character of the legal order as seen through a republican prism. I will, however, not address the issue of whether the republican theory is fundamentally at odds with liberalism in other aspects not related to sovereignty.

214  Ola Zetterquist

II.  The Enigma of Constitutional Pluralism – The Original Sin of the Court of Justice of the European Communities It is no exaggeration to say that the European Court of Justice (ECJ) started it all. In the seminal case van Gend en Loos,1 the court held that the (then) EEC was a ‘new’ legal order, separate from both Member State law and from the other bodies of international law. When the dust had settled, the question was inevitably raised which relationship that obtained between these distinct bodies of law. In particular the court held that the Community was an entity ‘. . . for the benefit of which the States have limited their sovereign rights’. In the same vein the court stated that the Member States had endowed the Community institutions with ‘sovereign rights’, thereby suggesting that the concept of state-sovereignty was now in a process of change. When the ECJ subsequently, in the case Costa v ENEL,2 added that EC law would prevail over national law – however framed – the constitutional drama was intensified and Member State sovereignty further undermined. Since the ECJ, tactfully, did not claim that the EC Treaties had replaced the national Constitutions or that these otherwise derived their validity from the Treaties, the claims laid down in these early constitutionalising cases was essentially one of constitutional pluralism: the EU Constitution co-exists with national Constitutions and with international law in general in a non-hierarchical fashion. If we are to believe the ECJ there are now 27 Member State Constitutions that co-exist with the EU Constitution. This Chapter proposes to examine the legalphilosophical obstacles to the idea of constitutional pluralism and a possible neo-Roman alternative for escaping the sovereignty trap.

III.  Constitutional Pluralism in the Shadow of Leviathan – the Stumbling-block of Sovereignty Not everybody has accepted the court’s argument about the EU as a separate and independent legal order. Scholars like TC Hartley, to name one of many, have accused the court of suffering from a ‘reality deficit’ when it has employed the language of constitutional pluralism.3 The basis of Hartley’s argument is that the Member States are still sovereign – the Lords of the Treaties with the famous wording of the German Constitutional Court – and can at any time control the   Case 26/62 Van Gend en Loos ECR [1963] 1.   Case 6/64 Costa v ENEL ECR [1964] 585. 3   TC Hartley, Constitutional Problems of the European Union (Oxford, Hart Publishing, 1999) 181. 1 2

Out with the New, in with the Old  215 application and, ultimately, validity of EU law on their territories.4 What may remain is an issue under public international law between the Member State that misapplies EU law and the other Member States Parties to the Treaties but this should not be confused with a constitutional conflict. As the House of Lords phrased that very issue in the Salomon case: If the terms of the legislation are clear and unambiguous, they must be given effect to, whether or not they carry out Her Majesty’s Treaty obligations, for the sovereign power of the Queen in Parliament extends to breaking treaties.5

The central point is that the national Constitution, conferring sovereign power on the state organs, trumps a treaty – however framed – if push comes to shove. A further point of interest in the reasoning of the Lords is that it seems to be irrelevant what material content the ‘legislation’ or the ‘Treaty obligations’ have. It is rather the locus of the decision-maker that matters, or, to be more precise, that the decision-maker is located at the national level. National sovereignty, it seems, is still alive and kicking despite the assertions to the contrary. In the words of the (then) British Prime Minister John Major in the House of Commons on the British accession to the European Union Treaty: ‘The sovereignty of this House is not a matter that is up for grabs – that is perfectly clear.’6 The Prime Minister presumably meant that the fundamental British constitutional principle of parliamentary sovereignty was in no way affected by the EU Treaty, nor that any such effect had followed from acceding to the European Communities in 1973. Logically such an argument also entails that no subsequent treaty is capable of derogating from parliamentary sovereignty lest the Parliament formally abdicates its sovereignty and conveys it on another political entity. Since the counterattack on the notion of constitutional pluralism is premised on the argument of sovereignty, it is essential to analyse the more precise meaning and origin of the latter concept. The presumed etymological background to sovereignty is the Latin superaneus, meaning top or highest (cf the Italian soprano). Sovereignty as a concept was first used by the French philosopher Jean Bodin (1530–1596), who regarded it as the unique attribute of the state, embodying the ultimate power to enact, apply and maintain the law.7 The more distinctly modern understanding of the concept, however, came with the English philosopher Thomas Hobbes (1588–1679), particularly formulated in his masterpieces The Citizen (1642) and Leviathan (1651) where the defining characteristics of sovereignty was an artificial person (= the state) with supreme power to ‘. . . make and abrogate laws, to determine war and peace, to know and judge of all controversies . . . to elect all magistrates, ministers and counsellors’.8   Ibid, Chs 7–9.   Salomon v Commissioners of Customs and Excise [1967] 2 QB 116 at 143. 6   From the debate on the European Union (Maastricht) Treaty held in the House of Commons on 12 May 1992, www.hansard.millbanksystems.com/commons/1992/may/12/engagements. 7   J Bodin, On Sovereignty (Cambridge, Cambridge University Press, 1992) 56; see also V Bogdanor (ed), The Blackwell Encyclopaedia of Political Institutions (Oxford, Basil Blackwell, 1987) 591. 8   T Hobbes, ‘The Citizen’ in Man and Citizen (Indianapolis, Hackett Publishing, 1991) 188. 4 5

216  Ola Zetterquist Sovereignty is often regarded as being the same as the independent control of a certain territory by one or more organs (a certain person, certain persons or certain assemblies).9 These criteria are very similar to those used for defining statehood.10 State and sovereignty are therefore concepts that belong very closely together and is indeed often expressed as ‘state sovereignty’. There are many facets to the concept of sovereignty and it is now used in many different senses.11 A first distinction can be made between political and legal sovereignty. Another division is between internal and external sovereignty. In the political sense sovereignty is territorially defined and constituted by control over a society of individuals, who customarily obey this sovereign, and that is not limited by any higher or equal power.12 The inhabitants in a given state usually obey their own state and not other states. In the legal sense, sovereignty may be defined on the basis of the legal norms that those who apply the legal system follow and recognise as valid and superior to other legal norms, that is the institution (or institutions) that issue these legal norms is regarded as sovereign.13 Within the state this is a more common way of identifying the sovereign than the question of purely physical control. Legal sovereignty is closely connected to the notion of formal legitimacy (validity) of the legal norms in the legal order. According to the Austrian legal philosopher Hans Kelsen (1881–1973), the legal order should be understood as a hierarchical order with different strata of legal norms and where the Constitution represents the highest level of positive law.14 A Constitution is characterised by the fact that there is no person or institution competent to proclaim the Constitution void without actually overthrowing it.15 In this way the notion of sovereignty is closely related to the Constitution since no other norm that belongs to the same legal order as the Constitution can be invoked to invalidate the Constitution. It is, as already mentioned, common to distinguish between external and internal sovereignty. This sub-division corresponds largely with the division of sovereignty in its political and legal senses.16 External sovereignty implies that within a state’s legal system foreign laws can apply only on the basis of that legal system’s own laws. The sovereign state’s legal system is, in other words, totally independent of other legal systems. Internal sovereignty on the other hand corresponds to   See Bogdanor, n 7 at 583.   Laid down in Art 1 of the Montevideo Convention of 1933; see further MD Evans, International Law (Oxford, Oxford University Press, 2006) 231–42. 11   A Dicey, Introduction to the Study of the Law of the Constitution (Indianapolis, Liberty Fund, 1982) 27. 12   As defined by, among others, J Austin, The Province of Jurisprudence Determined (Cambridge, Cambridge University Press, 1995) 166 and J Bentham, Of Laws in General (London, Athlone Press, 1970) 18. 13   Dicey, n 11 at 28; see also HWR Wade, ‘The Legal Basis of Sovereignty’ (1955) 13 Cambridge Law Journal 172 at 186. 14   H Kelsen, Introduction to the Problems of Legal Theory (Oxford, Oxford University Press, 1992) 58. 15   Dicey, n 11 at 39. The only possible exception to this rule is if the Constitution is scrapped in accordance with its own provisions. 16   See N MacCormick, Questioning Sovereignty (Oxford, Oxford University Press, 1999) 126 compared with A Dicey, n 11 at 27. 9

10

Out with the New, in with the Old  217 legal sovereignty as described above and focuses on the issue of which institution(s) are supreme within the legal system. The main distinction is that whereas external sovereignty is always absolute – a state is either sovereign or not – internal sovereignty can be located in different institutions (which derive their authority from the Constitution of the sovereign state). As can be seen, the concept of sovereignty is primarily concerned with locating a final authority within the political and legal order and it is thus a notion that is inherently hostile to competing supreme authorities. It is the unitary character of sovereignty that makes the doctrine of constitutional pluralism present a serious, virtually insurmountable, problem and which led Lord Denning to proclaim (presumably in despair): Our sovereignty has been taken away by the European Court of Justice. It has made decisions impinging on our statute law and says that we are to obey its decisions instead of our own statute law . . . It has put on the Treaty an interpretation according to their own views of policy . . . The European Court has held that all European Directives are binding within each of the European countries; and must be enforced by national courts; even though they are contrary to our national law . . . No longer is European law an incoming tide flowing up the estuaries of England. It is now like a tidal wave bringing down our sea walls and flowing inland over our fields and houses – to the dismay of all.17

A theory of constitutional pluralism is (at best) a mirage if the sovereignty argument holds true. The relation between sovereignty and law is also the fundamental issue that must be resolved if a theory of constitutional pluralism is to hold.

IV.  The Anatomy and Pedigree of Leviathan The various definitions of sovereignty, and their relation with the concept of formal legitimacy, have been given above. In this section the aim is to take a closer look at the pedigree and anatomy of the concept as it emerges from the theories of the originators Hobbes and Kelsen. The concepts of sovereignty and formal validity are core concepts of constitutional law today and even though Hobbes and Kelsen are approximately three centuries apart, they share some essential points of departure which in turn form the basis of constitutional monism. These common points are the separation of moral qualities from the question of law and freedom understood as non-interference. The starting point of Hobbes’ theory of law and state is the so-called state of nature, that is a state of affairs where there is no common superior authority present for those who happen to find themselves there. The state of nature is characterised by a total liberty for every agent to act in accordance with his/her 17   Lord Denning, in the Introduction to G Smith, The European Court of Justice. Judges or Policymakers (London, Bruges Group, 1990).

218  Ola Zetterquist capacity and expediency, meaning that there are no externally imposed restrictions on the scope of actions permitted in the state of nature. Liberty itself was ‘nothing else but an absence of the lets and hindrances of motion’.18 In the state of nature everyone has the right to everything which, in practice, will lead to the opposite – no one has the right to anything except by his or her own force.19 The way out of the destructive state of nature is, through a social contract, to create an artificial person – the sovereign of the commonwealth – with authorised power to restrict the original freedom of the state of nature of the individuals. The original liberty is instead transferred to the sovereign who now possesses the total liberty of the state of nature. The sovereign is defined in the following terms: [the Commonwealth] is One Person, of whose Acts a great Multitude, by mutuall Covenants one with another, have made themselves every one the Author, to the end he may use the strength and means of them all, as he shall think expedient, for their Peace and Common Defence. And he that carryeth this Person, is called SOVERAIGNE, and said to have Soveraigne Power; and every one besides, his SUBJECT.20

According to Hobbes’ understanding of liberty it follows that laws, which are the result of the actions of the sovereign, are restrictions (artificial chains was Hobbes’ own formulation) on the original liberty of the state of nature.21 The measure of liberty was the silence of the law and absence of restraint. Consequently, there was no necessary connection between a free (that is, sovereign) state and free citizens.22 It is also clear from the quoted paragraph that Hobbes did not attach any particular value to the moral qualities of the actions of the sovereign. He clearly emphasises that it is for the sovereign to decide on the expediency of the acts of the commonwealth. The benefit of having a sovereign is not that he is the wisest or the most congenial person but that he is authoritative and puts end to the plurality of wills characteristic to the state of nature. The Hobbesian (and Kelsenian) attack on the moral understanding of law were both premised on the belief that the introduction of moral arguments in the legal discourse fragmented the law, created divisive political tensions (possibly leading to civil war) and placed lawyers (or even theologians) above the legislator. Accepting that moral qualities influence the validity and meaning of positive law is in effect to introduce an additional set of constitutional norms in the shape of natural law of one flavour or the other.23 Hobbes accordingly argued that any 18   T Hobbes, n 8 at 216. This definition of liberty allows Hobbes to state that a prisoner in a large cell enjoys more liberty than one confined in a smaller cell. 19   Ibid 116. 20   T Hobbes, Leviathan (Cambridge, Cambridge University Press, 1996) 121. 21   Ibid 145. 22   Ibid 149. A contemporary example might be the Soviet Union, which certainly was a free state without this implying that Soviet citizens were free in any meaningful way. 23   All brands of natural law, with the exception of Hobbes’ own peculiar version, where he turns the canons of the natural lawyers on themselves, are pluralist in nature even though conflicts between the different orders are ultimately to be resolved in accordance with the natural law. This is, however, a material question, not a formal one.

Out with the New, in with the Old  219 attempt to subject the sovereign’s decisions to independent moral assessment, thereby qualifying the sovereign as good or evil, was in effect an attempt at the very integrity and existence of the state. The argument was illustrated with a colourful analogy to the original sin of the bible (where the apple of the tree of knowledge represented precisely the knowledge of good and evil): Who hath told thee that he was a tyrant? Hast thou eaten of the tree whereof I commanded thee that shouldst not eat? For why dost thou call him a tyrant, whom God hath made a King, except that thou, being a private person, usurpest to thyself the knowledge of good and evil?24

It is also specifically the absence of moral assessments that make the Kelsenian legal theory, with Kelsen’s own words, ‘pure’.25 Their common aim was a monistic system of rules (founded in one Constitution only) that is internally consistent and hierarchically structured with mechanisms for resolving conflicts of norms. The question of validity is a purely formal exercise of ascertaining whether the norm (general or individual) in question was adopted in accordance with the higher echelons of the legal system.26 Hobbes expressly refuses any independent assessment of whether the sovereign acts in accordance with the dictates of morality and goes as far as to say that the law is in itself the measure of good and evil.27 Even though neither of the two denied that moral arguments could be made even about legal matters, they were both in agreement that such arguments were distinct and separate from legal arguments. Both Hobbes and Kelsen strived for a unitary model of the law free from external distortions, whether political, moral or legal, where conflicts of norms could always be resolved without recourse to elements outside the system. The combination of formal legitimacy with sovereignty functions is a formidable stumbling block to constitutional pluralism as envisaged by the ECJ. It may be true that coherence can be achieved by adopting a certain perspective on the issue (and omitting the conflicting counter-perspective). From the EU perspective the vexed question can be solved by taking the position that EU law prevails over conflicting national law (as a matter of formal fact) whereas a proponent of the Member State view can argue that the Treaties derive their legitimacy from the constitutions of the Member States and that the Treaties are (equally as a matter of formal fact) subordinated to national law. Constitutional pluralism, however, is incompatible with both the Kelsenian and the Hobbesian constitutional models, which are unitary to the core. Nor can the EU, on a Hobbesian reading, have replaced the former constitutions since the EU is not a state. Consequently another model must be sought if constitutional pluralism is to be sustained. It is my claim that such a theory can be found, and be fruitfully applied, if we turn our   Hobbes, n 8 at 246.   H Kelsen, Reine Rechtslehre (Vienna, Franz Deuticke Verlag, 1960) 320. 26   Kelsen, n 14 at 64. 27   ‘[T]here are no authentical doctrines concerning right and wrong, good and evil besides the constituted laws in each realm and government’. See Hobbes, n 9 at 98. 24 25

220  Ola Zetterquist gaze towards the constitutional theory prior to the emergence of the sovereign state. Drawing inspiration from Quentin Skinner’s influential work Liberty before Liberalism28 the idea is here to conceptualise the idea of the Constitution before the sovereign state. It is on this point that the neo-Roman republican theory, which has a completely different point of departure than the Hobbesian, presents itself as an alternative to the models of Kelsen and Hobbes. The road ahead goes through the past.

V.  Roman Freedom and Constitution The republican tradition takes its point of departure in the constitutional theory of the Roman republic (509–27 BC). The essential distinction in Roman law was the one between a free man – a citizen of the republic – and a slave. This was formulated as a question of status (from which the word state is actually derived) – one was either a servus or liber homo.29 On the Roman scheme freedom was not understood as absence of external impediments (as Hobbes would have it) but rather as the absence of domination. A slave, even though he or she could enjoy a considerable amount of actual liberty, was at any point in time liable to have his/her situation changed at the arbitrary whim of the owner.30 It was this condition that rendered the slave unfree rather than physical coercion.31 The reason for this is that a slave, who is aware of this predicament, will shape and adapt his behaviour in order to avoid the risk of intervention by the master, even though such intervention may not even have been actually threatened.32 Likewise, not every interference with a citizen’s legal situation was considered to be contrary to freedom but only such that are arbitrary, that is, based on factional grounds and, therefore, contrary to the res publica. Translated to modern republican theory this is formulated as follows: in order for interference to be   Q Skinner, Liberty Before Liberalism (Cambridge, Cambridge University Press, 1998).   See Q Skinner, ‘Freedom as the Absence of Arbitrary Power’ in C Laborde and J Maynor (eds), Republicanism and Political Theory (Oxford, Blackwell Publishing, 2008) 83–101. 30   Expressed in the terminology of Hohfeld a slave could enjoy privilege (liberty) but never immunity. See further WN Hohfeld, ‘Some Fundamental Conceptions as Applied in Judicial Reasoning’ (1913–1914) 23 Yale Law Journal 16–59. 31   This view was echoed by many of the Whig philosophers of the 17th century: ‘For as liberty solely consists in an independency upon the will of another, and by the name of slave we understand a man, who can neither dispose of his person nor goods, but enjoys all at the will of his master.’ A Sidney, Discourses Concerning Government (Indianapolis, Liberty Fund, 1996) 17. ‘Freedom is not, as we are told, A liberty for every Man to do what he lists: (For who could be free, when every other Man’s Humour might domineer over him?) But a Liberty to dispose, and order as he lists, his Person, Actions, Possessions, and his whole Property, within the Allowance of those Laws under which he is; and therein not to be subject to the arbitrary Will of another, but freely follow his own.’ J Locke, Two Treatises of Government (Cambridge, Cambridge University Press, 1988) 306, § 57. 32   Skinner, n 28 at 90. 28 29

Out with the New, in with the Old  221 consistent with freedom it must be based on principled legislation that takes the affected individuals interests into account and that the interfering agent must be subject to some sort of constitutionally entrenched constraint and review.33 An important consequence of understanding freedom as non-domination is that no-one, not even the state, can hold the total freedom of the Hobbesian state of nature since this would entail domination of those subject to it (which was precisely what Hobbes held to be necessary in order to end the chaos of the state of nature). The freedom of a state was not in the first instance defined in its relation with other states but inwards. There was, in other words, a necessary connection between a free state and free individuals within the state – no state could claim to be free if its citizens were not.34 It could be formulated as that the republicans did not want to rule (à la Hobbes) in a certain way; they simply did not want to be ruled or, at any rate, not be ruled by a ruler with arbitrary power (usually understood as a monarch).35 To measure the freedom of the citizens according to the Hobbesian yardstick of absence of legal restraint seemed ludicrous to the neo-republican philosophers of the enlightenment. As John Harrington formulated the position: to say that a Lucchese [Lucca was an Italian republic, author’s remark] hath no more liberty or immunity from the laws of Lucca, than a Turk hath from those of Constantinople, and to say that a Lucchese hath no more liberty or immunity by the laws of Lucca than a Turk hath by those of Constantinople, are pretty different speeches. The first may be said of all governments alike, the second scarce of any two.36

As can be seen in this quote the point was to be free by the laws rather than from them. This point lies at the centre of the republican concept of freedom and has significant importance for the understanding of the nature of law. Law is primarily an instrument for securing non-domination and must therefore conform to the common good (res publica). The emphasis on the common good and laws being instrumental for securing non-domination means that there is a rather pronounced anti-majoritarian strand in the republican tradition.37 Majority voting certainly has a place in it but is seen in the first instance as the most simple and effective mechanism for preventing arbitrary interference rather than as a mechanism for achieving political freedom (as Rousseau had it38). The republic therefore places a strong emphasis 33   See further P Pettit, Republicanism – A Theory of Freedom and Government (Oxford, Oxford University Press, 1997) 65. 34   Machiavelli formulated the connection in the following terms: ‘[T]he common benefit derived from a free government is recognized by no one while it is possessed, that is, to be able freely to enjoy one’s possessions without worry.’ Discourses on Livy (Oxford, Oxford University Press, 1997) 63. 35   C Laborde and J Maynor, ‘The Republican Contribution’ in C Laborde and J Maynor (eds), Republicanism and Political Theory (Oxford, Blackwell Publishing, 2008) 3. 36   J Harrington, The Commonwealth of Oceana and a System of Politics (Cambridge, Cambridge University Press, 1992) 20. 37   Pettit, n 33 at 30. 38  JJ Rousseau, The Social Contract and Other Later Political Writings (Cambridge, Cambridge University Press, 1997) 54.

222  Ola Zetterquist on a stable and detailed framework for the legislative process that should, ideally, not be subject to changes. The constitutional system of the Roman Republic was built on the system of separation of powers in order to avoid existence of arbitrary power, even the power held by the people themselves, and such a system remains an indispensable component of most of the contemporary republican theories.39 Rome was often described as a mixed Constitution40 combining elements of monarchical, aristocratic and democratic power, and this fact was held out as one of the most import­ ant in explaining the rise of the Roman republic as a superpower.41 The idea of constitutional pluralism was at the heart of the republic and Machiavelli accordingly argued that a healthy republic was characterised by friction and conflict between these different powers. In this vein he argued that Rome was a ‘perfect republic’ because of the discord between the plebeians in the popular assemblies and the patricians of the senate.42 Far from trying to prevent conflict in the republic he welcomed it but at the same time cautioned against letting the republic be governed by extraordinary means that might seem tempting in a given moment and even produce short term benefits. Breaking the law, even for good reasons, would in the long run undermine the republic and open up the possibility of breaking it also for bad reasons. The laws of the republic must therefore ‘contain provisions for everything and establish a remedy for every circumstance and set up means for dealing with it’.43

VI.  Res Publica and the Legal Order The notion of res publica has a central role in republican theory and constitutes the material legitimacy (the reason of being) of the legal and political order, for which the members of the society in question submit to the authority of the norms and institutions set up by the legal order. Res publica is also the criterion for determining whether legislation, and interference in the personal sphere, accord with the ideal of non-domination.44 39  Cicero, The Republic and The Laws (Oxford, Oxford University Press, 1998) 71 and Pettit, n 33 at 177. Some modern republicans will dispute this point: Richard Bellamy and Adam Tomkins for instance believe that a Constitution designed on checks and balances with a strong judicial power (legal constitutionalism) will fail to secure freedom as non-domination. See further R Bellamy, Political Constitutionalism – A Republican Defence of the Constitutionality of Democracy (Cambridge, Cambridge University Press, 2007) and A Tomkins, Our Republican Constitution (Oxford, Hart Publishing, 2005). 40   Reflected in the republic’s official signature Senatus Populusque Romanus (SPQR) – the Senate and People of Rome. 41  Polybius, The Rise of the Roman Empire (London, Penguin Books, 1979) 317; N Machiavelli, n 34 at 26. 42   Machiavelli, n 34 at 26. 43   Ibid 95. 44   Pettit, n 33 at 182.

