General Principles of Law: European and Comparative Perspectives 9781509910717, 9781509910724, 9781509910694

Citation preview

GENERAL PRINCIPLES OF LAW Examining general principles of law provides one of the most instructive examples of the intersection between EU law and comparative law. This collection draws on the expertise of high-profile and distinguished scholars to provide a critical examination of this interaction. It shows how general principles of EU law need to be responsive to national laws. In addition, it is clear that the laws of the ­Member States have no choice but to be responsive to the general principles which are developed through EU law. Viewed through the perspective of proportionality, legal certainty, and fundamental rights, the dynamic relationship between the ingenuity of the Court of Justice, the legislative process and the process of Treaty revision is comprehensively illustrated. Volume 23: Studies of the Oxford Institute of European and Comparative Law

Studies of the Oxford Institute of European and Comparative Law Editor Professor John Cartwright Board of Advisory Editors Professor Mark Freedland, FBA Professor Stephen Weatherill Professor Stefan Enchelmaier Recent titles in this Series Volume 15: The Protection of Fundamental Rights in the EU After Lisbon Edited by Sybe de Vries, Ulf Bernitz and Stephen Weatherill Volume 16: The Involvement of EU Law in Private Law Relationships Edited by Dorota Leczykiewicz and Stephen Weatherill Volume 17: Current Problems in the Protection of Human Rights: Perspectives from Germany and the UK Edited by Katja S Ziegler and Peter M Huber Volume 18: Legal Challenges in the Global Financial Crisis: Bail-outs, the Euro and Regulation Edited by Wolf-Georg Ringe and Peter M Huber Volume 19: The Unitary EU Patent System Edited by Justine Pila and Christopher Wadlow Volume 20: The EU Charter of Fundamental Rights as a Binding Instrument: Five Years Old and Growing Edited by Sybe de Vries, Ulf Bernitz and Stephen Weatherill Volume 21: The Images of the Consumer in EU Law: Legislation, Free Movement and Competition Law Edited by Dorota Leczykiewicz and Stephen Weatherill Volume 22: Passing Wealth on Death: Will-Substitutes in Comparative Perspective Edited by Alexandra Braun and Anne Röthel

General Principles of Law European and Comparative Perspectives

Edited by

Stefan Vogenauer and Stephen Weatherill

OXFORD AND PORTLAND, OREGON 2017

Hart Publishing An imprint of Bloomsbury Publishing Plc Hart Publishing Ltd Kemp House Chawley Park Cumnor Hill Oxford OX2 9PH UK

Bloomsbury Publishing Plc 50 Bedford Square London WC1B 3DP UK

www.hartpub.co.uk www.bloomsbury.com 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 www.isbs.com HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published 2017 © The editors and contributors severally 2017 