Out with the New, in with the Old  223 The origin of the notions of res publica and republican constitutionalism lie mainly with the Roman statesman and philosopher Marcus Tullius Cicero (106– 43 BC) who held that legal agreement and community of interest formed the core of the concept.45 The legal and political orders are in this sense the property (res) of the people (publica). Cicero also formulated another classic of modern Western legal theory in his famous account of law to be viewed as ‘right reason [recta ratio] in harmony with nature’46 and that law must be universalistic (and not limited to a certain community) in its character.47 This was underlined by the fact that international law (ius gentium) was perceived as a law between individuals, rather than states, and thus a law that transcended national boundaries and bound Roman and non-Roman alike. Put differently, there was a difference between (morally) ‘good’ and ‘bad’ law in the republican Constitution48 and sharing law also meant sharing justice.49 More specifically this link between public power, law and public morality has ever since the days of the Roman republic been known as the res publica or the common good of the society from which the constitutional notion of the commonwealth derives. The notion meant to the Romans not as much a ‘republic’ as ‘the activities of the Roman people’.50 This dynamic concept of the res publica is underlined by the fact that the Roman republic did not possess a Constitution in the modern sense of a clear set of written rules that defined state institutions and their powers.51 The lasting appeal of the notion of res publica is that it appeals to a universalistic idea of a common good, morally superior to the self-interest of the single individual (or group of individuals). Rousseau famously stated that there was a significant difference between the ‘will of all’ and the ‘general will’, the proper res publica: ‘The [general will] looks only to the common interest, the [will of all] looks to private interest, and is nothing but a sum of particular wills.’52 Even though Rousseau in the end concludes that there is no other viable way of ascertaining the general will than by majority rule, his idea of a distinction between the general will and a cumulated will of all has been quite influential for a modern interpretation of the res publica. Much like the idea of a social contract, res publica aspires to the assent of rationally thinking individuals by appealing to principles of moral justice that are shared by most (if not all) individuals. Law in accordance with res publica entails a view of the law as a species of moral public dialogue, a dialogue conducted both through the political assemblies and 45   Cicero, n 39 at 19. In this passage Cicero can be seen as an early precursor of the core ideas of constitutionalism, the modern heir of the republican ideal; see Pettit, n 33 at 177. 46   Cicero, n 39 at 68. 47   Ibid 111. 48   Ibid 112. 49   Ibid 105. 50   P Jones and K Sidwell (eds), The World of Rome (Cambridge, Cambridge University Press, 1997) 84. 51   Still today there are Constitutions that develop in the Roman way. The British common law Constitution is the most important contemporary example. 52   Rousseau, n 38 at 60.

224  Ola Zetterquist the courts of law. De Tocqueville advanced such an argument in his famous account of the American republic: However weak a man may be, he can always compel a judge to listen to his complaint and give him an answer. That is inherent in the very nature of judicial power . . . The power of the courts has been at all times the securest guarantee which can be provided for individual independence, but this is particularly true in ages of democracy.53

A citizen gets the opportunity to speak in the process of legislation but also the chance to speak back through the courts and both venues are equally important for establishing non-domination.54 In this sense a republican substantial understanding of the law resonates with more contemporary theories that stress the importance of integrity55 and coherence 56 in legal reasoning, indicating that some legal solutions are better than others with reference to the underlying moral fabric (res publica) of the legal order. The focus is thus moved from the formal issue of the source of the rule in question to the more substantial question of whether the rule is in accordance with the values that the legal order enshrines. The dialogue principle of the republican Constitution is also echoed in modern theories of judicial review that stress the interaction between legislative and judicial bodies. The latter are primarily charged with the task of ensuring that the choices of the former are applied in a coherent and non-discriminating fashion in the interest of the whole.57

VII.  All Roads Lead to Rome – A Republican Rhyme in the Constitutionalising Cases of the ECJ If there cannot be a sovereign EU, can there be a republican one? In this section the purpose is to see whether the argumentation of the ECJ in the constitutionalising cases have contributed to freedom as non-domination at the European level. If so, then a case could be made for accepting constitutional pluralism based on republican values. The ECJ has consistently, since the 1960s, claimed that the EU has a constitutional legal order that is independent from the legal orders of the Member States   A de Tocqueville, Democracy in America (London, Fontana Press, 1969) 698.   As a point of curiosity, in this regard, it could be observed that the Forum in Rome was the central scene both for political activities and for the public trials. 55   See eg R Dworkin, Law’s Empire (Cambridge, Harvard University Press, 1986) 225, particularly at 243 where integrity ‘asks judges to assume . . . that the law is structured by a coherent set of principles about justice and fairness and procedural due process, and it asks them to enforce these . . . so that each person’s situation is fair and just according to the same standards’. 56   A Peczenik, ‘The Passion for Reason’ in L Wintgens (ed), The Law in Philosophical Perspectives (Dordrecht, Kluwer Academic Publishers, 1999) 173–224. 57   For an example of such a theory see J Hart Ely, Democracy and Distrust – A Theory of Judicial Review (Cambridge, Harvard University Press, 1980). 53 54

Out with the New, in with the Old  225 and that embraces individuals as well as states. By the same token it is from the status as an independent legal order that the problems associated with the ‘democratic deficit’ of the EU and the question of the compatibility with the concept of sovereignty seek their provenance. The starting point of the constitutional argument is found in the by now class­ ical statement in the preamble of the EC Treaty that the EC strives to create ‘an ever closer Union among the peoples of Europe’. This statement has even been said to form the ‘genetic code’ of the EC.58 Further guidance could be found in Article 6 TEU, which states that the EU is founded on certain fundamental values (liberty, democracy, respect for human rights and fundamental freedoms and the rule of law) and that these principles are common to the Member States. These constitute the moral foundations of the EU or, put in other terms, its res publica – its claim for moral authority as a commonwealth.59 The ECJ set out its famous arguments for the ‘new legal order’ of the EC in the decisive van Gend en Loos case.60 One could summarise the argument of the court as follows: the Treaty provisions could not be understood without an inquiry into its ‘spirit, the general scheme and the wording of [its] provisions’. The objective of the Treaty implied that the Treaty did not limit itself to creating mutual obligations between the states but also created rights and obligations for individuals, which became part of their legal heritage. The Treaties were adopted in legal form and created institutions with state powers. The Treaties furthermore made reference to individuals in the preamble and the nationals of the Member States were called upon to collaborate in the functioning of the then Community through the European Parliament and the Economic and Social Committee.61 As a consequence, the court held that, provided some other conditions are met, rights laid down in the Treaty can have direct effect in the Member States. The important constitutional point is that the effect of EU law follows directly from the Treaty and not from the Member State’s internal provisions. Direct effect is in this sense incompatible with the traditional view of the relationship between domestic law and international law since the traditional position is that the state’s constitutional provisions ultimately determine the internal effects of international law.62 It is thus the introduction of direct effect that opens up the perspective of constitutional pluralism since the internal effect of EU law is no longer contingent on the national Constitution(s) while, at the same time, the Member State Constitution remains valid.

58   F Mancini and D Keeling, ‘Democracy and the European Court of Justice’ (1994) 57 Modern Law Review 175, 186. 59   MacCormick, n 16 at 137. 60   van Gend en Loos, ECR, n 1. 61   Ibid, para 12. 62   See, inter alia, E Denza, ‘The Relationship between International and National Law’ in MD Evans (ed), International Law (Oxford, Oxford University Press, 2006) 428. This very argument was also put forward in van Gend en Loos, n 1 paras 5–8, by the Dutch, German and Belgian governments and was supported by the Advocate General but was not accepted by the court.

226  Ola Zetterquist In the likewise seminal case Costa v ENEL,63 the ECJ held that the terms and spirit of the Treaty were accepted by the states on a basis of reciprocity. The execu­ tive force of the Treaties could not vary between the Member States without giving rise to discrimination between the European individuals. The Treaty, as an independent source of law, could not be overridden without being deprived of its character as Community law and without the legal basis of the Community itself being called into question.64 The conclusion from this reasoning was that EC law had supremacy over Member State law within the jurisdiction of the ECJ, even in the case where the Member State law concerned was of constitutional character, something which runs counter to the traditional view on the relationship between international and domestic law. The principles of direct effect and supremacy together meant that EC law was a common law within the Member States and not only between them (as would have been the case with an ordinary Treaty) and, accordingly, that individuals in the Community now shared this law. A further paramount step in the development of the constitutional character of the EU was the introduction of protection of fundamental rights of the individual as a part of the general principles of EU law.65 The court found that fundamental rights were protected as rights under EU law, rather than as rights of a certain Member State legal order. Such protection can be seen as a necessary counterbalance to the principle of supremacy of EU law in relation to national law. If violations, by EU institutions, of fundamental rights of the individual could no longer be checked by national constitutional provisions an alternative protection under EU law would have to be found for the EU to enjoy constitutional legitimacy.66 To paraphrase Voltaire’s famous remark on the Deity, one could say that if constitutional rights protection did not exist in EU law before, one would have to invent it. It also falls to the European level to strike the balance between the interests concerned. The point of republican interest is that the general interest at issue here is one that embraces the entire EU and not only a part of it. It is thus possible that the general interest in the EU as a whole might call for a restriction that would not have been permitted in one part (typically a Member State) or vice versa. By taking the entire EU into consideration the balance between competing claims will ideally be based on more universalistic principles than if they were to be decided within a single Member State, thereby connecting to Cicero’s ideal that   Costa v ENEL, n 2.   Ibid 593. 65   The ECJ embarked on this path in a string of cases in the 1960s and 1970s starting with Case 29/69 Stauder ECR [1969] 419. The idea was further clarified in the seminal case Internationale Handelsgesellschaft, where the Court held that ‘respect for fundamental rights form an integral part of the general principles of law protected by the Court of Justice. The protection of such rights, whilst inspired by the constitutional traditions common to the Member States, must be ensured within the framework of the structure and objectives of the Community.’ Case Internationale Handelsgesellschaft 11/70 ECR [1970] 1125, para 4. A fuller account of the issue is given in T Tridimas, The General Principles of EC Law (Oxford, Oxford University Press, 1999) 202–43. 66   F Mancini, ‘The Making of a Constitution for Europe’ (1989) 26 Common Market Law Review 595, 608. 63 64

Out with the New, in with the Old  227 the law ought to be universal for all mankind rather than divided by different states. To put it briefly, it is clear from this cursory glance that the ECJ has moved from a pacta sunt servanda (state) to a res publica (citizen) approach to EU law. It has given considerable importance to the position of the individual in inter­ national law and the rule of law in its reasoning on why EU law constitutes a new constitutional legal order. The Treaties are concluded in legal form and create rights for individuals (although the Treaties themselves were concluded by states). This is said to be because of the ‘spirit and general scheme’ and the objective of the Treaty, that is because of its moral purpose or res publica. The foundation of the Union as a constitutional order thus echoes Cicero’s definition of the republic as a community based on legal agreement and community of interest. Admittedly, the contribution of the ECJ is short of a fully fledged republic but it is at least an emerging constitutional foundation based on the legal protection of individuals under a common legal order. No state can on its own grant its citizens the right to freely move and reside within other Member States. Even an agreement between two (or more) states on these matters would leave the rights of the individual as, at best, accessory to those of the state in question. The EU accordingly functions as an important source of genuine rights for individuals within the European territory, rights that could not be achieved by the Member States on their own. What is more, these rights do not only function as a constitutional parameter against one’s own state, they are also operative against other states as a matter of right and not as concession, that is, contestable according to the dialogue principle. The principles of direct effect and supremacy, which are the necessary conditions for asserting EU rights against conflicting national legislation, are accordingly instrumental in securing nondomination.67 These principles have been described by the late judge Federico Mancini as the most significant democratic improvements of the EU in the sense that the ECJ took the law out of the hands of bureaucrats and politicians and give it to the people,68 but perhaps the principles are more appropriately suited for a republican understanding of the constitutionalisation of the Union. Even though the interest in fundamental rights was to some extent forced on the ECJ by national courts (notably the German Constitutional Court), the challenge has been readily accepted and the view expressed by one Advocate General is very much to the point: [The fundamental principles of national legal systems] contribute to forming that philosophical, political and legal substratum common to the Member States from which through the case-law an unwritten Community law emerges, one of the essential aims of which is precisely to ensure the respect for the fundamental rights of the individual.69

67   Phrased in Hohfeldian terms the rights conferred by EU law are now enjoyed as immunities rather than liberties. 68   Mancini, Keeling, n 58 at 183. 69   Internationale Handelsgesellschaft, n 65.

228  Ola Zetterquist The significance of the legal/philosophical substratum of individual rights protection was at the centre of attention in Kadi.70 In this case the ECJ held that a regulation (EC/881/2002), which implemented the decisions of a UN Security Council Sanctions Committee, was invalid since it, inter alia, violated the fundamental values of the Community, specifically the rights to a fair trial and an effective legal remedy.71 The judgment thus features a clear example of the dialogue principle and a readiness to assess (otherwise validly enacted) norms against the fundamental values (res publica) of the legal order. The interplay between the national courts and the ECJ (and indeed, in the Kadi case, the interplay between the ECJ and the UN Security Council) that developed the fundamental rights protection in the EU also strikes one of Machiavelli’s home-keys: the interaction and discord between the two levels contribute to creating a better European republic.72 The friction between the national and European courts creates a dynamic dialogue that may even contribute to developing a common European legal system and strengthen the legitimacy of the integration process if the two levels are seen as comparatively equal, rather than an imperial court imposing its will on the national ones.73 It is certainly true that the institutional framework for the EU has not caught up with its judge-made development with its exodus from the origins in public international law to a constitutional supra-national legal order. The institutions still reflect a European Union where the states have a powerful position between the individuals and the institutions of the EU. On balance, however, the interplay between national courts and the ECJ has developed in such a way as to qualify as a constitutional legal order with individuals as well as states (and it should be recalled that according to the republican ideal the state has no moral standing besides that of its individuals). Introducing a constitutional bond between individuals in the EU strengthens freedom understood as non-domination as the rights conferred by the EU become contestable under common judges and legislators. The legal process has all through the history of the European Union proved one of the most effective ways of civic participation in the functioning of the Union and for ensuring accountability of the political bodies both at national and European level. A good example on this point was given in the second Defrenne 74 case where the principle of equal pay for men and women was enforced judicially on the not so enthusiastic Member States in the Council (where these states had unanimously agreed not to apply the principle). It is also true that the EU clearly falls short of a political parliamentary democracy of the sort found at the national level but it would be unfair not to recognise   Case C-402/05 P Kadi and Al Barakaat v Council and Commission [2008] ECR I-6351.   Ibid, para 303. 72   D Edward, ‘National Courts – the Powerhouse of Community Law’ (2002) 5 The Cambridge Yearbook of European Legal Studies 1, 1. 73   P Cramér, ‘Does the Codification of the Principle of Supremacy Matter?’ (2004) 7 The Cambridge Yearbook of European Legal Studies 57, 75. 74   Case 43/75 Defrenne (II) [1976] ECR 455. 70 71

Out with the New, in with the Old  229 the progress made over time in this domain. It is reasonable to see some degree of connection between advances in the European legal and political domain respectively and the ECJ has thus indirectly had an important role in this process. Even though the EU does not function as a parliamentary democracy we should not be blind to the possibility that it is, with the words of Neil MacCormick, adequately democratic75 and that it functions as a republic, with a res publica of its own, which aspires to be more than just the cumulating of national interests. Indeed returning to the republican idea of a mixed Constitution and power-­dispersal yields some rather interesting results as far as the EU is concerned which has been pointed out by, to give one example, Neil MacCormick.76 The EU combines the classical elements of aristocracy (the Commission), direct democracy (the European Parliament), indirect democracy (the Council, that is the Member States) and now, following the Lisbon Treaty coming into force, has a president to represent the monarchical aspect (without any dynastic pretensions, one hopes). Machiavelli and Cicero would have been pleased!

VIII.  Conclusion – Roma Locuta, Causa Finita Est There are some distinct advantages in viewing the EU through the republican lens. First, it places the emphasis on individuals rather than on states. In this sense it restores international law – ius gentium – to the people from its temporary hijackers, the states. Secondly, it focuses on substance rather than form and leads to value hierarchies rather than norm-hierarchies. To put it in republican terms, it is more interesting to discuss whether we are free by European laws rather than free from them. Thirdly, it broadens the concept of voice to include the legal domain in addition to the political. It is no secret that the EU has had considerably more success in the former area compared to the latter. The legal channel as a forum for a moral dialogue resonates well with the reasoning of the ECJ in the constitutionalising cases where emphasis has been laid on the legal protection of the individuals in the EU, thereby reducing the scope for arbitrary interferences in their legal position. In this sense an emerging European citizenship can be found in the case law of the ECJ. Finally, a republican lens provides a way out of the sovereignty trap where sovereignty is essentially a zero-sum game: either the Member States or the EU has it. These three elements combined form building stones for a pluralistic European republican Constitution. In sum: the republican perspective offers Leviathan the possibility of a well-deserved break after more than 350 years of hard work.

  MacCormick, n 16 at 148.   Ibid 146.

75 76

10 Institutional Dimension of Constitutional Pluralism JAN KOMÁREK

I.  Constitutional Pluralism and its Institutional Dimension A.  Constitutional Pluralism The concept of constitutional pluralism seems to be in fashion today; so much so that it was labelled a ‘movement’ by one of its critics.1 But, as recently noted, it ‘has paid a price for its popularity [since it] has gained so many meanings that often the participants in the debate talk past each other, each endorsing a different understanding of what constitutional pluralism actually means’.2 As I understand it, constitutional pluralism obtains when various constitutional authorities compete over the same territory and the same legal relationships. It differs from a mere plurality of constitutional sources in that these authorities have plausible claims to legitimacy and authority as perceived by those who are subject to them. Understood in this way constitutional pluralism seems to provide an accurate description of what we have in the European Union today, the Maastricht decision of the German Federal Constitutional Court3 being the ‘paradigmatic piece

1   See J Baquero Cruz, ‘The Legacy of the Maastricht-Urteil and the Pluralist Movement’ (2008) 14 European Law Journal 389. The notion of legal pluralism is much broader and legal sociologists had been using it long before it became the common parlance of European lawyers. See M Avbelj, ‘The EU and the Many Faces of Legal Pluralism: Toward a Coherent or Uniform EU Legal Order?’ (2006) 2 Croatian Yearbook of European Law and Policy 377. 2   Introductory notes in M Avbelj and J Komárek (eds), ‘Four Visions of Constitutional Pluralism – Symposium Transcript’ (2008) 2 European Journal of Legal Studies 325, 325 (a transcript of a symposium organised by the editors with contributions by Julio Baquero Cruz, Mattias Kumm, Miguel Poiares Maduro and Neil Walker, which also provides an accessible starting point for constitutional pluralism). 3   Case 2 BvR 2134, 2159/92 Maastricht [1993] BVerfGE 89, 155.

232  Jan Komárek of [its] dogmatics’,4 now well internalised by the constitutional courts of other Member States.5 But is constitutional pluralism tenable and normatively attractive for the European Union? It does not fit well with constitutionalism, since it denies order and organisation, while constitutionalism imposes them.6 Some pluralist theories propose important limitations on it – so much so that one can question their pluralist nature.7 Neil MacCormick did this most explicitly when he departed from his initial praise of pluralism expressed by the Maastricht decision,8 which he later labelled as ‘radical’, and suggested ‘pluralism under international law’ instead,9 plainly admitting that it was ‘a kind of “monism” in Kelsen’s sense’.10 In this Chapter I do not intend to defend constitutional pluralism as an attractive theory underlying European integration. What I want to do, however, is to suggest some virtues that constitutional pluralism brings through its institutional dimension.

B.  Institutional Dimension Constitutional pluralism, which challenges the hierarchical ordering imposed by the principle of primacy, opens doors to an examination of whether the institution which takes a particular decision is in the best position to adopt it and whether the effects of its decision should persist beyond the context of that particular situation. The decision of the Polish Constitutional Tribunal11 concerning implementation of the European Arrest Warrant Framework Decision12 is an example of this. The Tribunal examined whether the implementing laws breached the constitutional prohibition on the extradition of Polish citizens13 which was required, under certain circumstances, by the EU Framework Decision. It is important   See Baquero Cruz in Avbelj and Komárek, n 2 at 335.   See Baquero Cruz, n 1 at 397–403. Baquero Cruz, however, also questions the descriptive accuracy of constitutional pluralism. See also G Davies, ‘Constitutional Disagreement in Europe and the search for Pluralism’, Chapter 12 in this volume and A Somek, ‘Monism: A Tale of the Undead’, Chapter 15 in this volume. 6   See N Walker, ‘Constitutionalism and Pluralism in Global Context’, Chapter 2 in this volume and M Avbelj, ‘Can European Integration be Constitutional and Pluralist – Both at the Same Time?’, Chapter 16 in this volume. 7   See Avbelj and Komárek, n 2 at 343–45. 8   See Neil MacCormick, ‘The Maastricht-Urteil: Sovereignty Now’ (1995) 1 European Law Journal 259. 9   See N MacCormick, Questioning Sovereignty (Oxford, Oxford University Press, 1999) 97–121 (‘Juridical Pluralism and the Risk of Constitutional Conflict’). 10   Ibid 121. 11   Polish Constitutional Tribunal, Case P 1/05 judgment of 27 April 2005, European Arrest Warrant, www.trybunal.gov.pl/eng/summaries/summaries_assets/documents/P_1_05_full_GB.pdf. 12   Council Framework Decision 2002/584/JHA of 13 June 2002 on the European arrest warrant and the surrender procedures between Member States [2002] OJ L 190/1. 13   Article 55(1); English translation www.trybunal.gov.pl/eng/Legal_Basis/constitution.htm. 4 5

Institutional Dimension  233 to know that the Polish Government sought to amend the Constitution when implementing laws were being adopted; the proposal was, however, rejected by the Polish Sejm.14 Under the ‘simple’ EU primacy rule the Tribunal could never have reviewed the legislation which reflected requirements of EU law since, as the Court of Justice ruled in Internationale Handelsgesellschaft, ‘the law stemming from the Treaty, an independent source of law, cannot because of its very nature be overridden by rules of national law, however framed’, meaning by that ‘either fundamental rights as formulated by the Constitution of that State or the principles of a national constitutional structure’.15 The Tribunal, however, did not follow this logic.16 The Tribunal found the legislation contrary to the Constitution, but at the same time sought to avoid open conflict with the EU by postponing the effects of its decision for 18 months, urging the legislator to adopt a corresponding constitutional amendment. In this way it both avoided a breach by Poland of its obligations stemming from the EU Treaty and brought the constitutional legislator back into the process. Whether or not we praise the Tribunal’s decision depends on how much we want courts to change the meaning of Constitutions they are called to interpret; as is obvious, I favour a more restricted approach, which would require a much longer argument.17 Here I want to point out only the possibility of courts involving constitutional legislators in constitutional conflicts occasionally arising in the European Union. In other words, constitutional pluralism makes the institutions behind particular decisions ‘visible’. As I discuss below, institutions are very much overlooked by many theorists studying European integration. Consider Eric Stein, a pioneer of the constitutional reading of European integration, who focused entirely on the Court of Justice and the doctrines which have formed the basis of the EU Constitution. It was the court which ‘fashioned a constitutional framework for a federal-type structure in Europe’.18 Similarly Koen Lenaerts – himself a judge at the court – tells the story of the constitutionalisation of the Union in terms of the court’s judgments and takes its pronouncements as conclusive evidence for that. In 1990, for example, he observed: At present, the constitutional character of the European Community Treaties stands beyond doubt. In 1986, the Court of Justice even ruled that ‘the European Economic Community is a Community based on the rule of law, inasmuch as neither its Member 14  For a detailed discussion of the process of implementing the Framework Decision and the accompanying constitutional amendment in Poland see A Nußberger, ‘Poland: The Constitutional Tribunal on the implementation of the European Arrest Warrant’ (2008) 6 International Journal of Constitutional Law 162, 162–64. 15   Case 11/70 Internationale Handelsgesellschaft [1970] ECR 1125, para 3. 16   It must be noted, however, that the Framework Decision was adopted within the EU third pillar, where primacy of EU law was not clearly established by the Court of Justice. 17   See A Albi, ‘Supremacy of EC Law in the New Member States Bringing Parliaments into the Equation of “Co-operative Constitutionalism”’ (2007) 3 European Constitutional Law Review 25. 18   E Stein, ‘Lawyers, Judges, and the Making of a Transnational Constitution’ (1981) 75 American Journal of International Law 1, 1.

234  Jan Komárek States nor its institutions can avoid a review of the question whether the measures adopted by them are in conformity with the basic constitutional charter, the Treaty’.19

Both Stein and Lenaerts emphasise legality and the rule of law values. The EU Treaties are binding on the Member States (because they are, after all, legal documents). Regardless of the depth and scope of the Treaties’ claims, for this sole reason they are enforceable even against the Member States’ constitutions. The EC Treaty contains Article 220, which states that it is the Court’s task to ensure that in the interpretation and application of the Treaty ‘the law is observed’. To Lenaerts, this ‘in turn implies the tasks of “umpiring the federal system” and “drawing lines”’.20 The EU Constitution is therefore what the Court of Justice says it is. It implies not only the absolute supremacy of EU law; it also means judicial supremacy.21 Mattias Kumm offers a competing vision, which transcends the Court of Justice’s self-understanding. The European Union’s constitutional authority is based on the ‘principles of common European constitutionalism’. These principles ‘are a common heritage of the European constitutional tradition as it has emerged in the second half of the 20th century’,22 and comprise liberty, democracy, respect for human rights and fundamental freedoms, and the rule of law. These principles are to be balanced by courts so that they can be realised to the greatest extent possible.23 According to Kumm ‘citizens in Europe . . . are governed by national and European officials whose practices are ultimately guided by principles of republican constitutionalism that are shared by the European Union and its Member States’.24 By ‘officials’ Kumm means primarily courts, while the principles are to be defined (or specified) in the process of adjudication. Kumm does not single out the Court of Justice. Member States’ courts also have (at least sometimes) the ‘last’ word if their interpretation can satisfy the principles of common European constitutionalism to a greater extent than that offered by the Court of Justice. So in this way Kumm’s theory can be read as ‘pluralist’, in the sense that it presupposes competition between different constitutional authorities, one represented by the Court of Justice, the other by Member States’ (constitutional) courts. Nevertheless, it is single-institutional judicial supremacy,

19   K Lenaerts, ‘Constitutionalism and the Many Faces of Federalism’ (1990) 38 American Journal of Comparative Law 205, 210, quoting Case 294/83 Les Verts [1986] ECR 1339, para 23. 20   K Lenaerts, ‘Federalism: Essential Concepts in Evolution – The Case of the European Union’ (1998) 21 Fordham International Law Journal 746, 797 (footnotes omitted). 21   The term refers to the US debate. For an introduction see L Alexander and F Schauer, ‘Defending Judicial Supremacy: A Reply’ (2000) 17 Constitutional Commentary 455 with further references. 22   M Kumm, ‘Beyond Golf Clubs and the Judicialization of Politics: Why Europe Has a Constitution Properly So Called’ (2006) 54 American Journal of Comparative Law 505, 517. 23   Kumm develops this idea in his previous article, ‘The Jurisprudence of Constitutional Conflict: Constitutional Supremacy in Europe before and after the Constitutional Treaty’ (2005) 11 European Law Journal 262, 286–88. 24   Kumm, n 22 at 507. Kumm uses various terms for these principles, but if I understand him correctly, ‘principles of Common European constitutionalism’ and ‘principles of republican constitutionalism’ are synonymous.

Institutional Dimension  235 albeit of a different kind from that of Lenaerts.25 It is for courts to give content to the ‘principles of common European constitutionalism’. Many influential theories of European constitutionalism leave its institutional dimension unexplored, although they do not place courts (or any other insti­ tution) at the centre. Joseph Weiler, for example, argues that the principle of constitutional tolerance – whereby the Union’s Member States are invited to comply with the Union’s constitutional authority instead of being obliged to do so26 – must embrace a wide set of actors, like immigration officials overturning practices of decades and centuries and learning to examine the passport of Community nationals in the same form, the same line, with the same scrutiny of their own nationals . . . a similar discipline will be practised by customs officials, housing officers, educational officials and many more subject to the disciplines of the European constitutional order.27

In most of his writings, however, Weiler either remains on a rather abstract level or focuses on a particular institution, most often courts and their role in European integration.28 This is not an accusation of ‘institutional blindness’, in the sense that Weiler would either construct a grand theory without any concern for its institutional context or unilaterally favour one institution (as the theories based on judicial supremacy do).29 However, the complex relationships between constitutional authorities also involve different institutions which make claims to legitimacy of a different kind, and this fact does deeply influence the more abstract questions. It is not just the Union on the one side, and its Member States on the other. It is also the Court of Justice (sometimes together with domestic courts) versus domestic legislators,30 the Commission versus domestic courts,31 or courts versus the market32 – to mention just some of the possible relationships. 25   Kumm provides a justification for such a strong role for the courts in ‘Institutionalising Socratic Contestation: The Rationalist Human Rights Paradigm, Legitimate Authority and the Point of Judicial Review’ 1 European Journal of Legal Studies (December 2007). 26   This, of course, is just too short a summary of Weiler’s approach to European constitutionalism and the principle of constitutional tolerance, which is present in most of his writings. See particularly JHH Weiler, ‘In Defence of the Status Quo: Europe’s Constitutional Sonderweg’, in JHH Weiler and M Wind (eds), European Constitutionalism Beyond the State (Cambridge, Cambridge University Press, 2003). 27   Weiler n 26 at 21. 28   See JHH Weiler, The Constitution of Europe. “Do the New Clothes Have An Emperor?” and Other Essays on European Integration (Cambridge, Cambridge University Press, 1999). 29  See A Vermeule, Judging under Uncertainty. An Institutional Theory of Legal Interpretation (Cambridge, Mass; London, Harvard University Press, 2006) 16–18. 30   On this see the text accompanying nn 33–36. 31   Consider, eg, the duty of Member States’ courts to defer to the Commission’s determinations concerning violations of the competition rules. On this see particularly AP Komninos, ‘Effect of Commission Decisions on Private Antitrust Litigation: Setting the Story Straight’ (2007) 44 Common Market Law Review 1387. Courts’ deference to administrative agencies’ interpretation of law is subject to a continuous debate in the US (see eg Vermeule, n 29 at 207–08), which seems to be entirely missing in the EU. 32   See eg a rather rare instance of institutional analysis in the Opinion of AG Poiares Maduro in Case C-438/05 International Transport Workers’ Federation and Finnish Seamen’s Union [2007] ECR I-10779 (‘Viking Line’), paras 41 and 42.

236  Jan Komárek To illustrate the importance of the institutional dimension of European constitutionalism, take the following example. Some people suggest that the Court of Justice does not face ‘the countermajoritarian problem’, since there is ‘less dem­ ocracy at the European level’.33 However, the court often deals with decisions adopted in the domestic democratic process and the conflict between the EU and its Member States can be read as a kind of countermajoritarian difficulty too. In its Maastricht decision34 the German Federal Constitutional Court asserted that the final source of legitimacy of EU law must be found in domestic democratic procedures, because there is not sufficient democracy to satisfy the German stand­ ard at the European level. According to Joseph Weiler, ‘the German Court presents itself as a guarantor of the universal values of democracy rather than as a guarantor of German particularism’.35 On the other hand, however, Weiler alerts us to who the guarantor of ‘true’ democracy is: courts. ‘[D]efending the constitutional identity of the state and its core values turns out in many cases to be a defence of some hermeneutic foible adopted by five judges voting against four.’36 Miguel Poiares Maduro takes institutions and the institutional limitations of courts most seriously, taking his cue expressly from Neil Komesar.37 Institutional choice is one of Poiares Maduro’s contrapuntal38 principles, designed ‘to guide the ordinary state of affairs’ in the relationships between the EU and its Member States. It requires that ‘each legal order and its respective institutions must be fully aware of the institutional choices involved in any request for action in a pluralist legal community’.39 In relation to courts, Poiares Maduro highlights that they must increasingly be aware that they don’t have a monopoly over rules and that they often compete with other institutions in their interpretation. They have to accept that the protection of the fundamental values of their legal order may be better achieved by another institution or that the respect owed to the identity of another legal order should lead them to defer to that jurisdiction. This requires courts to both develop instruments for institutional comparison and to set the limits for jurisdictional deference at the level of systemic identity.40

33   M Rosenfeld, ‘Comparing constitutional review by the European Court of Justice and the US Supreme Court’ (2006) 4 International Journal of Constitutional Law 618, 631. 34   See n 3. 35  JHH Weiler, ‘Does Europe Need a Constitution? Demos, Telos and the German Maastricht Decision’ (1995) 1 European Law Journal 222. 36   Weiler, n 26 at 17. 37   Imperfect Alternatives. Choosing Institutions in Law, Economics, and Public Policy (Chicago, University of Chicago Press, 1994). 38   See M Poiares Maduro, ‘Contrapunctual Law: Europe’s Constitutional Pluralism in Action’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 532. I am aware that I am contributing to the terminological confusion which so often blurs much of European (constitutional) scholarship. However, Poiares Maduro refers to a musicological concept, a counterpoint. The correct adjective, then, is ‘contrapuntal’, not ‘contrapunctual’. 39   Ibid 530. 40   M Poiares Maduro, ‘Interpreting European Law: Judicial Adjudication in a Context of Constitutional Pluralism’ (December 2007) European Journal of Legal Studies.

Institutional Dimension  237 Taking a more critical perspective, however, one can ask whether Poiares Maduro truly goes beyond the single-institutionalism of the theories based on judicial supremacy and, if so, whether he takes the institutional dimension of constitutionalism more seriously than ‘institutionally indifferent’ theories. As regards the first question, I have already examined whether the principle of universalisability does not in fact prevent institutional choice and deny pluralism (another contrapuntal principle), at least in some instances.41 According to the universalisability principle courts should use reasoning, which may be adopted by other courts within the Union.42 Consequently, under this principle courts are not allowed to rely as justification on specific provisions of their Constitutions aimed at nationals of a particular state because that could lead to ‘evasion and free-­ riding’. However, this forces the courts to solve most conflicts by themselves and leaves no space for other institutions to step in. While I would praise the Polish Constitutional Tribunal for leaving the decision on a conflict between a clearly formulated constitutional prohibition to extradite Polish nationals on the one hand and the requirement to do so imposed by the European Arrest Warrant Framework Decision on the other to the constitutional legislator,43 Poiares Maduro would most probably condemn the Tribunal for taking a decision which cannot be universalised by other Union courts (since the constitutional prohibition protected Polish nationals only). As regards the second question – whether Poiares Maduro takes institutions (sufficiently) seriously – it can be unjust to answer it now, since Poiares Maduro has certainly not had his last word on the matter and his theory is still very much in the making.44 However, his conception of the EU Constitution as a remedy for Member States’ constitutions’ failures is dependent on a proper institutional analysis, which he still needs to do. For example, if Article 28 EC is to be read as a political right, giving traders a means to question the regulatory policies of a state to which they export [this has got to be ‘export’] their goods, forcing that state to justify its regulatory choices,45 one must seriously ask whether the judicial process in the Union is open enough not systematically to favour just some types of litigants.46 Similarly, when Poiares Maduro argues that moving decisions to the 41   See Komárek, ‘European Constitutionalism and the European Arrest Warrant: In Search of the Limits of Contrapunctual Principles’ (2007) 44 Common Market Law Review 9, 33. 42   Poiares Maduro, n 38 at 529–30. 43   Komárek, n 41; also Albi, n 17. 44   See Poiares Maduro, n 40 at 4. 45   See M Poiares Maduro, We the Court. The European Court of Justice and the European Economic Constitution A Critical Reading of Article 30 of the EC Treaty (Oxford, Hart, 1998) 156–59. 46   See eg D Chalmers, ‘The Dynamics of Judicial Authority and the Constitutional Treaty’ (2005) 3 European Journal of Constitutional Law 448, 472. This argument concerning the very ability of the judicial process to make such participation possible must be distinguished from a rather convincing argument made by A Somek, ‘The Argument From Transnational Effects I: Representing Outsiders Through Freedom of Movement’ (2010) 16 European Law Journal 315, 331, dealing with the question of what kind of interests are in fact promoted by the free movement rules: ‘[s]ubstantive economic due process, by contrast, is about protecting the economically active and enterprising vis-à-vis a caring state. It avails of a comparative dimension only inasmuch as it privileges the interest in the generation of wealth. This is not, arguably, a democracy-reinforcing concern, for it shuts down the voices and

238  Jan Komárek Union level can open deliberation on decisions ‘otherwise embedded with certain values and assumptions that are no longer subject to deliberation’,47 one must examine whether the Brussels institutions are truly so far away from (at least some) Member States’ capitals. And so on. But even if Poiares Maduro ultimately fails to provide answers to these questions, he has certainly opened the door to a very different kind of analysis of European constitutionalism. The following is an attempt to move Poiares Maduro’s framework further, and also shift its focus somewhere else, where I believe it can yield some useful results. I would also want to make clear, however, that the institutional dimension of constitutionalism in the European Union reaches beyond the sphere of ‘traditional’ constitutional pluralism which is concerned with constitutional conflicts in the European Union.48 Taking the institutional dimension seriously, one can for example ask whether the ‘Constanzo mandate’,49 which requires Member States’ administrative authorities to decide not to apply domestic legislative rules if they conflict with Community law, can be maintained. In Fratelli Constanzo the court stated that it would be contradictory to rule that an individual may rely upon the provisions of a directive which fulfil the conditions defined above in proceedings before the national courts seeking an order against the administrative authorities, and yet to hold that those authorities are under no obligation to apply the provisions of the directive and refrain from applying provisions of national law which conflict with them.50

The court finds support for this conclusion in the fact that obligations arising under the provisions of a directive are binding upon all the authorities of the Member States.51 The court does not distinguish between the various institutions of the state – all are equally bound to give primacy to the provisions of the directive over conflicting domestic legislative rules. Now imagining any tax law authority to be empowered to disregard domestic legislation which it believes to be fades out the interests of those who benefit from more restrictions that have been adopted by democratic means. Hence, substantive economic due process is non-democratic in its orientation’. 47   M Poiares Maduro, ‘How Constitutional Can the European Union Be? The Tension Between Intergovernmentalism and Constitutionalism in the European Union’, in JHH Weiler and CL Eisgruber (eds), Altneuland: The EU Constitution in a Contextual Perspective, Jean Monnet Working Paper No 5/04, http://www.jeanmonnetprogram.org/papers/04/040501-18.html. 48   According to M Poiares Maduro, ‘Courts and Pluralism: Essay on a Theory of Judicial Adjudication in the Context of Legal and Constitutional Pluralism’ in J Dunoff and J Trachtman (eds), Ruling the World? Constitutionalism, International Law and Global Government (Cambridge, Cambridge University Press, 2009) 356–57 and fn 1, ‘the core and starting point of traditional constitutional pluralism analysis in the context of the European Union’ consisted of an examination of ‘the acceptance of the supremacy of EU rules over national constitutional rules [which] has not been unconditional and has been even, at times, resisted by national constitutional courts [conferring] to EU law a kind of contested or negotiated normative authority’. 49   Case 103/88 Fratelli Costanzo [1989] ECR 1839. See M Claes, The National Courts’ Mandate in the European Constitution (Oxford, Hart, 2006) 266–78. 50   Fratelli Costanzo, n 49, para 31. 51   Ibid, para 30.

Institutional Dimension  239 incompatible with Community law does not seem to be a very attractive idea, especially as it can escape all judicial control. If the administrative decision grants rights to an individual, there is often nobody to challenge the legality of that decision in the courts. This problem was recently noted by Advocate General Colomer, who seemed to agree with one commentator that ‘there is something “deeply disturbing” underlying the Fratelli Costanzo case law’.52 Another example of the Court of Justice’s disregard for the institutional dimension of its rulings can be found in its Member State liability case law, especially that concerning breaches of Community law committed by Member States’ courts of final appeal. There the court also emphasises that ‘[i]n international law a State which incurs liability for breach of an international commitment is viewed as a single entity, irrespective of whether the breach which gave rise to the damage is attributable to the legislature, the judiciary or the executive’.53 According to the court, ‘[t]hat principle must apply a fortiori in the Community legal order since all State authorities, including the legislature, are bound in performing their tasks to comply with the rules laid down by Community law which directly govern the situation of individuals’.54 When the Member States pointed out the difficulties posed in their judicial systems (since the same court which committed the breach may adjudicate on a subsequent liability claim),55 the court responded that ‘[i]t is not for [it] to become involved in resolving questions of jurisdiction to which the classification of certain legal situations based on Community law may give rise in the national judicial system’.56 This of course does not require Member States’ courts to start to disregard the Court of Justice’s rulings in Fratelli Constanzo or Köbler. It is also for the Court of Justice to take the institutional dimension seriously. In this respect the importance of constitutional pluralism (in its narrow meaning) is only marginal, since problems concerning institutional choice are far more pervasive. They have been observed primarily in relation to Member States’ courts and their empowerment, for example by giving them the power of judicial review over national legislation.57 But the problem of institutional choice is a much wider phenomenon which should be reflected by the Court of Justice and other actors too.

52   Opinion in Case C-205/08 Alpe Adria Energia [2009] ECR I-11525, fn 38, referring to M Bobek, ‘Thou Shalt Have Two Masters; The Application of European Law by Administrative Authorities in the New Member States’ (2008) 1 Review of European Administrative Law 51, 63. 53   Case C-224/01 Köbler [2003] ECR I-10239, para 32. 54   Ibid. 55   See J Komárek, ‘Federal Elements in the Community Judicial System – Building Coherence in the Community Legal Order’ (2005) 42 Common Market Law Review 9, 29–30. 56   Köbler, n 53, para 47. 57   See B de Witte, ‘Direct Effect, Supremacy, and the Nature of the Legal Order’ in P Craig and G de Búrca (eds), The Evolution of EU Law (Oxford, Oxford University Press, 1999) 207–08.

240  Jan Komárek

II.  Institutions: From Conflict and Choice to Communication and Involvement A.  Institutional Analysis The institutional dimension of constitutionalism has received more attention in the United States.58 Although Cass Sunstein and Adrian Vermeule called for an ‘institutional turn’ as late as in 2003,59 Richard Posner pointed out that their claims were not as novel as the authors suggested.60 William Eskridge even asserted that ‘[t]he institutional turn started almost a century ago, with the recognition that judges should consider relative institutional competence when interpreting statutes or even the common law’61 and identified several institutional turns in the history of US legal thought.62 The problem does not seem to lie in the ‘institutional blindness’ of various theorists.63 The paradigmatic example of the scholarship studying institutions and their relative competences is Henry Hart and Albert Sack’s The Legal Process, a set of teaching materials prepared by the authors in 1958 (although actually published only posthumously in 1994).64 It ‘was a part of a larger collective effort to synthesize the lessons of pre-war American law – the realist legacy of law as function and policy, the institutional competence idea central to the regulatory state, and the rationalist view of law as reasonable and coherent’.65 The civil rights revolution of the 1960s revealed limitations on the traditional legal process analysis, and was perhaps among the reasons why The Legal Process 58   Although it is on the rise in Europe as well; see particularly JA King, ‘Institutional Approaches to Judicial Restraint’ (2008) 28 Oxford Journal of Legal Studies 409. 59   ‘Interpretation and Institutions’ (2003) 101 Michigan Law Review 885. Significant parts of this article appear in Vermeule, n 29, which I use as my primary source. 60   ‘Reply: The Institutional Dimension of Statutory and Constitutional Interpretation’ (2003) 101 Michigan Law Review 952. See also Sunstein and Vermeule’s response, ‘Interpretive Theory in Its Infancy: A Reply to Posner’ (2003) 101 Michigan Law Review 972. 61   See ‘No Frills Textualism’ (2006) 119 Harvard Law Review 2041, 2045 (a review of Vermeule, n 29). 62   See Eskridge, n 61 at 2044–51. I believe that it does not make much sense to talk about ‘turns’, since institutional analysis (in different forms) has been present in much of the scholarship throughout the 20th century. On the legal process school (and related approaches) see also N Duxbury, The Patterns of American Jurisprudence (Oxford, Clarendon Press, 1995) at 205–99 (making the same point at 206–07). 63   There are, of course, theorists who do not pay attention to the institutional dimension, a view which is not justified once they start making practical pronouncements on what, for example, courts should do. See Posner, n 60 at 955, who defends Ronald Dworkin against some of the charges raised by Sunstein and Vermeule, n 29 at 902–4, since ‘scholars are permitted to discuss one question at a time’, but then admits that Dworkin can be faulted ‘in using his partial analysis as the basis for confident evaluations of particular interpretive issues, such as the assisted-suicide issue that Sunstein and Vermeule discuss’. 64   See HM Hart and AM Sacks (WN Eskridge and PP Frickey eds), The Legal Process. Basic Problems in the Making and Application of Law (Westbury, NY, Foundation Press, 1994). 65   See WN Eskridge and PP Frickey, ‘An Historical and Critical Introduction to The Legal Process’, in Hart and Sack, n 64 at c.

Institutional Dimension  241 was never published by its authors (despite its wide influence on US legal academia).66 The main weaknesses consisted in ‘its polarized categorizations (for example, substance/procedure), its undue optimism about the competence and public-spiritedness of state institutions, and its failure to recognize the ideological and non-neutral nature of its own positions’.67 But the legal process’s influence endured, and both critical legal studies and law and economics ‘defined themselves, in part, by attacking legal process’.68 Moreover, another iconic casebook of that period, The Federal Courts and the Federal System, authored in its first edition by Henry Hart with Herbert Wechsler,69 has remained hitherto the primary teaching material in the field.70 Neil Komesar’s Imperfect Alternatives,71 already mentioned in relation to Miguel Poiares Maduro’s work in the context of the EU, and Adrian Vermeule’s Judging under Uncertainty,72 both distance themselves from the traditional legal process school. For Komesar, ‘Hart and Sacks presented a largely idealized image of institutions. Each institution was assumed to be a contemplative, deliberative, rational decision-maker’,73 while in Vermeule’s opinion, ‘Hart and Sacks’s elaborate talk about institutional competence is undercut by their stylized, nonempirical treatment of actual institutions and their capacities, and by their crude treatment of the systemic effects of competing interpretive approaches’.74 Both Komesar and Vermeule emphasise the need for comparative dimension of institutional analysis: ‘[w]hether, in the abstract, [any of the institutions] is good or bad at something is irrelevant’. ‘The correct question is,’ according to Komesar, ‘whether, in any given setting, [a particular institution] is better or worse than its available alternatives.’75 Moreover, the comparison must be ‘symmetrical’; it is a mistake ‘to take a cynical or pessimistic view of some institutions and an unjustifiably rosy view of others’.76 A related requirement on a sound institutional analysis is therefore that it must be ‘evenhandedly empirical’ – ‘realistic about the capacities of all relevant actors’.77

  See Eskridge and Frickey, n 65 at xcviii.   Ibid civ. 68  EL Rubin, ‘The New Legal Process, the Synthesis of Discourse, and the Microanalysis of Institutions’ (1996) 109 Harvard Law Review 1393, 1398. 69   (Brooklyn, Foundation Press, 1953.) Eskridge and Frickey, n 65 at cii place this book among ‘the central works of what is now recognized as a legal process tradition’. 70   See RH Fallon, ‘Reflections on the Hart and Wechsler Paradigm’ (1994) 47 Vanderbilt Law Review 953. The book is now in its sixth edition (RH Fallon, Jr et al, Hart and Wechsler’s The Federal Courts and the Federal System 6th edn (New York, Foundation Press, 2009). 71   See n 37. 72   See n 59. 73   Komesar, n 37 at 12. 74   Vermeule, n 29 at 27. See also Sunstein and Vermeule, n 59 at 900–02. 75   All quotations from Komesar, n 37 at 6 (I generalised the point made by Komesar in relation to two concrete institutions: the market and political process). 76   Vermeule, n 29 at 17. 77   Ibid 18. 66 67

242  Jan Komárek

B.  Problems of Institutional Choice The following is not intended as a sweeping critique of institutional choice. I want only to show the problems it has (both theoretical and practical), which leads me to try to find other ways in which the institutional dimension can usefully be explored. The lack of consensus on how to evaluate institutional competence, resulting in different authors’ preference for different institutions, is most troubling. Two (somewhat related) problems concern, first, the practical feasibility of institutional choice and, second, its orientation on the decision makers who are often supposed to limit their actions because of their relative institutional weakness. Adrian Vermeule does not present any clear criteria for institutional choice. When he discusses interpretive choice,78 which implies institutional choice, he mentions only some relevant considerations. They are ‘information and superior competence to translate information into sound legal policy’,79 ‘trustworthiness of agents and delegates’ (will agents ‘act to promote the principal’s interest or their own?’)80 and ‘the likely reaction of other institutions to the rules or standards chosen by a decisionmaker engaged in interpretive choice’.81 After having chosen a formalist interpretation (which should limit the scope of discretion on the part of courts),82 Vermeule presents ‘institutional variables’.83 He asks whether ‘a formalist or nonformalist judiciary, in one or another domain, produce more mistakes and injustices’.84 ‘Cognition and motivation of interpreters’,85 the possibility of error correction by another institution,86 and, finally, decision costs and costs of uncertainty87 are other institutional variables. They all sound relevant, but how is one to measure them? Vermeule does not say. For example, how do we know whether or not the Supreme Court’s decision is ‘correct’ and ‘just’? We ask courts such questions, and the court of final appeal gives the final and ‘correct’ legal answer.88 Of course people can disagree on whether the decision was ‘right’,   See Vermeule, n 29 at 67–70.   Ibid 69. 80   Ibid. 81   Ibid 70. 82   Vermeule’s book can equally be read as a defence of interpretive formalism and preference for administrative agencies. See C Nelson, ‘Statutory Interpretation and Decision Theory’ (2007) 74 University of Chicago Law Review 329 (a review of Vermeule, n 29). But Vermeule seems not to remain faithful to his own pronouncements and the choice he makes is not based on any empirical evidence (see Eskridge, n 61 at 2046). 83   See Vermeule, n 29 at 76–79. 84   Ibid 77. 85   Ibid 77–78. 86   Ibid 78. 87   Ibid 79. 88  Thus Brown v Allen, 344 U.S. 443, 540 (1953) (Jackson J concurring): ‘We are not final because we are infallible, but we are infallible only because we are final.’ It is one of the peculiar features of EU law and its judicial system that this does not quite apply there, since the Court of Justice constantly undermines hierarchical features of Member States’ judicial systems. See J Komárek, ‘“In the Court(s) We Trust?” On the Need for Hierarchy and Differentiation in the Preliminary Ruling Procedure’ (2007) 32 European Law Review 467 at 486–89. 78 79

Institutional Dimension  243 but they would have to agree on some non-legal criteria according to which to do this – in other words, on some substantive theory. And that is something Vermeule (and Sunstein too) wants to avoid.89 Similarly, what are decision costs? They are among the central concepts of the theory of judicial minimalism, proposed by Vermeule’s intellectual fellow-traveller, Cass Sunstein.90 Yet, neither of them gives a useful answer to this question. Finally, Vermeule’s institutional choice can be as ‘institutionally blind’ (to use Vermeule’s own label) as other theories of interpretation. To ask real-life decision makers, particularly the courts, to make such complex empirical assessments (which moreover Vermeule himself fails to do) can simply be to ask them to exceed their actual competences. In his reply to Sunstein and Vermeule’s article (which applies with equal force to Vermeule’s book), Richard Posner noted: Casual empiricism is often unavoidable in law. Sunstein and Vermeule deny this. They say the fact that ‘relevant empirical and institutional variables are costly to measure’ (they are speaking of the variables relevant to the choice between loose and strict construction) is ‘hardly an argument for nonempirical interpretive theory.’ It is, in fact, a compelling argument. Unavailability of empirical data does not excuse the judge from having to interpret statutes in the cases that come before him for decision; and to decide how to interpret them he will perforce have to decide whether he is a loose or a strict constructionist. He may not be articulate about the choice, but he will make it nonetheless.91

Neil Komesar differs from Vermeule in several ways, yet I do not think he escapes the problem of the lack of some feasible criteria for evaluating institutions. Komesar presents a ‘participation-centered approach’, which ‘identifies the actions of the mass of participants as the fact that in general best accounts for the variation in how institutions function’.92 The participants comprise consumers, producers, voters, lobbyists and litigants. In Komesar’s opinion, ‘[a]t least initially, official actors in the political process and the judiciary [the two of the institutions Komesar examines, together with the market] – legislators and judges – play a secondary role’.93 ‘[T]he interaction of these many actors rather than the will of a few officials receives central attention in [Komesar’s] analysis.’94 Komesar honestly acknowledges the difficulties his approach meets when applied to concrete problems; difficulties facing not only the real decision makers, but also theorists.95 Apart from the practical feasibility of his analytical framework, one can further ask whether participation should be the only criterion for

  Vermeule, n 29 at 79–85 and Sunstein and Vermeule, n 29 at 914–19.   See CR Sunstein, One Case at a Time: Judicial Minimalism on the Supreme Court (Cambridge, Mass, Harvard University Press, 1999) 47–50. 91   Posner, n 60 at 969, quoting Sunstein and Vermeule, n 59 at 907. 92   Ibid. 93   Ibid. 94   Ibid. 95  See N Komesar, ‘The Perils of Pandora: Further Reflections on Institutional Choice’ (1997) 22 Law and Social Inquiry 999, a response to a book review of Imperfect Alternatives, HS Erlanger and TW Merrill, ‘Institutional Choice and Political Faith’ (1997) 22 Law and Social Inquiry 959. 89 90

244  Jan Komárek institutional comparison.96 In this respect I would like to pray in aid Daniel Halberstam’s idea of constitutional heterarchy. Halberstam suggests that pluralism in the European Union, with its contestation of the final authority, is in some respects similar to that in the United States, where different institutions (the President, the Congress, the Supreme Court, and ultimately, the ‘People themselves’) compete for the final authority to interpret the US Constitution. Constitutional conflict is central to both systems.97 Now as is crucial for my argument against Komesar’s participation-centred approach, each actor makes different claims based on three values, essential to constitutionalism: voice (representation of the relevant political will), expertise (knowledge or the instrumental capacity to decide a particular issue) and protection of rights.98 These values of constitutionalism to a great extent resemble Bruce Ackerman’s ‘three great principles that motivate the modern doctrine of separation of powers – democracy, professionalism, and the protection of fundamental rights’.99 According to Halberstam, none of these values is exclusively or even reliably associated with one or another of the contending actors. At different times, different actors can lay claim to be vindicating any one or more of these values. If an actor can maximize all three values in any given case, that actor’s claim to authority within the system becomes paramount. If, as is more frequently the case, different actors can lay only partial claim to one or the other of these values, the stage is set for constitutional confrontation.100

The participation-centred approach does not use this value-laden language. This has a great advantage since it avoids questions which it cannot (and does not want) to answer – for example, what the ‘correct’ level of protection of rights is. But at the same time it seems to me that it overlooks something which is import­ ant about the role of institutions in constitutionalism and which goes beyond the value of participation. Perhaps it shows the limitations of the institutional analysis, which can never give complete answers to questions which arise. The impossibility of answering some questions, once we do not agree on any substantive theory, is well illustrated by Halberstam’s discussion of the protection of rights. Using the example of the Supreme Court’s decision in Dred Scott v Sandford 101 and also the court’s obstruction of Congress’s implementation of the 96   In this respect one can also link this question to the critique by JH Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge, Mass, Harvard University Press, 1980). Ely’s central argument is that judicial review should be primarily democratic-participation (re)enforcing. Despite Komesar’s critique of Ely’s single-institutionalism (see Komesar, n 37 at 198–215), they share the same basic commitment to participation. 97   See Daniel Halberstam, ‘Constitutional Heterarchy: The Centrality of Conflict in the European Union and the United States’ in J Dunoff and J Trachtman (eds), Ruling the World? Constitutionalism, International Law and Global Government (Cambridge, Cambridge University Press, 2009). 98   Ibid 328. 99   ‘The New Separation of Powers’ (2000) 113 Harvard Law Review 633, 634. 100   Halberstam, n 97 at 337–38. On the link between pluralism in the US Constitution and institutional competence see CL Eisgruber, ‘The Most Competent Branches: A Response to Professor Paulsen’ (1994) 83 Georgetown Law Journal 347. 101   60 U.S. 393 (1857) (ruling that Congress could not forbid slavery in the federal territories).

Institutional Dimension  245 Reconstruction Plan after the War, most famously by its ‘separate but equal doctrine’,102 Halberstam shows how the court ‘declined to provide much meaningful rights protection’.103 The problem, however, arises when one is to assess whether individual rights are ‘better protected’ when there is a deep (and perhaps irreconcilable) disagreement over their content. If courts favour the right to the personal autonomy (or human dignity, or equality) of women over the right to life of an unborn child, are we really sure that they have a claim to rights protection superior to that of a legislator which prohibits abortion?104 The only way to ‘evaluate’ rights protection seems to be institutional analysis. As Christopher Peters notes, Whether the Court or Congress has been more successful at protecting rights depends upon entirely subjective, and not at all empirical, conclusions about which rights are worth protecting. A satisfying assessment of comparative institutional advantages in the protection of rights must, it seems to me, be divorced from particular subjective evaluations of rights, and thus from subjective evaluations of ‘success’ or ‘failure’ in particular cases.105

That, however, meets similar problems of insufficient criteria, as I suggest above. Apart from the absence of criteria for institutional choice, the second kind of problem concerns its orientation on the decision-making institution, which is supposed to make the comparative calculus about its competence and possibly restrain itself from taking the decision or at least limit its scope. I can illustrate the difficulty with the thus conceived institutional analysis on the already mentioned example of the Polish Constitutional Tribunal and its review of the European Arrest Warrant.106 Was it an honest choice the Tribunal had made, or was it only an easy, ‘buck-passing’ way of avoiding adjudication on an issue which was to be duly decided by the constitutional adjudicator?107 Finally: can the decision-making institution ever be persuaded that its own competence is limited and therefore that it must duly defer to another institution? In his (rather severe) critique of Vermeule, Richard Posner observes that ‘[a]cademics who are not seriously engaged with the judiciary urge judges to change by adopting this or that approach, and usually it is an approach designed to clip judges’ wings’.108 That is what Vermeule’s institutional analysis suggests by proposing a more formalist approach to interpretation. But, as Posner argues, ‘[j]udges are not interested in having their wings clipped, but will happily adopt restraintist approaches as rhetorical tools to persuade others that what looks like judicial assertiveness is obedience’.109 I think that here Posner questions   Plessy v Ferguson, 163 U.S. 537 (1896).   See Halberstam, n 97 at 349. 104   See J Habermas, ‘On Law and Disagreement. Some Comments on “Interpretative Pluralism”’ (2003) 16 Ratio Juris 187. 105   ‘Assessing the New Judicial Minimalism’ (2000) 100 Columbia Law Review 1454, 1492, fn 180. 106   See the text accompanying n 11. 107   See Komárek, n 41 at 38. 108   RA Posner, How Judges Think (Cambridge, Mass, Harvard University Press, 2008) 215. 109   Ibid 215–16. 102 103

246  Jan Komárek institutional choice as a meaningful interpretive strategy for constraining the courts’ discretion, the very aim which Vermeule wants to pursue.

C.  Towards Involvement and Communication Approaching the institutional dimension in terms of choice is connected to emphasising the finality and irreversibility of the decisions taken by particular institutions. But very often (not always, of course), it only seems that a particular institution is making a momentous decision which cannot be overturned. In many instances that is not the case. The process of norm creation is not linear, from one institution to another, for example from the legislator which adopts a piece of legislation to courts which only apply it. Nor is judicial review; by striking down a piece of legislation the court only starts a process in which other actors of the constitutional system react.110 As Mark van Hoecke observes, law is to a very great extent circular.111 Constitutional pluralism, with its contestation of finality, only reinforces this circular exchange among various actors.112 Theories of constitutional dialogue (or communication) within a constitutional order are now widespread on both sides of Atlantic – in the US, Canada and Europe.113 Although few of them identify themselves with constitutional pluralism, they share with it the belief that no institution has a superior claim to define or protect a particular constitutional value. The theories can differ in their concrete institutional arrangements and the weight they give to each of the actors – some, for example, favour courts, some favour legislators, but they all emphasise communication among different constitutional actors and their mutual involvement. This turn to dialogue responds to the lack of clear answers and the lack of consensus among those who are concerned by them. It is nicely illustrated by Barry Friedman’s response to some reactions to his article concerning congressional control over the jurisdiction of federal courts – an ever recurrent topic of doctrinal debate since the very founding of the United States. Friedman says: I am presenting an argument of a different genre than the traditional advocacy over the ‘correct’ or even ‘best’ interpretation of article III [which establishes jurisdiction of federal courts]. I am arguing that what I observe ‘just is’ and that our explorations might

110   See in particular B Friedman, ‘Dialogue and Judicial Review’ (1993) 91 Michigan Law Review 577 (in the context of the US) and M Van Hoecke, Law as Communication (Oxford, Hart, 2002) 172–78. 111   See Van Hoecke, n 110 at 37–39. 112   Ibid 209. See also Friedman, n 110 at 653–80 (discussing mechanisms of dialogue in the context of the US Constitution – and at 658–68 giving a concrete example of a dialogue on abortion initiated by the court’s decision in Roe v Wade, 410 U.S. 113 (1973)) and MC Dorf, ‘Foreword: The Limits of Socratic Deliberation’ (1998) 112 Harvard Law Review 4. 113  See C Bateup, ‘The Dialogic Promise. Assessing the Normative Potential of Theories of Constitutional Dialogue’ (2006) 71 Brooklyn Law Review 1109 for an overview and evaluation of such theories.

Institutional Dimension  247 be more useful if we began from this jumping-off point, instead of from insisting that the answer can be found in the Constitution’s text, and the Original Plan.114

Institutional involvement and communication allow each of the actors’ com­ petence to be combined so that a better and more legitimate decision (in comparison to a decision taken by a single actor) can be taken. ‘Better’ because institutions can combine their varied competences and legitimacy claims when arriving at it.115 ‘More legitimate’ since communication itself can have such effects.116 An important point is, of course, that these vitally depend on concrete institutional settings existing in a particular constitutional system.117

III. Conclusion Closely related to emphasising institutional involvement and communication over institutional choice and conflict is shifting the focus from an institution which takes the decision (and should ideally, but in my opinion unrealistically, take its institutional competence into account when making it) to other institutions. How can they react? How can they claim a space for their own decisions? In other words, this means moving from a single-institutional perspective, combined with the illusion of finality, to a multi-institutional and circular perspective. Then, it is important not only how the decision-making institution constrains itself, but also how other institutions can react to its decision and possibly claim space for their own decisions. Institutional choice then appears less important than involvement. It is less important to choose the right institution (if only because such a choice is difficult to make in practice in the absence of clear criteria) than to achieve the involvement of more institutions and find ways of involving them. A promising line of inquiry into the institutional dimension of constitutional pluralism therefore lies in examining how this can be achieved.

114  Barry Friedman, ‘Federal Jurisdiction and Legal Scholarship: A (Dialogic) Reply’ (1991) 85 Northwestern University Law Review 478, 479–80 (containing references to the previous article and the reactions to it). 115   This is essentially based on a ‘many-minds argument’, which has many variants which I cannot explore here. See A Vermeule, ‘Many-Minds Arguments in Legal Theory’ (2009) 1 Journal of Legal Analysis 1 and CR Sunstein, A Constitution of Many Minds (Princeton, Princeton University Press, 2009) (who takes a more positive approach to ‘many minds’ than Vermeule). 116   See particularly Van Hoecke, n 111 at 203–15, who in this respect builds very much on J Habermas, Between Facts and Norms (Cambridge, Polity Press, 1997). 117   See A Vermeule, Law and the Limits of Reason (Oxford, Oxford University Press, 2009) 41–55.

11 Legal Pluralism and Institutional Disobedience in the European Union JULIO BAQUERO CRUZ

I. Introduction In this Chapter I will argue that the different forms of judicial resistance to the absolute supremacy of European Union law that we find scattered in the case law of a number of constitutional courts or courts carrying out constitutional review around Europe are best understood as exceptional instances of institutional diso­ bedience rather than as examples or a more or less virtuous general practice of legal or constitutional pluralism. The argument proceeds as follows. First, I briefly review the main understand­ ings of the relationship between European Union law and national constitutional law, concluding that none of them gives a totally accurate description of current or past practice, and also that none of them is a satisfactory model in normative terms for the future of legal integration in the Union. Secondly, I use the theory of civil disobedience to propose the rudiments of what I should like to call institu­ tional disobedience. In a refined version of the traditional ‘law of integration’ approach to the relationship between Union law and state constitutional law, I shall argue that institutional disobedience best explains and frames those acts of resistance on the part of the constitutional and supreme judicial bodies of some Member States. Rather than seeing resistance and disagreement as a structural and normal part of the legal system, with potentially very negative consequences for the sustainability of integration, for legal certainty and for the rule of law in the Union, institutional disobedience tends to see it as a limited phenomenon directed towards the public voicing of disagreement in the interest of law reform. This understanding, I hope, may pave the way to a more ordered and pro­ ductive view of the relationship between European Union law and state constitu­ tional law, reducing the risks of incoherence and legal uncertainty which are implicit in legal pluralism, particularly in its radical versions. To conclude, I will reflect on the consequences of this approach for the practice of law in the European Union.

250  Julio Baquero Cruz

II.  Three Conceptions of the Relationship between European Union Law and State Constitutional Law1 The three predominant conceptions of the relationship between European Union law and state constitutional law, which roughly correspond to the three main political theories of European integration, are the state-centred conception, the law of integration, and constitutional pluralism. I shall consider them in turn in a brief review.

A.  State-centred Conceptions There are several state-centred conceptions out there, but they all share a com­ mon trait: the dogmatic preference for the state and its people as the predominant political community, and for constitutional law as the predominant structuring law of that community. European Union law is seen and actually allowed to exist only as part of that framework, in the terms and within the limits established by national law. The Union is not and cannot be an autonomous entity. It has not yet severed the umbilical cords that link it to the constitutional laws of the various Member States, the true ‘masters of the Treaty’. Its law is just a more effective kind of international law, but not fundamentally different from it. These theories recall the intergovernmental theories of political integration, such as those devel­ oped by Milward or Moravcsik, which also explain integration from the exclusive point of view of the state’s power and preferences.2 As a result of that predominant point of view, state-based constitutionalism, an essentially rational product of the modern mind, usually remains pre-modern when it looks beyond its boundaries – when its purely national perspective defines an identity by opposing itself to other identities. Beyond the state, state-centred constitutionalism becomes suffused with political theology, with notions such as state, people, identity, sovereignty and constituent power, which work as axioms: the unproven dogmatic foundations of the whole constitutional building and also the generally unconscious basis of a form of legal nationalism. Unconscious indeed and one could even say almost invisible: it is so deeply engrained in the legal cultures of the Member States that it usually passes unnoticed, at least among lawyers. The most radical state-centred conception is a sort of constitutional solipsism aimed at avoiding tackling the issue. This position considers European Union law 1  This section is partly based on my essay ‘An Area of Darkness: Three Conceptions of the Relationship between EU Law and State Constitutional Law’ in N Walker, S Tierney and J Shaw (eds), Europe’s Constitutional Mosaic (Oxford, Hart Publishing, 2011) 49–71. 2   See AS Milward, The European Rescue of the Nation-State 2nd edn (London, Routledge, 2000); A Moravcsik, The Choice For Europe: Social Purpose And State Power From Messina To Maastricht (Ithaca, Cornell University Press, 1998).

Legal Pluralism and Institutional Disobedience  251 as something intrinsically foreign and different. It would always move, as it were, on another plane. It could never be part of the same picture. It would be extraconstitutional or a-constitutional. As a result, the two could never meet or collide. Issues of Union law would then be characterised as issues of mere statutory law, to be resolved by ordinary courts, taking into account ordinary law. Constitutional courts could not possibly bother with them.3 There are more sophisticated and realistic versions of the state-centred approach, of course, normally in the form of a general acceptance of the suprem­ acy of Union law with a number of reservations and the possibility of a last resort intervention of a constitutional court if in an exceptional case there is a need to preserve the core values and principles of the state constitutional order. This is the case in which the Schmittian decision on the exception stands as the small shrine in which sovereignty is preserved, in general – but not always – without actual consequences for integration. The judgment of the German Constitutional Court on the Treaty of Lisbon,4 a sophisticated and rather convoluted restatement of the Maastricht-Urteil with some additional teeth and dubious practical consequences for the decision-making process of the Union, is the most recent example of this approach. The courageous reader of this piece of legal writing will certainly notice that in the world it represents everything is seen through the peculiar lens of German constitutional law, which becomes the beginning and the end of an axi­ omatic and ultimately, in spite of all its turns, very simple circular reasoning. The European Union and its law are reinterpreted and sometimes distorted through that curious lens. The state and its constituent power are presented as untouch­ able mythical figures. Their withering away would require an unthinkable dra­ matic act. There is no effort on the part of the German court to see itself as a participant, in the same way as German political institutions, in a larger institu­ tional and legal community that transforms itself gradually without any need for dramatic events. There is no hesitation on its part about what course to take and no consideration at all of the very negative consequences that its behaviour could have on all the other actors and systems concerned, not to speak of the common enterprise of integration. Its normative horizon really starts and ends with the German Constitution. The so-called ‘eternity clause’, the ultimate ground of the case law of the German Constitutional Court on European integration, is clearly suffused with political theology, and points to an unreconstructed legal eschatol­ ogy in which the historically grounded mistrust of democratic politics leads to a 3   That was, for a long time, the position of the Spanish Constitutional Court, which variously con­ sidered ‘infraconstitutional’ or ‘extraconstitutional’ the possible conflicts between Community law and internal law (see, eg, judgments 64/1991 and 45/1996). 4   Case 2 BvE 2/08, judgment of 30 June 2009, Lisbon, English translation http://www.bverfg.de/ entscheidungen/es20090630_2bve000208en.html. See also K Auel and J Baquero Cruz, Karlsruhe’s Europe (Paris, Notre Europe, 2010). The German court later softened its approach in its more deferen­ tial and pragmatic judgment in Case 2 BvR 2661/06, judgment of 6 July 2010, Honeywell, http://www. bundesverfassungsgericht.de/en/decisions/rs20100706_2bvr266106en.html limiting the scope of its ultra vires review of Union law to acts which are manifestly in breach of the limits of the powers of the Union and which lead to a structurally significant shift in the division of powers between the Union and the Member States.

252  Julio Baquero Cruz blind trust of the politics of law carried out by the very German court, with very negative potential consequences for the European project. Radical and less radical, sophisticated and less sophisticated, all state-centred conceptions present a common impediment: their incapacity and sometimes out­ right unwillingness to ‘think European’, to embrace integration, to adopt an internal point of view with regard to the Union and its law, and to work towards the consolidation of a common European legal culture. They tend to see the European Union and its law from a purely external perspective, as the mere sum of the Member States, as a political and legal complement to them. The nationstate is thus taken as the natural political community which strives to maintain itself for an indefinite and ideally eternal time, sometimes and only in a number of areas with the help of integration, and to achieve a just and stable social, eco­ nomic and political order within its borders. Only the democracy of the nationstate is natural and truly democratic: that of the Union lacks the necessary preconditions of democracy, and can only and will always be an artificial and imperfect ersatz. In the same way, legal things beyond the state also lack that natu­ ral basis: they are artificial constructs to be regarded with diffidence and to be narrowly construed. As a result, only the Constitution of the nation-state is a true Constitution and can reign supreme. In the last analysis and as a matter of principle, European Union law must always be subordinate to it, even if that sub­ ordination only becomes visible in the unlikely but crucial exceptional case.

B.  The Law of Integration or Constitutionalism Beyond the State The law of integration, in contrast, strives hard to ‘think European’ and to lay down the foundations of a European legal culture that can travel across borders and become truly common ground as a technique, as a language, and as a carrier of shared values and principles.5 According to its logic, the progressive reception of European Union law in the national legal cultures would lead to the establish­ ment of an ordered relationship along the lines devised by Union law itself, in an automatic process that can be seen as the legal counterpart of the spillover of neo­ functional theories. The acceptance of its supremacy would be part of the European path of civilisation. Inclusion in a larger, more comprehensive, political community would imply the acceptance of its law and an enlargement of the hori­ zon of national legal cultures. Union law and its institutions would have to be in charge of policing the borders between the Union legal system and national legal orders, for that function could not possibly be decentralised without putting inte­ gration into danger. Those institutions should be trusted to do that, since they are as democratic as they can be in a system of integration and not fundamentally undemocratic or less democratic than the institutions of the Member States – 5   See P Pescatore, The Law of Integration (Leiden, Sijthoff, 1974); M Cappelletti, M Seccombe and JHH Weiler, Integration through Law: Europe and the American Federal Experience, five volumes (Berlin and New York, W de Gruyter, 1985–88).

Legal Pluralism and Institutional Disobedience  253 with which they are in close contact through the interface between systems that is the Council. At the ideal end of the process of legal integration, there would be a harmonious coming together to form a new composite legal system, based on common principles and on a large consensus about the way in which legal orders should interact with each other, especially in case of conflict. The law of integration is thus a modern enterprise through and through, and is deeply imbued with the usual trust the modern man puts in reason and rational­ ity. In many ways, it is a project that takes to a continental plane the modern project of constitutionalism, which was originally circumscribed to the nationstate and had and still has a pre-modern attitude towards anything that is beyond the state. Thus, it was clear that sooner or later it would end up adopting the vocabulary of constitutionalism and clashing with the basic tents of state-based constitutionalism. The ambitious project of the law of integration has obviously encountered a number of difficulties, and some of them may well have become endemic. The law of integration has always had a problematic and ambivalent mode of being. Torn between the domestic, the federal and the international, it has always seemed much more effective than most international law and yet somewhat less effective than most domestic and federal law. Even today, its authority does not seem to be as solidly established as that of other more traditional legal systems. It can never be taken for granted. It is always changing, never stable, sometimes improving, sometimes decaying – inevitably linked to the ups and downs of political integra­ tion. Union law itself and its doctrinal elaboration may be well developed in the­ ory and in the case law, but social reality and the behaviour of legal actors in the Member States do not always correspond to it. The more Union law develops in words and structure, the more the social and legal reality of some Member States seems to resist that development, and the wider becomes the gap between the ideal order of things expressed in Union law and stubborn things as they are and tend to remain. That resistance may sometimes be rebellious but most of the time it is due to ignorance and habit. And even though the law of European integration may be seen as a success in terms of acceptance and effectiveness, especially when compared to the law of other international and regional organisations, Union law still presents many shortcomings in its actual application: it is a rather imperfect system of law. These difficulties have to do with legal culture rather than with legal principles and techniques. As I have already argued, national legal cultures are deeply engrained. Like the state, they become habits and tend to be perceived as some­ thing natural – the conspicuous ‘law of the land’. Thus, most national legal actors still live in a self-contained word in which external objects, including European Union law, are seen with suspicion if they are seen at all. Quite often Union law is not applied when and as it should. This problem may be graver in peripheral countries, in countries that still have a short experience as Member States of the Union, or in countries with judicial systems that do not perform well in general, also with regard to state law. To a variety of extents, however, it is present in all of

254  Julio Baquero Cruz them, even in the founding Member States that have well-functioning judicial and administrative institutions, and it does not take much to find cases in which Union law is blatantly ignored. Hence if one reads a handbook on European Union law one will not have a realistic view of it, but a highly idealised view of the legal system that is only part of the story. To have the complete picture one needs to examine carefully its practical reception in the legal orders of the Member States. And there the picture usually becomes a bit blurred and sometimes very blurred if not bleak. Confronted with these difficulties, Union law has developed a number of strat­ egies of ‘internalisation’ and ‘incorporation’, which render it flexible and plural from within and more palatable to the constitutional orders of the Member States.6 Such strategies can only work, however, to the extent that the guardians of those constitutional orders are willing to engage in a productive dialogue with the institutions of integration. Before unilateral and direct threats to supremacy, however, Union law has only been able to repeat the predictable and ineffectual response: supremacy! Like most legal systems, European Union law does not have the legal instruments needed to cope with extreme cases of disagreement, other than ignoring them. But by ignoring them it runs the risk of drifting further apart from reality, of becoming a voice in the wilderness.

C.  Legal or Constitutional Pluralism Following a third way of sorts, some have tried to bridge the gap between Statecentred conceptions and the law of integration through dynamic and bidirec­ tional models.7 They are the legal counterpart of third-generation political theories of integration such as multilevel governance.8 They have a clearly postmodern flavour in their rejection of hierarchy and their insistence on complexity, openness, discourse and the like. They take into account the shifts and evolutions through time of the systems of law at play, seeing them as reciprocally engaged in an open-ended process of informal dialogue. They have come to be known as legal or constitutional pluralism. There are several pluralisms out there, more or less radical in the consequences they extract from their approach, but they are all characterised by a number of common traits. First, the pluralists would like to stand beyond the European point of view and also beyond any particular national perspective. They would 6   On these strategies see L Azoulai, ‘The Future Constitutional Role of the European Court of Justice’, in J Baquero Cruz and C Closa (eds), European Integration from Rome to Berlin, 1957–2007: History, Law and Politics (Brussels, PIE–Peter Lang, 2009) 229–47. 7   See M Poiares Maduro, ‘Contrapunctual Law: Europe’s Constitutional Pluralism in Action’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003, 501); N Walker, ‘The Idea of Constitutional Pluralism’ (2002) 65 Modern Law Review 317; M Kumm, ‘The Jurisprudence of Constitutional Conflict: Constitutional Supremacy in Europe before and after the Constitutional Treaty’ (2005) 11 European Law Journal 262. 8  See eg L Hooghe and G Marks, Multi-level Governance and European Integration (Lanham, Rowman & Littlefield, 2001).

Legal Pluralism and Institutional Disobedience  255 like to see things from far away, and seek to adopt a neutral stance. Second, this point of departure leads them to the recognition of the normative claims of both sides and to the search for a compromise. Third, such a compromise is sought on the basis of what is common, and on the recognition of what is different. This may explain the usual emphasis of pluralists on the existence of ‘common constitu­ tional traditions’ to be discovered through deliberation. Hence, also, the need to protect what is not common, the specific traits of national constitutional identi­ ties, for plurality is perceived as a value in itself, often quite independently from the merits of its underlying substance. Finally, this leads them to prefer to leave forever open the issue of supremacy, an openness that becomes embedded in the compromise. Thus, the supremacy or primacy of Union law would generally be recognised by national constitutional orders, but unilateral exceptions would exist, through judicial or political means, whenever they are justified by the need to preserve a particular constitutional identity and hence the pluralism of the sys­ tem as a whole. Like the two other accounts, pluralism is not without problems, especially when it is presented as a normative ideal and not just as a description of the relationship between European Union law and state constitutional law.9 As a description of the current state of affairs, it may be plausible, at least in some Member States. But then constitutional pluralism could just be seen as a second best solution for a slow transitional period towards legal orthodoxy, something like Wittgenstein’s famous ladder, to be thrown away once one has climbed up on it and gained the height needed to see the world correctly.10 Then, in other words, it would not be the sort of dialectic Aufhebung that some want it to be, but a more modest and pragmatic attempt to bridge the seemingly intractable gap that lies between state-centred conceptions and the law of integration. As a normative ideal, like a ladder where some would like to stay and an Aufhebung of sorts, however, pluralism might be somewhat uncomfortable and less appealing when its possible consequences are considered. First, the neutrality it purports to adopt is difficult if not impossible in the world of law. Confronted with actual conflicts that call for a solution, neutrality is unsustainable. With their contempt for hierarchy, the pluralists actually seem to favour practical solutions which are closer to the exceptional reservations of sover­ eignty favoured by State-based constitutional theories than they are to the position of the law of integration. So far we have not seen any convincing proposal of how a non-hierarchical legal order would work in practice while remaining a legal order. For if we want it to remain an order, a system of law needs at least a measure of normative hierarchy that necessarily implies a measure of institutional hierarchy. In most cases, the questioning of the supreme character of European Union law only leads to the replacement, in a number of exceptional cases, of one hierarchy by another. Second, the compromise legal pluralism tries to achieve may be positive 9   For a more detailed critique, see my article ‘The Legacy of the Maastricht-Urteil and the Pluralist Movement’ (2008) 14 European Law Journal 389, 412–18. 10   L Wittgenstein, Tractatus Logico-Philosophicus 2nd edn, trans by D Pears and B McGuinness (London, Routledge, 2001), proposition 6.54.

256  Julio Baquero Cruz in political terms but it would sometimes sacrifice the position of individuals, who would not know what institution will ultimately adjudicate on their rights and obligations and on the basis of what rules. This is problematic, for the position of individuals should be at the centre of our reflection. This deserves emphasis, since in its quest for institutional compromises pluralism tends to marginalise the posi­ tion of individuals, which the legal systems involved and their institutions are sup­ posed to serve. Third, the tension between the search for common constitutional traditions and the need to preserve constitutional diversity may fatally undermine deliberation, if the emphasis on specificity, difference and identity – all of them easily constructible and open to political manipulation – leads to irreconcilable positions. Finally, pluralists tend to consider that the absence of a final decision on the issue of supremacy is an essential part of the structure of European Union law, even its very foundation, ignoring the fact that such ‘pluralistic’ episodes are rela­ tively rare in the actual practice of law in Europe. *   *   * It is that fact that made me think that the phenomenon sometimes described as legal or constitutional pluralism could not be seen as the normal state and even less as the very foundation of the relationship between European Union law and state constitutional law. My intuition was always the opposite: that if we want the institutional and legal machine to function at all, its normal state should be supremacy and co-operation through the preliminary rulings procedure; that express disagreement should remain exceptional, an extraordinary escape valve from the system, not part of its essence. At present, however, the European Union system cannot cope with or make sense of express disagreement. It has no space for it. For some time I just thought that there was no solution to this problem, that the two first positions were incommensurable, that none of them was fully accurate and acceptable, that plu­ ralism was vague, confusing and unworkable, and that we were just stuck with what seemed to be not really a substantive problem but a problem related to the incompatible points of view adopted by different actors in the discourse around supremacy. A problem of legal culture, therefore, that could only be resolved through the difficult and slow evolution of the various legal cultures involved. Only later on, and quite coincidentally, did I start to think about the issue from the angle of civil disobedience. In 2008 I wrote that the unilateral selective ‘disapplication’ of EU law in very exceptional cases could at most be justified as a – preferably unwritten – final right of resistance against EU measures that would be manifestly inimical to the basic values and principles of a national consti­ tutional system.11

  Baquero Cruz, n 9 at 416.

11

Legal Pluralism and Institutional Disobedience  257 It seemed to me dangerous to transform this exceptional right of resistance, which may indeed be healthy in extreme cases, into an essential element of the system.12 After that, in a discussion with three leading exponents of the pluralist movement – Mattias Kumm, Miguel Maduro and Neil Walker – I referred to this possibility as ‘political disobedience’, which is a form of ‘resistance’, and I expressed doubts as to whether the argument from civil disobedience could be extrapolated from individuals to institutions, and argued that we needed to think a bit more about that.13 That is what I set out to do in what follows.

III.  Basic Traits of Civil Disobedience Institutional disobedience can be seen as a special case of civil disobedience car­ ried out by institutions. Thus, in order to define institutional disobedience we must take into account the basic traits of civil disobedience. In this starting point of the reflection little new will be said. My purpose is not to break new ground on the theory of civil disobedience, which has remained more or less stable since the 1960s and 1970s, the period in which it received most attention in both theory and practice, but to think about its institutional counterpart. What is civil disobedience? Following Bedau and Rawls, it can be defined as ‘a public, nonviolent, conscientious yet political act contrary to law usually done with the aim of bringing about a change in the law or policies of government’.14 This can be done, Bedau notes, directly, by breaching the rule against which the disobedient protests, or indirectly, as Thoreau did, by breaching an unrelated rule to protest against the state. It is, then, a kind of performance, an act through which the actor does or does not do something in order to make a political statement which generally has a larger significance than the ‘literal’ and purely legal meaning of the act itself. That is why acts of civil disobedience are often rich in rhetorical dimensions. Civil disobedience belongs to the ‘cluster concept’ of ‘conscientious illegality’, which would cover the various kinds of unlawful political strategies that lie between revolution or rebellion, on one side of the spectrum, to ordinary obedi­ ence to the law, on the other.15 As with most political concepts, its definition is mostly a matter of distinguishing it from other forms of political action. A dis­ tinctive feature of civil disobedience is that it must be done in public and not in   Ibid.   M Avbelj and J Komárek (eds), ‘Four Visions of Constitutional Pluralism – Symposium Transcript’ (2008) 2 European Journal of Legal Studies 325, 346 and 360–61. 14   J Rawls, A Theory of Justice (Oxford, Oxford University Press, 1999), section 55, reproduced in HA Bedau (ed), Civil Disobedience in Focus (London and New York, Routledge, 1991) 104. For the definition, Rawls follows HA Bedau, ‘On Civil Disobedience’ (1961) 58 Journal of Philosophy 653. 15   B Smart, ‘Defining Civil Disobedience’ in Bedau (ed), n 14 at 199. 12 13

258  Julio Baquero Cruz secret. Covert disobedience is generally equated with conscientious objection, the aim of which is not primarily to bring about a change in an unjust law, but to escape its individual application in the case in hand. Civil disobedience is also dif­ ferent from mere criminal conduct: the illegal conduct of the civil disobedient sub­ ject is based on conscience, ‘motivated by respect for some moral principle which [. . .] was violated by the law and which deserved greater respect than the law itself’.16 Civil disobedience is also different from but akin to obedience while voicing disagreement with the law. It is different as well from test cases and from lawful protest, which are accepted by the legal order and protected by fundamental rights (rights of defence, freedom of speech, freedom of association and right to pro­ test). Sometimes it is difficult to distinguish between these forms of individual attitude towards laws that are considered to be unjust. Since much depends on individual intention, as distinct from mere criminal behaviour, it is not always easy to decide to which category a given action belongs. That decision is import­ ant insofar as we may want to use the notion of civil disobedience as a normative concept, to see it as a form of unlawful but legitimate political action which is different from political acts which are both unlawful and illegitimate. As already suggested, civil disobedience can also be distinguished from rebellion and revolution, which would be the result of a systematic and persistent disobedi­ ence aimed not at promoting a punctual change of the law in a generally just sys­ tem but at changing radically or overturning the whole legal and political systems. According to Rawls, civil disobedience ‘expresses disobedience to law within the limits of fidelity to law’.17 He argues that ‘a theory of civil disobedience [. . .] is designed only for the special case of a nearly just society, one that is well-ordered for the most part but in which some serious violations of justice nevertheless do occur’. In such a nearly just society, where citizens ‘recognise and accept the legit­ imacy of the constitution’, the problem of civil disobedience would only arise when there is a severe conflict of duties related to the limits of majority rule: At what point does the duty to comply with the laws enacted by a legislative majority (or with executive acts supported by such a majority) cease to be binding in view of the right to defend one’s liberties and the duty to oppose injustice?18

From this perspective, civil disobedience, together with judicial review, could help in extreme cases in which the ordinary political process was of no avail to preserve a just political system, as ‘a final device to maintain the stability of a just constitution’.19 In the same vein, we must also distinguish civil disobedience from general nonco-operation. In the theory and practice of Gandhi, non-co-operation is indeed directed at opposing and overthrowing an ‘evil system’.20 Civil disobedience is   HA Bedau, ‘Introduction’, in Bedau (ed), n 14 at 9.   Rawls, n 14 at 106.   Ibid 103. 19   Ibid 115. 20   MK Gandhi, Nonviolent Resistance (New York, Schocken Books, 1961) cited in V Haksar, Civil Disobedience, Threats and Offers: Gandhi and Rawls (Delhi, Oxford University Press, 1986). 16 17 18

Legal Pluralism and Institutional Disobedience  259 directed at a particular policy or law of a ‘nearly just’ political community. General non-co-operation is more deleterious to the system. It does not accept it as a nearly just system. Besides its public character, an important condition of civil disobedience is that individuals who intend to have recourse to it must accept the negative ‘lawful consequences of their illegal conduct’, thus showing ‘that they respect the rule of law generally’.21 Rawls remarks that protesters and authorities share some moral principles, while rebellious or revolutionary actors do not share them and would not accept the punishment inflicted for their disobedience. They tend to deny the system as a whole, not just a particular part of it. Another important requirement of civil disobedience is that ‘not just any moral principle will suffice to justify illegal conduct’ in any circumstance. ‘For surely some acts of civil disobedience are unjustified, and the reason is that sound moral principles do not permit, much less require, illegal conduct of certain kinds in certain kinds of circumstances.’22 Rawls takes up and develops that argument, considering that civil disobedience can thus be seen as a plea for reconsideration . . . a political act not only in the sense that it is addressed to the majority that holds political power, but also because it is an act guided and justified by political principles, that is, by the principles of justice which regulated the constitution and social institutions generally.23

Therefore, it could not be justified if it were ‘grounded solely on group or selfinterest’. It needs to be based on ‘the commonly shared conception of justice that underlies the political order’.24 This leads to the question, also essential for institutional disobedience, of whether civil disobedience can ever be legitimate in a developed and well-func­ tioning constitutional democracy. There have been strong arguments against civil disobedience in the context of democratic societies, or aimed at restricting it, and they cannot be ignored. One author has contended that the ‘institution of judicial review, in which the acts of governmental authorities are tested in the light of the higher law of the Constitution, provides for a kind of tamed or civilised ‘civil dis­ obedience’. One of the consequences of this institution is ‘to divert disobedience and even revolution into the channel of the law’.25 Indeed, if civil disobedience is, as John Rawls would have it, about the limits of majority rule, that clearly overlaps with the function of judicial review, which is the formal mechanism designed to ensure that majority rule does not overstep the constitutional preconditions of democracy, endangering the political system and democracy itself. One could indeed argue that in such a context civil disobedience can never be justified: it would amount to mere disobedience and beget just more and more disobedience,   Bedau, n 14 at 8.   Ibid 10. 23   Rawls, n 14 at 105–06. 24   Rawls, n 14 at 105–06. 25   HJ Storing, ‘The Case Against Civil Disobedience’ in Bedau (ed), n 14 at 87. 21 22

260  Julio Baquero Cruz endangering law and civil society.26 Even in the hands of those who are not repre­ sented in the political process, civil disobedience would at most be ‘a feeble instrument’.27 Lawful political action would always be preferable, since it would be equally effective without undermining respect for the law. This is very much the position of Joseph Raz, of particular interest for institu­ tional disobedience and for the case of the European Union. For him, in a liberal state ‘there can be no right to civil disobedience which derives from a general right to political participation. One’s right to political activity is, by hypothesis, ade­ quately protected by law. It can never justify breaking it.’28 According to Raz, the denial of a right to civil disobedience is not necessarily negative or a bad thing for pluralism. He contends that his argument does not disregard ‘the desirability of encouraging pluralism in the society’: one cannot just suppose that pluralism must lead to dissent and disobedience. It will do so if the law does not allow for pluralistic forms of life to flourish. If the law encourages and respects pluralism it need not lead to dissent from law. It can find adequate expression within it.29

His conclusion is telling: civil disobedience is an exceptional political action. It is exceptional, in liberal states, in being one beyond the bounds of toleration, beyond the general right to political action . . . one type of political action to which one has no right.30

Raz’s argument, in my view, could justify limiting the permissible or legitimate forms of civil disobedience, but it is not sufficient to rule out completely civil dis­ obedience as a form of legitimate and permissible political action. In some extreme cases, even in the framework of generally just and well-functioning con­ stitutional democracies, recourse to civil disobedience may be justified and indeed useful. And those constitutional democracies may have good reasons to accept it as an exceptional element of dynamism in their political systems. History, which is the site of experimental politics, proves abundantly that sometimes civil disobe­ dience has been the only way to attract serious attention over an essential problem and to prompt legal and political changes which seemed impossible and almost unthinkable at the time and which we now cherish. In any event, civil disobedience should only be used as a last resort: We may suppose that the normal appeals to the political majority have already been made in good faith and that they have failed. The legal means of redress have proved of no avail . . . Since civil disobedience is a last resort, we should be sure that it is necessary.31   Ibid 94.   Ibid 99. 28   J Raz, The Authority of Law (Oxford, Oxford University Press, 1979) 266–75 reproduced in Bedau (ed), n 14 at 166. 29   Ibid 168. 30   Ibid 168–69. 31   Rawls, n 14 at 110. 26 27

Legal Pluralism and Institutional Disobedience  261 It cannot be a first resort or the only resort: one should not take things to the extreme from the very beginning. Since it is an extreme form of political action, actors must decide to engage in it only after a serious consideration of its possible consequences and with much prudence. For Rawls, ‘there is a limit on the extent to which civil disobedience can be engaged in without leading to a breakdown in the respect for law and the constitution, thereby setting in motion consequences unfortunate to all’.32 This would specially be the case in political communities in which several minorities could legitimately have recourse to civil disobedience: in such a case the combined effect of the action of each could be catastrophic for the political community at large. Rawls’ words also have important implications for the European Union: This kind of case is also instructive in showing that the exercise of the right to dissent, like the exercise of rights generally, is sometimes limited by others having the very same right. Everyone’s exercising this right would have deleterious consequences for all, and some equitable plan is called for.33

To conclude this section, I would like to reflect about what seems to me to be the most remarkable thing about the definition of civil disobedience: how theo­ retical works have contributed to transform the notion into a normative and almost regulatory concept. Some may actually think that there is something odd and even contradictory in wanting to frame civil disobedience with a sort of code of conduct that would serve to decide when and how it can be used legitimately and when and how it should not be used. It is like wanting to include within the world of law and rules something that is inherently unruly and outside the scope of the law – like wanting to tame a wild animal. Should it not rather stay outside of the law? Should we not rather avoid talking about it, as if it did not exist, as if the law should not refer to such extraordinary things? It is not lawyers, however, that have tried to regulate civil disobedience. The theory of civil disobedience has been mainly developed in political theory as an attempt to frame and civilise the exercise of this potentially boundless energy that exists within political systems. Rightly so, perhaps, for otherwise it would not be possible to distinguish legitimate disobedience from other forms of transgression, and in drawing those distinguishing lines what they were doing is very similar to regulating the phenomenon. Also, perhaps, because they perceived, like Rawls, that even democratic political systems could sometimes need this exceptional mechanism to avoid stagnation and the entrenchment of grave injustices. And finally, in a more unconscious way, as an attempt to extend the realm of the law to cover all sorts of action, even those that at first sight seem to have nothing to do with it, to be its very negation. There is, indeed, an existential link between the law and the violation of the law. Without the second, the first cannot really be seen, it becomes invisible, it cannot show itself as law, with all its force and effectiveness. To establish a limit,   Ibid 111.   Ibid 112.

32 33

262  Julio Baquero Cruz and that’s what law generally is about, always means to open the way to its trans­ gression, and transgression is potentially boundless. A single act of transgression, for example an act of civil disobedience, is perhaps almost nothing from the point of view of the effectiveness of a rule which is respected in all other cases, but it represents a tendency that, if it became general, would be the end of law. That is the misgiving of Socrates in Crito: that one violation, even of an unfair rule, potentially and implicitly includes all the possible violations of all rules, fair and unfair, and prefigures the end of the polity. That’s indeed the reason that leads him to drink the hemlock instead of escaping and violating the law, for ‘who would care for a city without laws?’34 As in the case of total obedience, the law would also become invisible, this time on the opposite side of the spectrum. Of course we need not be as strict as Socrates, but near total obedience we may trace a line of some situations in which we may want punctual disobedience for good reasons that may be healthy for the political system as a whole.

IV.  From Civil Disobedience to Institutional Disobedience The idea of transposing the theory of civil disobedience to the institutional arena may seem quite daring to some. Institutions are creatures of the political system and of the legal order to which they belong. They are supposed to respect and enforce the framework to which they owe not only their legitimacy but also their very existence. In other words, they are supposed to work for it and not against it. They are meant to behave coherently together and not against each other. Imagine that one or the other institution of a polity started to behave in contradiction with its mandate and to disobey the commands which it is supposed to follow. Consider, for example, the case of an administrative body refusing to apply a law which it considers unjust, or that of a court deciding quite simply to ignore a law which it considers unconstitutional in spite of the fact that the constitutional court has declared it constitutional. These are strange situations that could create chaos, uncertainty, disorder and fragmentation. These are situations that we may want to avoid. There are, therefore, strong arguments against the very possibility of institutional disobedience, even if one is willing to accept the strategy of civil disobedience in extreme cases. A different and somewhat cynical argument would be to argue that, unlike individuals, institutions do not need to disobey to voice their disagreement openly. They can and indeed do voice such disagreement through interpretation, by adopting decisions which are formally compatible with and loyal to the legal order but which implicitly challenge the predominant legal position. Thus, an  Plato, Crito, 54a, 4–5.

34

Legal Pluralism and Institutional Disobedience  263 administrative body or a court could adopt a new interpretation which challenges the position of other courts or of the legislator. If it can do that through lawful avenues, why would it want to follow the more risky and unlawful path of institu­ tional disobedience? Critics would argue, in sum, that institutions and individuals are in a very dif­ ferent position with regard to the political and legal systems. Individuals often find themselves in a position of weakness, of subjection: they are subjects. Disobedience may sometimes be their last desperate resort. Institutions, in con­ trast, are often in a privileged position: they are vested with special responsibilities and the power to decide. It is that power and those responsibilities that would bar the possibility of institutional disobedience. Against these objections, I should like to defend the idea that, like individuals, institutions may also sometimes, with some provisos and in very exceptional cir­ cumstances, carry out ‘a public, nonviolent, conscientious yet political act con­ trary to law with the aim of bringing about a change in the law or policies of government’. Institutions are, after all, composed of human beings endowed with a conscience. Those human beings should not necessarily be and in any event are not mechanical and blind executors of the law. Sometimes they may feel that for very important reasons they are bound to disobey the law to attract the attention of the public to an important issue. For these purposes, implicit disagreement through interpretation would clearly be insufficient. In view of the special powers and responsibilities of institutions, however, insti­ tutional disobedience should be even more exceptional than civil disobedience. If one takes into account the conditions of civil disobedience, it is clear that institu­ tions will find themselves much less often in situations in which they may legiti­ mately use that strategy. The institution and its members have to accept the likely consequences of their conduct, including financial consequences, disciplinary sanctions, legal and political responsibility, etc. A higher and very important moral principle and not just any principle should be at stake. It can only be used as a last resort, after having tried all the possible remedies offered by the political process and the legal system. Finally, and this is the point in which the exceptional character of institutional disobedience will become most clear, an analysis should be carried out between the costs of the action and its possible benefits. Unlike the actions by individuals, the overall costs of institutional disobedience would tend to be very high, which means that the interests on the other side of the balance have to be very important and weighty. In a well-functioning society, therefore, institutional disobedience is likely to be justified only rarely, if ever. There are polities in which institutional disobedience may be more easily accepted, at least as a matter of principle: federal and divided-power systems. In such systems, a number of institutions may have a double loyalty, and that may create tensions. In such a context and in some situations, an institution of a divided-power system may be tempted to obey the commands of one authority and disobey those of another for important reasons. From the point of view of one of the subsystems involved, this would be an exceptional case of institutional

264  Julio Baquero Cruz disobedience. From the point of view of the other subsystem at play, it would be a case of an exceptional and justified self-defence. In a federal system, institutional disobedience could thus arise in the form of a conflict of duties, which is after all the basic situation behind civil disobedience. The definition of the proper scope of legitimate institutional disobedience would then amount to finding the point after which the duty to comply with the laws of one system ceases to be binding because of the duty to defend the basic principles of the other system. In divided-power contexts, the limits and conditions of civil disobedience have a special character. First, the act of disobedience is likely to attract no legal sanc­ tion or responsibility, but merely political consequences. If there is a sanction, it will probably be imposed on the political subsystem as a whole and not on the particular disobedient institution of that subsystem. Thus, before having recourse to it the institution in question should take into account the potential externalities of that course of action both for the subsystem and its institutions, for the other subsystems, and for the divided-power system as a whole. Second, the higher principle at stake could be either a principle of the divided-power system or a principle of the subsystem which is not shared by all. The first case is of course clearer, in as much as disobedience can be seen as a corrective action for a system which is drifting away from its correct path. The second, based on a particularity of the subsystem, may not be understood by those concerned in the other subsys­ tems and in the common system, and will probably attract no reconsideration on their part: it will just amount to disobedience, with all its costs but no beneficial effects. Third, divided-power systems normally have various procedural mechan­ isms, legal and political, formal and informal, to mediate and resolve these tensions. These mechanisms should be tried in good faith before turning to insti­ tutional disobedience, which must remain a truly last resort. Finally, the analysis of costs and benefits which will finally dictate the course of action to be followed should take into account the likely consequences of the disobedience for the pol­ ity as a whole and for its other parts, not only for the disobedient subsystem alone. Account should be taken of the potential destructive effect of such action if it were followed by institutions in many or all of the subsystems. The institution con­ cerned should assess whether the damage to the principle it intends to protect is important enough to justify the damage inflicted, with its risks of systemic crisis, escalation, secession and disintegration, to the polity as a whole. This kind of assessment, of course, requires a change of perspective. The institution must see itself as part of the divided-power system and not just as an organ of the subsys­ tem. Its normative and institutional horizon must be enlarged. A particularistic horizon in which the interests and concerns of the common system are discounted could lead to irresponsible destructive behaviour that has not properly taken into account all the stakes involved. In summary, institutional disobedience can be seen as a possible but risky political voice that should remain exceptional and mostly silent. Used scantly, properly and intelligently, it can be a stabilising device, a sort of escape valve in complex systems when they are confronted with extreme situations in which the

Legal Pluralism and Institutional Disobedience  265 usual channels of change are blocked or of no avail. Used negligently, however, it can damage the very foundations of the polity. The particular configuration of institutional disobedience in divided-power or federal systems, with their prob­ lems of double loyalty, is of especial interest to the concrete situations that may arise in the European Union, to which I now return.

V.  Institutional Disobedience in the European Union What national constitutional courts or courts that carry out constitutional review are doing when they voice actual or potential reservations to the absolute suprem­ acy of European Union law is, in my view, closer to an attempt at institutional disobedience than it is to a defence of state-centred constitutionalism or to a vir­ tuous open-ended pluralistic dialogue. The same would apply to comparable action on the part of the political institutions of the Member States, and also to popular protest through the institutional mechanism of referendums. These insti­ tutions seem to be saying: if there is a higher principle in danger we may want to reserve ourselves the possibility of disobeying Union law. Should the case arise, we will not be doing it because we want to overturn the European Union or to abandon it. We actually see the Union as a generally just system. We would be doing it as a plea for reconsideration in a particular case. It would indeed be a public, non-violent and conscientious breach of the common law with the aim of bringing about a change in the law or policy of the Union. The higher moral rules justifying such a departure could be found in the quintessence of the common constitutional law. The state would be ready to accept the consequences of its breach within the sphere of Union law (for example, liability, an infringement action, a judgment against the state, a fine, etc). And the European Union legal order and its institutions could be convinced by the plea for reconsideration, changing the law, or could take the argument into account and reject it with good reasons. Thus, the idea of institutional disobedience could complement and improve the traditional approach of the law of integration to the relationship between European Union law and state constitutional law. It could provide a way to think about, frame, and cope with exceptional cases of disagreement. As in the other contexts, in the European Union such a conduct can never be seen as the normal state of law, politics and the interaction between the two. It is indeed highly anomalous and risky. It is also a highly ambiguous course of action, bordering with punctual exit and disloyalty, at the limits of acceptable political and legal discourse. But sometimes, in very extreme cases, it may be a useful way to foster debate, to lead to positive changes and to unveil certain deficits. The normative framework of civil and institutional disobedience imposes a number of limits on the occasions in which to use it and the way to do so. An institution of a Member State that considers following that course of action should not do it lightly.

266  Julio Baquero Cruz First, a higher principle should be in grave, actual and imminent danger. This principle should in principle be common to the systems concerned. If it is a prin­ ciple peculiar to the legal order of the Member State concerned, the disobedient institution should give convincing reasons explaining why it is a higher principle that deserves the respect of all. Second, the institution has to see whether the available legal and/or political procedures (for example, preliminary references, including a preliminary refer­ ence asking the European Court of Justice to reconsider its position,35 and politi­ cal negotiations) to express dissent and to prompt change have been used and with what result. In other words, institutional disobedience has to be a last resort and not a first and only resort, as is sometimes the case in the Union. For example, in the European arrest warrant case,36 the German Constitutional Court annulled the German law transposing the framework decision without referring a question to the European Court of Justice, as suggested by the German government. The annulment was, supposedly, limited to issues falling within the margin of appre­ ciation of German authorities when transposing the framework decision, but such a finding was only possible after a complex and dubious interpretation of the framework decision itself, an issue that should have been referred to the European Court. In other situations, such as the prior constitutional review of a Treaty amendment, in which a preliminary reference is not possible because the new Treaty is not yet part of European Union law, national institutions have to behave with special prudence. In such cases, it might be more prudent to reserve consti­ tutional review for concrete cases arising once the new Treaty is in force, and limit themselves to flagrant abstract problems and violations which are quite unlikely to exist. This was done by the Spanish Constitutional Court with regard to funda­ mental rights protection in its declaration on the Treaty establishing a Constitution for Europe.37 In contrast, in its judgment on the Treaty of Lisbon,38 the German Constitutional Court took the opposite route, putting forward very concrete pro­ visos and conditions for the constitutionality of that Treaty, instead of reserving its judgment for concrete cases in which it would have the opportunity to request the guidance of the European Court of Justice. If every court exercising constitu­ tional review in the European Union engaged in such a procrustean exercise before a Treaty amendment came into force, I wonder whether the European Union could still function in practice, subject to such strain. Third, the institution concerned has to weigh the pros and cons of the route of institutional disobedience in the particular case, and be ready to accept all the political and legal consequences of its disobedience both for itself, for the Member State, for the other Member States and for the Union as a whole. We would also 35   A preliminary ruling does not have to be the end of the judicial dialogue in the framework of Art 234 EC (now Art 267 TFEU): Case 14/86 Pretore di Salò [1987] ECR 2545 and Case 69/85 Wünsche [1986] ECR 947. A new submission cannot be used as an appeal against a preliminary ruling, but it can ask for a clarification or even reconsideration in the light of new facts or arguments. 36   Case 2 BvR 2236/04, European Arrest Warrant [2005] BVerfGE 113, 273. 37   Spanish Constitutional Court, Declaration of 13 December 2004, 1/2004 [2005] 3 BOE 5. 38   See above n 4.

Legal Pluralism and Institutional Disobedience  267 have to consider certain limits based on political and legal prudence. Since the institutions involved consider the system to which they belong to be ‘nearly just’, maintain their general loyalty to it and do not intend to overthrow it but only to adjust it in one particular aspect which they deem unjust, they should avoid endangering or destroying it. Now, the European Union is a fragile system in which there are just minorities, all of which can be disobedient at some point for objectively legitimate reasons. Indeed, each Member State is a minority and all the peoples in the Member States are in a minority position within the Union. In such a system, the parties concerned may well have to exercise the possibility of institu­ tional disobedience with even greater prudence. And finally, if it is done, it has to be done publicly and explicitly, and not by paying lip-service to European Union law. The cases of sheer silent ignorance of Union law would never qualify as institutional disobedience. Indeed, non-cooperation in the Union is more common than is thought, not because of an open opposition to a system that is considered to be unjust in whole or in part, but because of the mere ignorance and inertia of national legal actors. This form of passive resistance may be more damaging than institutional disobedience. It is, in any case, not a legitimate form of institutional resistance, but mere disloyal behav­ iour in a system of integration. *   *   * To summarise, I think the notion of institutional disobedience is a good descrip­ tion of what is going on when national institutions, including courts exercising constitutional review, resist in one way or another, actually or potentially, the supremacy of European Union law.39 I also think it is a fairly good guide for what should and should not be done in such difficult and rare circumstances. Finally, from the perspective of the emergence and consolidation of a shared European legal culture, the notion of institutional disobedience may be a useful tool to think and discuss about these problems, not only from the particular point of view of the law of integration, but from the three possible perspectives outlined above, which may not be as incommensurable as they sometimes seem.

39   It is also very much what the European Court of Justice itself did when it resisted the supremacy of international public law in the Kadi case, in view of the lack of protection of fundamental rights within the system of the United Nations (Case C-402/05 P Kadi and Al Barakaat v Council and Commission [2008] ECR I-6351).

12 Constitutional Disagreement in Europe and the Search for Pluralism GARETH DAVIES

I. Introduction This Chapter has two purposes. One is to suggest that constitutional pluralism is an empty idea. Where there are multiple sources of apparently constitutional law one always takes precedence and the other is then no longer constitutional. Dialogue may help legal the legal sources reconcile, but it does not change the normative hierarchy between them. The second purpose is to make a concrete proposal for embedding pluralist thinking within EU law. The proposal is in the spirit of Poiares Maduro’s suggestion that national courts should take account of EU interests in interpreting national law, and also in the spirit of Kumm’s suggestion that EU law should be self-policing.1 However, unlike Poiares Maduro it focuses on the need for a more pluralist approach within EU law, rather than national law, and unlike Kumm it focuses on the need to prevent EU law becoming a threat to national constitutions, rather than mechanisms for defusing conflict if things get that far. The two purposes are linked by a common perception: that the investment in constitutional pluralism by scholars has not brought satisfactory returns, yet pluralism is too attractive an idea to be abandoned in haste. The Chapter begins by outlining why constitutional pluralism is attractive to scholars, and why it is unconvincing. It provides a summary of the argument about constitutional pluralism. The Chapter then moves on to develop this in more detail. It first describes the disagreement between some national supreme courts and the Court of Justice about the relative status of national constitutional and EU law. It then considers the likely practical consequences of these disagreements, concluding that the greatest risk is not of a clash of grand constitutional principles, but rather just that EU law may spread to the point that it is seen as no longer constitutionally legitimate. The major need, therefore, in a Europe where 1   M Poiares Maduro, ‘Contrapunctual [sic] Law: Europe’s Constitutional Pluralism in Action’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003); M Kumm, ‘The Jurisprudence of Constitutional Conflict: Constitutional Supremacy in Europe before and after the Constitutional Treaty’ (2005) 11 European Law Journal 262.

270  Gareth Davies the Treaties co-exist alongside national Constitutions, is for a mechanism within EU law which prevents such law and policy spread. This would be a mechanism integrating pluralism within EU law. The suggestion is thus that pluralism of legal systems, or of Constitutions, leads us nowhere, but pluralism within a legal system – a pluralist legal system – is an option worth exploring.

II.  Constitutional Pluralism The initial attractiveness of constitutional pluralism is as a description of the apparently unstable, or unresolved, hierarchy between (certain) national constitutions and EU law.2 Since neither bows to the other, and each is supreme on its own terms, a description of the overall state of affairs in terms of pluralism seems more convincing than one which concedes to the claims of one side or the other.3 Yet the symmetry of the situation is illusory. National courts control the outcome of actual cases, and in most cases they still consider that their ultimate allegiance, in the event of conflict, is to national constitutions and national supreme courts.4 EU law may assert, but it lacks the means to enforce its assertions.5 National Constitutions are the superior authority in practice. Since one of the conventional attributes of a Constitution is that it is the highest source of law within its jurisdiction, EU law is hardly constitutional in most states. Its constitutional status is limited to the Court of Justice itself and the courts of those states, if there are any, that accept unreservedly the absolute supremacy of EU law. In these states, national Constitutions hardly deserve their name any more, and are now subordinate law. However, in the states that retain their own Constitutions as the highest legal document, it is EU law that is ultimately sub­ ordinate. To speak, then, of constitutional pluralism, as if the Treaties and the Constitutions of Member States were equal partners, is more deceptive than 2   See generally N MacCormick, ‘The Maastricht-Urteil: Sovereignty Now’ (1995) 1 European Law Journal 259; N Walker, ‘The Idea of Constitutional Pluralism’ (2002) 65 Modern Law Review 317; NW Barber, ‘Legal Pluralism and the European Union’ (2006) 12 European Law Journal 306; M Poiares Maduro, ‘Interpreting European Law: Judicial Adjudication in a Context of Constitutional Pluralism’ (2007) European Journal of Legal Studies 1; N Krisch, ‘The Case for Pluralism in Postnational Law’ LSE Law, Society and Economy Working Paper 12/2009, http://www.lse.ac.uk/collections/law/wps/wps1. htm. 3   MacCormick, n 2; N MacCormick, Questioning Sovereignty (Oxford, Oxford University Press, 1999). 4   See for overviews of the position of the various Member State supreme courts J Baquero Cruz, ‘The Legacy of the Maastricht-Urteil and the Pluralist Movement’ (2008) 14 European Law Journal 389, 397–403; M Kumm, ‘The Jurisprudence of Constitutional Conflict: Constitutional Supremacy in Europe before and after the Constitutional Treaty’ (2005) 11 European Law Journal 262, 263–6; A Albi, ‘Supremacy of EC Law in the New Member States’ (2007) 3 European Journal of Constitutional Law 25. There may be some exceptions. See B de Witte, ‘Do Not Mention the Word: Sovereignty in Two Europhile Countries’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003). 5   D Chalmers, ‘Judicial Preferences and the Community Legal Order’ (1997) 60 Modern Law Review 164.

Constitutional Disagreement in Europe  271 descriptive. It describes the rhetorical independence of the two legal orders, but ignores the fact that in actual situations – as sources of applicable law – there is in most states an unequivocal ultimate hierarchy. EU law is at best a Constitution without a jurisdiction: it may be the final authority in its own sphere, but that sphere is largely virtual rather than actual. This is uncomfortable for those reared on the optimistic assertions of the Court of Justice about the nature of EU law, and those who value the EU and fear the resurgence of nationalism and protectionism. Yet to keep faith with the Court is also uncomfortable. One can assert that national courts are simply ‘wrong’, but this is to refuse any concessions to reality, and risk rendering EU law ridiculous, even pitiful, while also ignoring the democratic problems of placing the less accepted European order above the more accepted national one. Alternatively, one can explain that the Court of Justice is entirely correct, but is talking about the EU legal order, not the national, so that there is no doctrinal conflict.6 This Jesuitical argument is coherent, but entirely unconvincing. It assumes that when the Court of Justice answers preliminary references it is not providing an instruction to national courts which it expects to be followed, but instead making abstract assertions about the nature of the EU legal order, without any concern about their effect or role in national courts. First, this is manifestly incorrect: the court does not conceive of itself in such a philosophical role. Second, it would be irresponsible if it did. The Treaties are goal-oriented, not theory-oriented. The court’s clear attributed function is not to wash its hands of the relationship between EU law and national law but to provide judgments which can and will be accepted by national courts, in order that EU law is effective in practice. Hence there is a great attractiveness in solutions which seem to offer a third way between blind allegiance to one order or another. Constitutional pluralism is presented as such a third way, and its apparent descriptive nature belies an invariable normative undertone. Constitutional pluralists think that neither national constitutions nor the EU Treaty should dominate the other, and both should exist side by side in a non-hierarchical way.7 Yet normative pluralism is as unattractive as descriptive pluralism is inadequate. There is no finessing the choice between legal chaos and hierarchy.8 When a court is faced with a conflict between the Treaties and the national Constitution it either chooses one consistently, in which case there is hierarchy, or it chooses arbitrarily, in which case there is chaos.9 Perhaps the basic error of constitutional pluralism is to forget that law is only meaningful, and only interesting, when it is applied. A non-applicable law is merely a castle of propositions. Pluralism can be used to describe the relationship 6   This is sometimes called ‘radical pluralism’. See N MacCormick, Questioning Sovereignty (Oxford, Oxford University Press, 1999) 117. 7   See above, n 2; P Kirchhof, ‘The Balance of Powers between National and European Institutions’ (1999) 5 European Law Journal 225. 8   Baquero Cruz, n 4. 9  See R Dworkin, Law’s Empire (Cambridge Mass, Harvard University Press, 1986) 179 et seq; P Eleftheriadis, ‘Pluralism and Integrity’ (2010) 23 Ratio Juris 365.

272  Gareth Davies between EU law and national Constitutions right up to the point at which these are applied to actual situations, whereupon it collapses and melts away. Nevertheless, pluralism remains an attractive idea, with its implications of tolerance and accommodation. Hence, even if it cannot be applied to the relationship between EU law and national law it is worth seeing if it can be applied within these systems. For example, as is expanded upon below, EU law could be made pluralist in that it could be constructed in a way that showed respect for and accommodation of the different national – and international – legal orders with which it must interact. It could display self-restraint. This Chapter makes a concrete proposal for how such pluralist self-restraint could be incorporated in EU law. Such internal EU pluralism would not affect the hierarchy between the national and the European. Internal EU doctrines cannot affect these inter-systemic issues, at least not in the short term. Perhaps in the long term they may have an indirect effect by rendering the ‘surrender’ of national systems to EU supremacy more or less likely. However, internal pluralism can make the chances that actual conflict arises between the legal orders much less likely. By increasing the acceptance of EU law within national orders it could also help EU law integrate into national law, so that it better achieves its goals and becomes a more present part of national legal life. Integrating the possibility of concession to national law into EU law may, paradoxically, increase the status and effectiveness of that EU law.10 It is thus, on the goal-oriented terms of the Treaties, a good interpretation of the law.

III.  The Disagreement between National Supreme Courts and the Court of Justice There is an apparent disagreement between several national supreme courts and the European Court of Justice about the status of EU law. The national courts, notably the Bundesverfassungsgericht, take the view that if EU law infringes aspects of the national Constitution it will no longer apply on their territory, at least to the extent of the infringement.11 The European Court of Justice, by contrast, claims that EU law is not subject to national Constitutions.12 In the event of a conflict, EU law should nevertheless be applied. Each court is correct according to its own legal order. Each is the authoritative interpreter of its own document, in the first case the national Constitution and in the second case the European Treaties. The national courts notably do not claim 10   D Ross Phelan, Revolt or Revolution: The Constitutional Boundaries of the European Community (Dublin, Round Hall/Sweet and Maxwell, 1997); G Martinico ‘Preliminary Reference and Constitutional Courts: Are you in the Mood for Dialogue?’ University of Tilburg TICOM Working Paper 2009/10, ssrn.com/abstract=1483664 on the attractions for national supreme courts of normative flexibility in EU law. 11   See n 4. 12   Case 11/70 Internationale Handelsgesellschaft [1970] ECR 1125.

Constitutional Disagreement in Europe  273 an interpretative jurisdiction over EU law. They do not claim that EU legislation or judgments which infringe the Constitution are invalid or wrongful. They simply say that the national Constitution prohibits the application of such acts. Since supreme courts are the authoritative interpreters of their Constitutions, this view must be accepted as correct as a matter of constitutional law – whether or not it is wise or desirable. Similarly, if the ECJ, as the body that is unquestionably the authoritative interpreter of the Treaties, claims that EU law is unaffected by national Constitutions, then this is also the case, as a matter of EU law. It may seem odd that both sides can be correct. However, this difference of perspective is not between two courts within a common legal order, who take different views on what that legal order entails. Rather, there is a clash of legal orders. Two sets of rules exist, and they say different things. Since each order determines its own interpretation, there is no doctrinal reason why the two orders, or their courts, cannot say contradictory things but on their own terms each be correct. This is nevertheless a disagreement, rather than just a difference, because both legal orders claim jurisdiction over the same circumstances. When the ECJ makes its claim of supremacy, it is not talking about EU law in the abstract or only insofar as it applies in the court in Luxembourg. It is making the very concrete normative statement that, as a matter of EU law, national judges faced with a conflict between the Constitution and EU law should prefer and apply the latter. This instruction – a particularly appropriate word since the claim was formulated in the context of the preliminary reference procedure – is directly contradictory to the instruction which national constitutional courts do, would and will provide if they are asked the same question by the same judge, as may well occur. In fact the ECJ is also making an implicit claim about national law. It is asking judges to transfer their allegiance from the Constitution to the Treaties. However, it is clear that a judge cannot switch from subjection to legal order A to subjection to legal order B unless he takes the view that legal order A permits this. Legal order A may do so ‘voluntarily’, on its own terms, in a surrender to the new order, or it may be the case that in some sense legal order A has become weakened or diminished so that it is no longer ‘capable’ of binding a judge against legal order B. However, a slave cannot choose his master – he must be released. The ECJ must be understood as inviting and instructing national judges to find that national law in fact permits them to transfer ultimate allegiance to EU law, and prefer it over the Constitution. This may be voluntary, or because national law has in fact been ‘conquered’ by EU law. But in any case, for the ECJ’s instruction to be more than posturing, it must be taking the view that national law is no longer capable of binding national judges against contrary EU law. If the ECJ does not take this view, then it is making statements which it knows to be irrelevant. It is answering a question about what a judge should do with something that it accepts that judge cannot do. This would not be a very constructive use of the reference procedure, nor would it be a very purposive or useful approach to EU law. The court is not paid to provide interpretations that are, as a matter of principle, unconnected to adjudicative reality. It is therefore quite

274  Gareth Davies implausible to think that the judgments on supremacy should be read as ‘you must obey EU law (although we are not saying anything about whether you can)’. Rather they are ‘you must and can obey EU law’.13 These are therefore attempts at legal colonialism, at absorbing the national legal order within the European, and at removing the national Constitution from its supreme position in the national courtroom – at de-constitutionalising it. They are, however – alas for the ECJ – attempts that have failed. National courts do not, at least in many states, regard the Constitution as having surrendered.14 Since the ECJ has chosen not to retreat from its position, a principled disagreement remains. This leaves us in a position well known in international law. Sometimes states, like people, make promises that they had no right to make – according to their own law – and which they cannot keep – because of their own courts. This does not make the promises any less binding in the eyes of international law.15 Nor does it make the national laws any less valid. It just means there is a problem. At the moment that problem in the European context is largely theoretical. However, if EU law crosses the lines that national constitutional courts draw, the problem will become very real.

IV.  The Chances of Concrete Conflict If EU law does not in fact conflict with national Constitutions, disagreements over supremacy become considerably less urgent. Until now it has only conflicted with one or two, and then only in fairly marginal and easily resolved ways.16 However, there are a number of reasons to fear that this happy situation may be temporary. First, EU law is spreading into ever more areas, including particularly sensitive ones such as criminal law, security, immigration and family law. The chance of a collision with basic national values becomes greater. Secondly, post-enlargement there are many more supreme courts in the Union, some of which have already shown themselves to be capable of assertiveness. Thirdly, in its recent Lisbon judgment the German Federal Constitutional Court (FCC) reiterated a possible new basis for rejection of EU law; that it has limited the sovereignty of the state so much that the state can no longer define the socio-economic circumstances of its citizens.17 This national capacity and autonomy was seen by the court as a part of   See also the loyalty obligation, Art 4(3) TEU.   See n 4. 15   See Art 27 of the Vienna Convention on the Law of Treaties; Eleftheriadis, above n 9, 12. 16   See J Komárek, ‘European Constitutionalism and the European Arrest Warrant: Contrapunctual [sic] Principles in Disharmony’ Jean Monnet Working Paper 10/05, http://centers.law.nyu.edu/ jeanmonnet/papers/05/051001.html; Baquero Cruz, above n 4. 17   Case 2 BvE 2/08 et al Lisbon [2009]. See C Wohlfahrt, ‘The Lisbon Case: A Critical Summary’ (2009) 10 German Law Journal 1277, 1284–85; ‘Karlsruhe has Spoken. “Yes” to the Lisbon Treaty but . . .’, editorial in (2009) 46 Common Market Law Review 1023; D Thym, ‘In the Name of Sovereign Statehood: A Critical Introduction to the Lisbon Judgment of the German Constitutional Court’ (2009) 46 Common Market Law Review 1795. 13 14

Constitutional Disagreement in Europe  275 the democracy which the Constitution protects. Thus for EU law to collide with the German Constitution it is not necessary that it violate more conventional human rights, but sufficient that it just ‘go too far’ so that there is not enough national freedom left to define life in the German Republic. This interesting idea takes a fairly conventional constitutional value – demo­ cracy – and turns it into an extremely broad tool for policing the spread of EU law. Even conventional socio-economic EU legislation may, just by virtue of its accumulated mass and effect, become a constitutional issue, perhaps without any particular rule being notably offensive or odd. This is in contrast to early perspectives on the supremacy debate which located it very much in the human rights corner, and emphasised the constitutional objection that the EU did not have an adequate system of rights protection.18 This was an easy one for the ECJ to address, and it did. However, the more general demand that EU law not limit state freedom too much will require a more subtle, and a broader, approach. This will be discussed further below.

V.  Cheerful Pluralism The situation sketched above is sometimes described in terms of constitutional pluralism, and when these words are used it is very often in a broadly approving way. Such pluralism is seen as protecting diversity, and as preferable to the domination of one legal order by another. Since both enjoy a certain legitimacy, but represent different interests and perspectives, surely their courts should engage in an ongoing and respectful dialogue, seeking conciliation, but retaining the possibility of disagreement. In this latter possibility lies the autonomy of the legal order, and its legal self-respect, and its judges’ capacity to fulfil their duty to those who have entrusted them with jurisdiction. Thus a pluralist perspective on the EU–national judicial relationship may welcome the consensual resolution of specific disagreements, but does not seek to resolve any of the ‘ultimate’ questions, or to remove the capacity for future differences. Poiares Maduro has recently taken this approach.19 One of the virtues of his analysis is that it is realistic, in that it accepts that differences about the question of ultimate supremacy will not go away. This is certainly the case for at least the near future. The ECJ shows no sign of discovering that the EU Treaties are in fact subject to constitutional law, and supreme courts 18   The emphasis began to move from rights to competence around the time of the Maastricht judgment of the FCC (Case 2 BvR 2134, 2159/92 Maastricht [1993] BVerfGE 89, 155). See M Kumm, ‘Who is the Final Arbiter of Constitutionality in Europe? Three Conceptions of the Relationship between the German Federal Constitutional Court and the European Court of Justice’ (1999) 36 Common Market Law Review 351; L Besselink, ‘Entrapped by the Maximum Standard: On Fundamental Rights, Pluralism and Subsidiarity in the European Union’ (1998) 35 Common Market Law Review 629. 19   Poiares Maduro, n 2.

276  Gareth Davies seem to be becoming only more assertive in their claims that Constitutions determine what applies within the national jurisdiction. The disagreement can only be finally resolved by the surrender of one side, which does not seem imminent. However, the nonchalant charm of a pluralist world view can easily lead to a rather too sanguine assessment of the actual state of affairs. Tolerance of the other is a virtue, and thoughtful dialogue is a good thing, but they are not always sufficient to provide a fair and predicable legal regime.20 In particular, a defence of constitutional pluralism inevitably attempts to gloss the stark choice between a hierarchy of law, and a breakdown of law. It is suggested here that this attempt is hopeless, and the attachment of a pleasant-sounding pluralist label to the venture makes it no less so. For the reasons described earlier, constitutional pluralism turns out to be inadequate as either a description or a justification of the way things now are. The only situation which might be described as constitutionally pluralist is where national judges sometimes prefer EU law and sometimes the national Constitution.21 Then both Treaties and Constitutions would still have a claim to be constitutional, at least sometimes, while co-existing. However, a consideration of this possibility shows why pluralism is a virtue mostly reserved for the political and social sphere. It is, in law, not a good thing. A legal system in which it is uncertain, or at the discretion of the judge, which rules apply is not compatible with most ideas of legal certainty. One might speak of a situation of constitutional arbitrariness. It is in fact the simple breakdown of law, and its replacement by judicial preference – the one judge being more pro-European than another.

VI.  Respect and Communication The policy risk which results from the European judicial standoff is that cold war becomes hot war: that EU law is set aside in national jurisdictions, on constitutional grounds, and that this becomes a trend which harms both the status of EU law and its application and effectiveness. Even if the concrete effect of constitutional judgments is not of great European importance – were the European Arrest Warrant not to take full effect in one land it would be regrettable, but not in itself a great impediment to European functioning or development, for example22 – the breaking of a taboo by actually setting EU rules aside may lead to a general loss of judicial respect for EU law. In many contexts the effectiveness of EU law is not just a question of application or not, but of the way that the rules are applied to particular facts, and the spirit in which they are interpreted. A diminution of respect may have hard-to-quantify but nevertheless important effects.   Baquero Cruz, n 4.   See above, n 9. 22   See Komárek, above n 16. 20 21

Constitutional Disagreement in Europe  277 This in turn may mean that EU policies do not achieve their goals, so that the capacity of the EU to influence economic and social development is limited, which in turn may heighten cynicism about the EU, and encourage policy-makers to look to other forums and means. Europe is the first port of call when crossborder problems arise only insofar as people believe that Europe is capable of effective action, and of harnessing the forces of application within Member States. How serious these risks are, and the likelihood of serious problems, is difficult to assess. It is really a matter which is best decided by empirical research into the attitudes and behaviour of judges, by psychologists and anthropologists. However, given that the political foundations of and support for the EU are not rock-solid, and its law is often viewed with some suspicion, dislike or mystification by national judges, it seems plausible that the risks should at least be taken seriously. That is to say, the EU and the ECJ should at least consider what measures they could take to address them and minimise the chance that a destructive conflict or sequence of conflicts occurs.23 In fact, both national and European courts have a legal obligation to minimise the chance of conflict. Even though national courts may express scepticism about EU law at the limits of its competence or legitimacy, they do not express doubt that most aspects of national participation in the EU are constitutionally legit­ imate. Thus the very same ideas of democracy and autonomy which threaten the EU in extreme circumstances may help it in everyday ones. The fact that the national democratic organs have chosen to participate in it is a reason to accept its rules wherever this is legally possible. Moreover, it is a fairly uncontroversial principle of good judging that as far as possible one seeks to reconcile the different rules relevant to a situation. On the whole, national courts should, and do, try to avoid stark choices. Nevertheless, such a soft approach has its limits, particularly where constitutions are the bearer of vested meanings and values. There are plenty of imaginable situations where a constitutional court might feel that conflict cannot be interpreted away. This leaves an important role for the ECJ. It is, moreover, in the position of supplicant. The national courts ask nothing of it, but it wants national courts to apply its law. Its law is also the law at risk. The ECJ is therefore the actor with the most obvious responsibility for reconciling the national and the European, by developing EU law in a way that avoids conflicts in the first place. It may also be noted that any interpretation of EU law which leads to its rejection by national courts is a bad interpretation, as a matter of EU law, since it manifestly does not achieve the goals of that law. It must moreover be open to doubt whether such an interpretation corresponds to the intention of the Treaty signers, insofar as that may matter. Poiares Maduro and Kumm have both suggested ways in which national and European courts can get along with each other. Poiares Maduro’s ideas are 23   N MacCormick, ‘Risking Constitutional Collision in Europe’ (1998) 18 Oxford Journal of Legal Studies 517.

278  Gareth Davies communicative and respect-based. He would like to see courts – his emphasis is on national courts, although not exclusively so – paying more attention to the nature of their reasoning.24 When they take decisions they should take account of other legal orders, show them respect, and reason their judgments using arguments accessible to courts in other states. Instead of just focusing on the local context, they should explain in more general terms, European terms, why their decisions are justified, to show that they have taken account of the non-national interests involved, such as EU policies, or actors established in other states. Poiares Maduro would like to see judicial reasoning become less parochial in its perspective and conceptual vocabulary. Kumm addresses what happens when conflicts occur.25 He suggests that there should be a principle of EU law which authorises national constitutional courts to set aside EU law on specific constitutional grounds, that is to say when there are specific types of conflict with important constitutional rules. This is in the spirit of Ross Phelan’s suggestion that EU law be set aside in circumstances where it would lead to national legal revolt or revolution.26 This Chapter proposes a variation on Kumm’s argument. What is taken from him is the idea that EU law should try to police itself, and not leave this to national constitutions. While national courts will undoubtedly continue to regard their constitutions as a line which may not be crossed, they may well be prepared to consider whether conflicts can be avoided by first voicing their concerns within the framework of EU law, if that framework allows such concerns to be heard. It is this giving of a voice to national constitutional concern to which EU law should now aspire. However, addressing national concerns purely by reference to the national Constitution, as Kumm does, has disadvantages. First, it cuts the ECJ out of the loop. Even if they were to be asked to contribute to the decision, via a reference, the framing of the issue purely in terms of national law would limit what they could contribute. Secondly, this approach encourages national legal insularity. It stimulates exactly what Poiares Maduro warns against: national courts which focus exclusively on their local doctrine and interests and fail to interpret their national law in a European context. Thirdly, it is implausible to expect constitutional courts to show any interest in this doctrine. It amounts to an EU law acknowledgment of what they had already asserted the independent right to do anyway. As such, it can be seen as an attempt to force constitutional courts to concede EU law supremacy even while they reject European rules, by claiming that their rejection is a part of EU law. The unshakeable importance which these courts attach to their Constitutions would probably lead them to refuse this offer to come into the European legal order. They would, it is suggested, continue to locate any reasoning about the interaction of the Constitution and the Treaties firmly within the constitutional context, as a matter of constitutional principle.   Poiares Maduro, n 1.   Kumm, n 1. 26   Ross Phelan, above n 10. 24 25

Constitutional Disagreement in Europe  279 But the most substantive objection is that by placing his conciliation process at the constitutional level Kumm offers a last line of national defence whereas it is a first line that is perhaps more urgently needed. It was suggested above that future constitutional conflicts may arise because ordinary EU law has spread so much that national autonomy and democracy are perceived to be threatened. The aim of the ECJ should be to prevent this point from being reached. EU law needs to be policed before it becomes a constitutional threat. It is precisely the escalation of everyday EU law to a national constitutional issue that the ECJ should try to avoid.

VII.  Policing by Proportionality and Procedure In fact direct conflicts with traditional constitutional law are likely to be relatively trivial matters, and easy to resolve. In the event that an EU measure violates some fundamental right it can be dropped or amended without loss of face as a matter of EU law.27 Where EU law directly contradicts other aspects of the Constitution, this is likely to be legal accident. For example, where the European Arrest Warrant was incompatible with several Constitutions, this was simply because states had not completed their constitutional amendments on time.28 Although the cases are discussed in terms of grand principle, the problem was caused by procedural hiccups, relatively easily sorted out with little long-term impact. In short, the principles and structures of classic constitutionalism are open enough, and unobjectionable enough, that complying with them is not a significant policy constraint for the EU and should not raise any structural problems. Conflicts will be incidents, not clashes of ideology. The more serious problem is what the FCC identified as the growing threat to national autonomy posed by EU law. It was concerned that the law may so limit the capacity of national governments to define the life circumstances of their citizens that national democracy could no longer be said to be effective.29 Given that national democracy remains, in the eyes of most constitutional courts, essential, and constitutionally protected, the policy-constraining effect of the growth of EU law becomes a potential constitutional issue. The most serious risk of constitutional conflict is therefore raised by everyday EU law – socio-economic, criminal, environmental – which constrains national policy freedom. The challenge of preventing open rejection of EU law is the challenge of ensuring that EU law as a whole does not invade autonomy to an extent that crosses the constitutional courts’ red lines.30 In other words, it is the interaction of ordinary   Internationale Handelsgesellschaft, n 12; Charter of Fundamental Rights of the European Union.   See Komárek, above n 16. 29   See above, n 17. 30   See G Davies, ‘Subsidiarity: the wrong idea, in the wrong place, at the wrong time’ (2006) 43 Common Market Law Review 63. 27 28

280  Gareth Davies EU law with ordinary national law and policy that needs policing, with the aim of avoiding the circumstances where national courts feel the need to invoke the constitutional protection of democracy. As examples of serious substantive policy constraints, one may consider free movement, competition law, state aid and the euro rules, all of which have farreaching and often unexpected consequences. As examples of intervention in areas of national law apparently removed from EU policy one may consider the rulings in which rules on the naming of children,31 the language of court cases,32 and questions of procedural administrative law were found to conflict with free movement,33 or in which state aid law has set limits to regional autonomy.34 Less surprisingly, but more importantly, one may consider the impact on welfare states and social policy of economic regulation as a whole.35 It is understandable that states may feel that their autonomy is increasingly circumscribed, to a point that they may find almost intolerable. Yet at the same time, a simple concession to these feelings of powerlessness would amount to the abandonment of important EU policies, to which those same states have agreed. What is needed is therefore a policing process which is effective, and demonstrably effective – showing that the law is policed is almost as important to calming national nerves as actually policing it. On the one hand, therefore, there must be a system ensuring that EU law does not cause unnecessary destructive effects on national policy, and on the other hand there must be a system ensuring that national concerns as well as European interests are voiced and articulated in the decision-making process, so that it is apparent why judicial decisions go the way they do. Following the two paths indicated by Poiares Maduro and Kumm, there must be better communication and respect between orders, and there must also be rules mediating conflicts of interest, rules which may be no less important for almost never being used.

VIII. Proposal The inclusion of state interests and of the value of autonomy should begin at the legislative phase. However, legislation, and the Treaty, will often have unintended   Case C-148/02 Garcia Avello [2003] ECR I-11613.   Case C-274/96 Bickel and Franz [1998] ECR I-7637.   Case C-224/97 Ciola v Land Vorarlberg [1999] ECR I-2517. 34   Case C-88/03 Portugal v Commission [2006] ECR I-7115; Joined Cases C-428/06 to C-434/06 Unión General de Trabajadores de La Rioja [2008] ECR I-6747; R Greaves, ‘Autonomous regions, taxation and EC state aids rules’ (2009) 34 European Law Review 779. 35   See Fritz Scharpf, ‘The European Social Model: Coping with the Challenges of Diversity’ (2002) 40 Journal of Common Market Studies 645; G Davies, ‘The Process and Side-effects of Harmonisation of European Welfare States’, Jean Monnet Working Paper 2/06, http://centers.law.nyu.edu/jeanmonnet/ papers/06/060201.html. 31 32 33

Constitutional Disagreement in Europe  281 consequences, particularly within a specific national institutional context.36 There is therefore also a need for a principle of substantive law which addresses the situation where EU law has unusually destructive or chaotic effects, and balances the interests involved. Such a balancing belongs naturally within proportionality, and as such is already a part of EU law. Asking whether the application of that law is in fact disproportionate, because it has particularly dramatic or harmful consequences, is not doctrinally new. However, in the application of proportionality national autonomy and the national capacity to formulate and carry out policy is rarely seen as a value in itself.37 That is precisely what needs to change. It is suggested that the statement below should be applied by the court: States often have to adapt their policies and institutions to comply with EU law. That is an unavoidable consequence of membership. However, if the application of EU rules makes achievement of important and legitimate national policy preferences effectively impossible, or unreasonably difficult, then, depending upon the degree of EU or other interest in full application of that rule, it may be disproportionate to apply the rule in the particular context in question. States must provide evidence that amending their systems or policies to achieve their goals in a way compatible with EU law would either be unreasonably difficult or disproportionately harmful to other interests.

The application of this would take place in the context of the preliminary reference procedure. As ever, application of EU rules is for the national court, and ultimately it would be they who would decide on specific cases of setting aside an EU rule. However, it would be appropriate to ask a question to the court, and it could even be suggested that a national court must do so if it is considering setting aside on these grounds. The court would then provide general guidance on the meaning of proportionality and the nature and strength of the EU interests to be taken into account in the balancing process. It should also provide guidance on the kind and level of evidence that may be expected from Member States. The ambiguous and slippery nature of the division of powers in the reference procedure means that both national and European courts have an important role. Neither is emasculated, and it is in the interest of each to formulate their concerns and the relevant interests at their level in the most clear and complete way. Should their collective deliberation lead to the conclusion that application of EU law would be disproportionate there should then be an automatic procedure for investigation of the situation, consultation with the Commission and so on. This proposal is presented in terms of proportionality, and the adoption of national autonomy as a value to be weighed within that principle. However, it can equally be seen as a generalised public policy exception, or as a subjection of the supremacy principle to the principle of proportionality; the court should acknow­ ledge that under certain circumstances to grant supremacy to an EU rule, over a 36  See J Trachtman ‘Regulatory Competition and Regulatory Jurisdiction’ (2000) 3 Journal of International Economic Law 331, on the inevitable imperfection of economic regulation. 37   See Davies, above n 30.

282  Gareth Davies far more important national one, may be disproportionate. It matters little which description is chosen. The key idea reflected in the proposal is that good policy and constructive European integration require the EU to take account of national interests in the formulation, interpretation and application of EU law, just as national courts and institutions are supposed to take account of EU interests when developing and using their domestic rules. The aim is to defuse conflict by making interests and balances explicit. Instead of national judges setting aside national law because they feel they have to, they would be setting aside national law because, having heard evidence, they are satisfied that the state has been unable to provide a pressing reason not to. Occasionally, they would be setting aside EU law, because the state has made a convincing argument for exceptional circumstances, whereupon the Commission and other parties would begin working upon an acceptable solution. It is of course a dangerous proposal. It concedes the possibility of letting substantive EU law fail, in some contexts, at some times. However, the interests involved at national level are real, and ignoring them is a politically untenable as well as an undemocratic option. At any event, they will be discussed and asserted by national courts. The EU interest is therefore in bringing that discussion as far as possible within the European legal order, to ensure that European as well as national interests have a voice in resolving conflicts. There are (at least) three more criticisms. One is that the proposal amounts to re-opening the Treaties. In accepting, and continually reaffirming, the substantive content of the Treaties Member States must be taken to have accepted the enforcement of the rules they contain. To introduce derogatory principles other than those which the Treaties already contain would be no more or less than a step backwards in integration. The response to this criticism is that it underestimates the importance of the openness of the Treaty texts. They are more convincing as commitments to a process than as precise policy statements, and it seems reasonable to suggest that as interpretation develops and extends the scope of EU law it should also, in parallel, develop and extend the doctrines of control and restraint. Indeed, this is precisely Cassis de Dijon: the price for bringing equally applicable rules within the law on free movement was the parallel extension of the possibility of derogation.38 This suggests the second obvious criticism: that the proposal above is no more than is already accepted in the law on free movement, where proportionate derogations on grounds of legitimate national interests are part of the law.39 Indeed there is much similarity, but the proposal goes two steps further. First, it generalises the idea of such derogations to all EU law, including secondary legislation. Second, it makes the balancing of EU and national interests more explicit. The national judge, in particular, is invited not just to consider the national interest at stake, but the extent of the European one.   Case 120/78, Rewe-Zentral (Cassis de Dijon) [1979] ECR 649.   Ibid; Case C-55/94 Gebhard [1995] ECR I-4165.

38 39

Constitutional Disagreement in Europe  283 Ironically, the greatest threat to EU law may then be national judges who fail to consider national interests. By over-applying EU law they raise the risk of its extension beyond the legitimate. On the contrary, such open rules need policing, and it will be the involvement of national judges in this process which will prevent constitutional conflict from becoming real. The third criticism is that this proposal would be bad for legal certainty. Particularly if it is applied to secondary legislation it will render all EU legislation conditional. Given the value placed on reciprocity by Member States, if they see other states successfully evading rules on the basis of special national circumstances this could lead to a spiral of ‘special cases’. This is, in fact, precisely the point. There are two kinds of legal uncertainty which may arise: uncertainty over whether the law will in fact be properly applied, and uncertainty about the content of the law. The aim here is to replace the first danger by the second, to internalise the tension between national policies and EU law. If states are not to ignore threatening rules, a mechanism is necessary to allow the threat to be addressed within the law. Having created that mechanism, the need is then to make the resulting uncertainty about content acceptable. That is done by exposing the rationale behind it. There is a trade-off between certainty and ambition. Rules which are clear and accepted will inevitably be limited in their capacity to create change. The EU, however, aspires to change Europe. Yet at the same time, it does not want to destroy the individuality of national legal systems or institutions. To reconcile these goals it is necessary to admit both ex ante and ex post assessments of the working and effect of EU law. The challenge is then to get states to accept the risks of ex post assessment, and that challenge is met by making it as reasoned and transparent as possible, and involving national courts as much as possible, so that the possibility of setting aside EU law after a reasoned judicial process becomes an integral part of its legitimacy, and also of its own goal of reconciling unity and diversity.

IX. Conclusion Accepting limits is part of legal maturity. EU law should not just recognise the lines drawn by doctrines of human rights and attribution, but also those resulting from the legitimate desire of communities to define their own life circumstances. If that means EU policies must sometimes make concessions, so be it: EU law has goals, but so do Member States, and EU law has no monopoly of legitimacy. An inability to compromise is usually fatal to relationships. The practical importance of taking proportionality seriously can be summed up without resort to pluralism: it makes the EU reasonable, and as such makes it a polity with which national law can work towards the shared goals of a better European society.

13 The Silent Lamb and the Deaf Wolves Constitutional Pluralism, Preliminary References and the Role of Silent Judgments in EU Law DANIEL SARMIENTO* Viola: What else may hap to time I will commit, Only shape thou thy silence to my wit.

Captain: Be you his eunuch, and your mute I’ll be: When my tongue blabs, then let mine eyes not see.

Viola: I thank thee: lead me on. W. Shakespeare, Twelfth Night, Act I, Scene II

I. Introduction Constitutional malaise is on the rise in the European Union, but researchers and doctors are experimenting with a new and revolutionary treatment under the name of pluralism. Magic formula for some and scientific truth for others, pluralism conveys all the ills and wrongs of groundbreaking discoveries: intense debate, miraculous results, angry and sceptical criticism and a loud herd of fervent believers. Among other claimed heresies, pluralism supports a non-hierarchical approach to the law, a conception of democracy based on discursive procedures and not majoritarian outcomes, an individualistic and even rebellious role of courts and other institutions, and a vision of legal principles close to moral relativism. However, the gravest danger posed by pluralism does not reside in its prescriptions. The threat lies in its methodological foundation, for pluralism is not only a normative theory, but also a descriptive account of law, courts, judicial politics and constitutional conflicts in contemporary Europe. In other words, pluralism is not only a recipe; it is the world we live in. * All opinions are strictly personal. Many thanks to Ricardo Alonso, Miguel Azpitarte, Víctor Ferreres, Xavier Groussot, Jan Komárek, Luigi Malferrari and Suvi Sankari for very useful comments. Of course, all errors remain my own.

286  Daniel Sarmiento As the world celebrates the 150th anniversary of the publication of On the Origin of Species, it is important to ponder on the importance of descriptive and normative theories that touch upon the foundations of previously well-affirmed beliefs. Although Darwin struggled to prove that his findings were based on empirical observation and not religious scepticism, it was the very factual basis of his theory that caused alarm. In the same vein, while critics cry against pluralism’s anarchic proposals for the traditional ordered and disciplined domain of the law, its defenders insist upon the fact that pluralism is a reflection of the legal order as it stands today. Despite the normative elements underlying some of its propositions, pluralism is substantially an account of the law and of its practice in the European Union. This Chapter will portray how pluralism is a fact in contemporary European law, and not only a normative agenda for lawyers and courts. It will be claimed that pluralism is a valid explanatory tool in dealing with conflicts of a constitutional nature, where issues of legitimacy and sovereignty underpin litigation. By studying the way in which the preliminary reference procedure, established in Article 267 TFEU, introduces a communicative tool that enhances discourse among courts in Europe, I will describe how the ECJ employs this device in the context of constitutional pluralism. Attention will be paid not so much to the content of the court’s preliminary decisions, but to its silences. Contrary to what ordinary litigation allows, the preliminary reference procedure enables the ECJ to hand down answers in hetero­ geneous formats: detailed replies that give an exact account of the referring court’s queries; imaginative answers that strive to furnish responses beyond those originally posed; or, as will be portrayed in this Chapter, half-answers or no answers at all. I will support that the way in which the ECJ grants what will be called ‘silent judgments’ can be an expression of pluralism. Section II of the Chapter will develop the different forms of silent judgments present in the ECJ’s case law. Section III explores the role of national courts when facing silent judgments or pre-empting the ECJ’s ability to render a decision of the kind. Sections IV and V analyse the virtues and dangers of silent judgments in the context of constitutional pluralism, purporting several normative theses in support of a jurisprudence that safeguards both coherence and authority in EU law. The ECJ’s silence can be a useful tool in developing co-operation, complicity or exit strategies, but it is also a dangerous device in the present historical circumstances. Surrounded by national legal orders claiming the ultimate authority on issues of constitutional relevance, EU law must strive to resolve and pacify in an ever more politically, socially and culturally complex Union. Not an easy task when 27 wolves are awaiting a wrong decision from the ECJ, in a contingent judicial environment in which authority and coherence are at stake. As Thomas Jefferson wrote, in correspondence with John Holmes: ‘We have the wolf by the ear, and we can neither hold him, nor safely let him go. Justice is in one scale, and self-preservation in the other’.1 1  T Jefferson (MD Peterson ed), Writings (New York, Literary Classics of the US, 1984) 1434. Although Jefferson was referring to slavery and its moral, political and economic consequences in postcolonial North America, the imagery (and the imagery only) suits the purposes of this Chapter well.

The Silent Lamb and the Deaf Wolves  287 In this Chapter I will argue that justice and self-preservation are both possible in the EU’s current judicial framework. Constitutional pluralism and silent judgments will prove useful devices in obtaining such a goal, by shaping one party’s silence to the other one’s wit.

II.  Setting the Scene: the Role of Legal Pluralism in EU Law A fable or not, the world was a candid place for European lawyers until 1992. The European Communities were the successful outcome of almost four decades of economic and legal integration, whereby a new legal order had been created, enjoying direct effect and primacy over national provisions.2 Despite a few rebellions in the early 1970s,3 the authority of Community law was well founded and its effectiveness proved true throughout the 12 Member States. The Single European Act and eventually the Treaty on European Union drastically changed the scene, introducing an unprecedented degree of tension in the process of integration that was ultimately confirmed by the German Constitutional Court’s Maastricht decision.4 As of then, serious doubts were cast upon the new Union’s ability to assume an authoritative discourse in sensitive areas that touched upon hard-core elements akin to Member States’ sovereignty: fundamental rights, institutional autonomy and competences become areas of EU influence that states were not so eager to sacrifice.5 The answer to this tension has been fabricated in the course of the past 15 years, but it can be resumed in a paradox: on the one hand, the European discourse claims authority over issues of constitutional relevance; on the other, member states assert the ultimate decision on the same matters. In other words, both parties argue to be the holders of the Kompetenz-Kompetenz, and in such tension we have lived, in parallel narratives, fragmenting the traditional vision of a hierarchical legal order in which a single Grundnorm presupposed the source of all authority. The pragmatic result of it all is that, in contrast with pyramidal legal systems, co-existence becomes a task based on mutual co-operation between legal orders. In Miguel Poiares Maduro’s words:

  Cases 26/62 Van Gend and Loos [1963] ECR 1 and 6/64 Costa v ENEL [1964] ECR 585.   See in general K Alter, Establishing the Supremacy of European Law. The Making of an International Rule of Law in Europe (Oxford, Oxford University Press, 2001). 4   Case 2 BvR 2134, 2159/92 Maastricht [1993] BVerfGE 89, 155, published in English as [1994] 1 CMLR 57. 5   ‘“The true world has become a fable”: that’s how we felt at the time. We had thought that the law of integration was for real, that it was an essential component of the more civilised supranational political community created in the 1950s. Now it seemed that we had been misled, that the Community law we knew was at risk, that it had became in part a fable, that perhaps it had always been one and never a true world. We were baffled.’ J Baquero Cruz, ‘The Legacy of the Maastricht-Urteil and the Pluralist Movement’ (2008) 14 European Law Journal 389, 391. 2 3

288  Daniel Sarmiento [I]n spite of their claims to ultimate authority and legal sovereignty, both the EU and national legal orders make more or less explicit concessions towards the claim of authority of the other legal order. They make the necessary adjustments to their respective claims in order to prevent an actual collision. EU law has introduced substantive constitutional changes such as fundamental rights in order to accommodate the claims made by national constitutions. National constitutions have been interpreted in a manner that tends to prevent the review of specific EU acts.6

To solve this conundrum, pluralism provides assistance in elucidating the current tension and purports a way forward. However, it is important to insist that this theory portrays a tension of a specific type of conflict: disagreement between institutions on issues of constitutional relevance. Pluralism is not a useful theory when explaining the intricacies of, say, directives, member state liability or national implementation of agricultural regulations.7 The task assigned to pluralism is an explanatory discourse that shows how two autonomous legal orders coexist, clash and keep operating hand in hand, when each legal order claims authority over a constitutional issue.8 This assertion implies the following consequences. First, pluralism is a legal theory that operates in the area of the law; it has no pretensions beyond this terrain, although it can eventually communicate with political philosophy, political science and sociology.9 Second, it is a theory specific­ally designed to solve claims of authority on issues of constitutional relevance. These can be briefly resumed in three categories: fundamental rights, institutional autonomy and competences. Third, by operating in this domain, pluralism is also a theory of European law that gives account of other crucial theoretical issues, such as normativity. Portraying the inner workings of EU law in the context of constitutional conflict, pluralism has the potential of becoming a fully fledged theory of law that poses uncomfortable questions for those who purport a state-centred vision of a legal order. For this reason, a pluralist theory of European law has the potential of envisioning a theory of sovereignty in the European Union.10 Also, there are different strands of pluralism that put their attention on different actors. The most extended expression of pluralism focuses on judicial constitutional 6  M Poiares Maduro, ‘Contrapunctual Law: Europe’s Constitutional Pluralism in Action’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 524. 7   Although authors have attempted to address the complexities of EU legal institutions through the lenses of pluralism, their claims are still deeply attached to a theory of EU law and to the issues underlying constitutional conflict. For example, A von Bogdandy, ‘Pluralism, Direct Effect, and the Ultimate Say: On the Relationship between International and Domestic Constitutional Law’ (2008) 6 International Journal of Constitutional Law 397. 8   Neil Walker has distinguished constitutional pluralism from other strands of pluralism, claiming that ‘not everything which meets the test of legal or other qualifying normative order under the rubric of these various pluralisms also qualifies as constitutional discourse’. N Walker, ‘The Idea of Constitutional Pluralism’ (2002) 65 Modern Law Review 317. 9   Walker, n 8, 319. This assertion will obviously depend on the concept of law being held, which may embrace a wide array of facts and conducts, sometimes beyond the formal domain of positive legal provisions. 10   N Walker, ‘Late Sovereignty in the European Union’ in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 31.

The Silent Lamb and the Deaf Wolves  289 conflict, whereby courts engage in claims of authority over the definitive application and interpretation of the law. However, there are other variations of the theory that concentrate on the tensions between other institutional actors besides courts, namely executive, legislative and (exceptionally) non-state actors.11 For the present purposes, this Chapter will develop the judicial branch of pluralism, with a special emphasis in the study of constitutional conflicts arising between the ECJ and national courts. Other writers have purported a wider approach to judicial pluralism, by giving voice to national courts in the conflicts that may arise among them, claiming that judicial dialogue is not only a matter of two, but of a varied array of judicial interlocutors.12 This approach will not be followed in this Chapter, on the understanding that the substantial discourse in the resolution of constitutional conflict corresponds to two competing legal orders. Although I do not deny the import­ ance of judicial discourse à vingt-huit, such an approach will not be employed in the following pages. In the context of judicial constitutional pluralism, the main descriptive results of this theory can be resumed in the following: first, legal orders in the EU operate under shifting grundnorms, depending on the scope of application of each one;13 second, and despite the separation among legal systems, the criteria for the resolution of constitutional conflicts are to be found in mutually enhancing normative texts;14 and third, litigation finds no authoritative resolution based on the competence of one of the judicial actors at play, but on the result of constructive communication over specific issues.15 The last two features presuppose the ability of a national court to disregard EU law when its provisions find no reflection in national constitutional rules and discursive engagements provide no solution to such tension. For example, if a national Constitution contains a rule that proves clearly incompatible with EU law, domestic courts will be empowered to put aside European provisions as long as no interpretative effort can possibly save the conflict. This outcome is not the result of the primacy of national constitutions, but of a system of mutually dependant legal orders that take cooperation to its limits, and allow courts to change the grundnorm when conflicts become unsolvable.16 As the German Constitutional Court expressed in the Maastricht decision: 11   See generally Poiares Maduro, n 6, 501; M Rosenfeld, ‘Rethinking Constitutional Ordering in an Era of Legal and Ideological Pluralism’ (2008) 6 International Journal of Constitutional Law 415; J Komárek, ‘European Constitutionalism and the European Arrest Warrant: In Search of the Limits of Contrapunctual Principles’ (2007) 44 Common Market Law Review 9, 37–39. 12   See, eg, the discursive approach developed by Poiares Maduro under the heading of ‘vertical and horizontal coherence’: Poiares Maduro, n 6, at 527–29. 13   See N Barber, ‘Legal Pluralism and the European Union’ (2006) 12 European Law Journal 306, 316–18. 14   M Kumm, ‘The Jurisprudence of Constitutional Conflict. Constitutional Supremacy in Europe before and after the Constitutional Treaty’ (2005) 11 European Law Journal 262, 286. 15   Poiares Maduro, n 6, 524. 16   According to Kumm, ‘The task of national courts is to construct an adequate relationship between the national and the European legal order on the basis of the best interpretation of the principles underlying them both. The right conflict rule or set of conflict rules for a national judge to adopt is the one that is best calculated to produce the best solutions to realise the ideals underlying legal practice in the European Union and its Member States.’ Kumm, n 14 at 286.

290  Daniel Sarmiento Whereas a dynamic expansion of the existing Treaties has so far been supported on the basis of an open-handed treatment of Article 235 of the EEC Treaty [now Article 352 TFEU] as a ‘competence to round-off the Treaty’ as a whole, and on the basis of considerations relating to the ‘implied powers’ of the Communities, and of Treaty interpretation as allowing maximum exploitation of Community powers . . . in future it will have to be noted as regards interpretation of enabling provisions by Community institutions and agencies that the Union Treaty as a matter of principle distinguishes between the exercise of a sovereign power conferred for limited purposes and the amending of the Treaty, so that its interpretation may not have effects that are equivalent to an extension of the Treaty. Such an interpretation of enabling rules would not produce any binding effects for Germany.17

Constitutional pluralism has also a normative side. A legal system that operates within a shifting parameter of legality, and in which legal texts have to be interpreted in ways that reflect mutual deference and respect, must articulate ways and means that give shape to the peculiar discourse that gives answer to constitutional conflicts.18 To this purpose, pluralism offers distinct perspectives that enrich the overall use of the theory. Mattias Kumm provides, pursuant to ‘ideals that have provided the underpinning for liberal-democratic constitutionalism’,19 normative criteria that determine the conditions under which EU law must give way to the autonomy of national provisions. Miguel Poiares Maduro offers the procedural version, by purporting the need of national courts to develop, by active co-­ operation between them and with the ECJ, together with a claim of universalisability, principles that best serve the goals of integration and safeguard national identity.20 Jan Komárek has delved with the role of pluralism in the reasoning of courts that engage in constitutional conflict. While Kumm and Poiares Maduro offer substantive criteria that should eventually resolve the clash between legal orders, Komárek proposes an analytical framework in the context of a pluralist theory of legal reasoning. All three strands represent the foundations of what Leonard Besselink and Tom Eijbouts have generically described as ‘a Law of Laws’,21 and struggle to put order in the ways in which courts should solve constitutional conflicts.   Maastricht, n 4, 105.   The need to distinguish between a descriptive and normative strand of constitutional pluralism has been stressed by Walker, n 8, 338. He adds, a third strand, ‘epistemic pluralism’, according to which ‘the very representation of distinct constitutional sites . . . as distinct constitutional sites implies an incommensurability of the knowledge and authority (or sovereignty) claims emanating from those sites’. 19   Kumm, n 14. 20   Poiares Maduro, n 6. 21  L Besselink and T Eijbouts, ‘Editorial: The Law of Laws – Overcoming Pluralism’ 4 (2008) European Constitutional Law Review 395, 397, contributes to the debate with a maxim that gives shape to the ‘Law of Laws’, according to which ‘A judge or another public authority may qualify the validity, in his own jurisdiction, of a rule of more general circumscription than his own and binding on him. He shall do so not by mere reference to the autonomy or the supremacy of his own legal order, nor by reference to a legal hierarchy. He shall do so by reference to fundamental substantive norms valid in the wider circumscription also, or by putting forward such substantive norms that he holds to be applicable also there.’ 17 18

The Silent Lamb and the Deaf Wolves  291 The reasons why the descriptive and normative strands of pluralism have found fertile soil in EU law are not hard to guess. First, European provisions stand on brittle ground as far as competences are concerned, whilst the ECJ has developed principles that encourage authorities to stretch its area of influence beyond what Member States might be willing to tolerate.22 This combination of ambiguous attribution of competence and of judicial expansion of European rules causes unavoidable reactions from domestic instances, which will eventually focus their attention on the ECJ. Second, national Constitutions transfer competences to the EU in such a way that it is practically impossible to trace the ‘red lines’ of national autonomy. National constitutional language uses a state-centred tone, but its vocabulary is filled with EU-centred words and expressions, so many that one wonders where the limits of the Union’s powers are vis-à-vis the state. And third, honest and bona fide discourse is an essential part of the resolution of a constitutional conflict, inasmuch as the ECJ’s communicative channel lies mainly in the preliminary ruling procedure, which gives voice to the Luxembourg court if and only if a national jurisdiction is willing to raise the question. For this last reason, the ECJ must give answer to the queries posed, but national courts must also comply with their duty to refer.23 This is a subtle exercise of judicial diplomacy in which a court refuses to be the relevant deciding actor in a point of constitutional relevance, and another must play such a role but prove to be humble and thankful. A graphic way through which the ECJ handles this delicate task can be depicted by the detail of its answers, for Article 267 TFEU is flexible enough to admit exhaustive replies, partial answers, or no answers at all. By doing so in the context of a constitutional conflict, the ECJ will implement a conduct that has been well described by pluralism, according to which the resolution of such conflicts require a European and a national interlocutor, mutual deference, and a constructive effort in the interpretation of both legal systems. The scope of the ECJ’s answer will prove to what extent it is willing to defer to its domestic counterpart, and in what terms. Therefore, silence and the use of silence by the ECJ is a valuable explanatory tool among the utensils of pluralism. In his theory of judgment, Immanuel Kant claimed that empty decisions were cognisable and always had a meaning, and this will prove truer than ever when analysing a series of preliminary decisions rendered by the ECJ in the context of constitutional conflict. In order to do so, I will purport a typology of silence that may portray with greater clarity the role of ‘silent judgments’ in EU law. 22  See, eg, the ECJ’s case law on environmental protection and criminal law (Cases C-176/03 Commission v Council [2005] ECR I-7879 and C-440/05 Commission v Council [2007] ECR I-0000), or on the principle of non-discrimination on the grounds of age (Case C-144/04 Mangold [2005] ECR I-9981), both the cause of roaring criticism from Member States. 23   A duty not only purported by a pluralist approach to EU law, but also by the Treaties themselves. Article 257 TFEU establishes that a national court against whose decisions there is no judicial remedy under domestic law, when faced with a question on the interpretation or validity of EC law shall bring the matter before the Court of Justice.

292  Daniel Sarmiento

III.  The Silent Lamb: Classifying the ECJ’s Laconic Case Law Being a flexible instrument as it is, the preliminary reference procedure grants the ECJ a wide margin of manoeuvre when facing interpretative queries coming from national jurisdictions. In the same way that the referring judge is free to determine in what ways his question will be of use for the case at hand, the ECJ is also at liberty to consider in which manner his reply will be of the best utility. When the case touches upon issues of constitutional principle, the ECJ’s ability to give a reply might eventually clash with the autonomy of a Member State, and thus lead to a constitutional conflict. It is not surprising, thus, that the case law’s approach to these affairs deposits much importance on the way in which the answer is framed, the intensity of its normative content, the deference it grants to the referring court, the needs of uniformity and coherence, etc. These factors are all balanced through a subtle and complex use of both language and silence, in a manner that would fit appropriately in a theory of judicial minimalism.24 I will develop three forms of silence that portray the way in which the ECJ employs Article 267 TFEU in the context of constitutional conflict. First, the case of complete silence will be explored, whereby the ECJ renders no solution to the question posed, either because it considers that it has been answered in other ways, or because it is not useful for the purposes of the query. Second, partial silence will be portrayed as a form of judicial minimalism, whereby the ECJ decides on particular points of law, leaving the concrete answer to the referring court. And third, the strange case of unheard replies will be analysed. This form of judgment appears as a consequence of the ECJ’s recent decision in Cartesio, according to which inferior national courts have a discretion to set aside decisions of superior national courts that quash referring orders. However, national practice shows that Cartesio is a difficult decision to implement, since inferior courts tend to follow the orders of their superior courts. When this last circumstance occurs, the ECJ can put the referring judge in a tragic dilemma and ultimately drive him to ignore the interpretative help. In this third and last circumstance, silence is not a matter of the speaker, but of the listener, because nobody might be listening any longer.

A.  Complete Silence: Struggling with Authority Unanswered references are not an exception in European case law. The ECJ is under an obligation to reply only when the conditions established in Article 267 TFEU are met. This leads the court to discard queries beyond the boundaries of   See C Sunstein, Legal Reasoning and Political Conflict (Oxford, Oxford University Press, 1996).

24

The Silent Lamb and the Deaf Wolves  293 EU law,25 fictitious or hypothetical conflicts,26 deficiently motivated references,27 or questions posed by authorities which do not come under the Treaties’ definition of ‘jurisdiction’.28 However, there is another strand of silence much more creative on the ECJ’s side, whereby it replies by only handing ‘useful answers’ to the referring court.29 This technique enables the ECJ to adapt the original question into a version that may indeed appear more useful for the national judge, but it also allows the former to avoid delicate points, delay an issue for future occasions, or grant a wider margin of action to the national court. In the context of constitutional conflicts, this technique proves even more practical for the ECJ’s purposes, inasmuch it sets the scope of the answer the Luxembourg court is willing to provide. Two graphic examples of this approach can be explored in the judgments rendered in Grogan30 and Centro Europa 7.31 The Society for the Protection of Unborn Children Ireland brought proceedings against Stephen Grogan and 14 other officers of students associations, in connection with the distribution in Ireland of specific information relating to the identity and location of clinics in another Member State where medical termination of pregnancy was carried out. Abortion was prohibited in the Irish Constitution since 1983,32 and, according to the Irish High Court and Supreme Court, to assist pregnant women in Ireland to travel abroad to obtain abortions, inter alia by informing them of the identity and location of a specific clinic or clinics, came under such prohibition.33 The ECJ was asked to determine if the activity of carrying out an abortion came within the definition of ‘services’ provided for in Article 49 TEC [now Article 56 TFEU] and, if so, whether the plaintiff’s actions constituted a legitimate restriction to the freedom to provide services. The reference was setting the ECJ on a collision course with the Irish Constitution, forcing the former to legalise, pursuant to EU law, an activity that was banned by both the national constitutional law and case law. The ECJ’s answer is a graphic example of how silence can be a helpful tool in a pluralist setting.34 In reply to the first question, the Luxembourg court declared 25   Cases C-144/95 Maurin [1996] ECR I-2909, para 12; C-291/96 Grado and Bashir [1997] ECR I-5531, paras 13–17; and Case C-299/95 Kremzow [1997] ECR I-2629, para 16. 26   Cases 104/79 Foglia v Novello [1980] ECR 745, 760; and 244/80 Foglia v Novello (No 2) [1981] ECR 3045, paras 29–31. 27   Cases C-320/90 to C-322/90 Telemarsicabruzzo and Others [1993] ECR I-393, para 6. 28   Case 61/65 Vaassen-Göbbels [1966] ECR 377. 29   Among others, see Cases C-379/98 PreussenElektra [2001] ECR I-2099, para 38; C-94/04 and C-202/04 Cipolla and Others [2006] ECR I-11421, para 25; C-35/99 Arduino [2002] ECR I-1529, para 25; and C-222/05 to C-225/05 Van der Weerd and Others [2007] ECR I-4233, para 22. 30   Case C-159/90 Society for the Protection of Unborn Children Ireland [1991] ECR I-4685 (Grogan). 31   Case C-380/05 Centro Europa 7 [2008] ECR I-349. 32  Constitutional amendment approved by referendum inserted in Art 40, s 3, of the Irish Constitution a third subsection worded as follows: ‘The State acknowledges the right to life of the unborn and, with due regard to the equal right to life of the mother, guarantees in its laws to respect, and, as far as practicable, by its laws to defend and vindicate that right.’ 33   High Court, judgment of 19 December 1986, and Supreme Court, judgment of 16 March 1988, The Attorney General (at the relation of the Society for the Protection of Unborn Children Ireland Ltd) v Open Door Counselling Ltd and Dublin Wellwoman Centre Ltd [1988] Irish Reports 593). 34   Grogan, n 30.

294  Daniel Sarmiento that the performance of an abortion can be normally provided for remuneration and may be carried out as part of a professional activity.35 Considering that medical services fall under Article 49 TEC, abortions were services of the same nature and, thus, under the scope of the Treaties. The ECJ was immune to arguments raised by the plaintiffs in the main proceedings, for whom the provision of abortion could not be regarded as being a service, on the grounds that it is grossly immoral and involves the destruction of the life of a human being.36 However, this was the only point on which the ECJ was willing to take a firm stance. When approaching the second query, the judgment wandered into a different tone, characterised by a subtle silence. The defendants in the main proceedings were not professionals performing an economic activity, but students engaged in the diffusion of information concerning an economic activity. According to the court, ‘the link between the activity of the students associations . . . and medical terminations of pregnancies carried out in clinics in another Member State is too tenuous for the prohibition on the distribution of information to be capable of being regarded as a restriction within the meaning of Article [49 TEC]’.37 This assertion enabled the court to resolve the case in favour of the plaintiff in the main proc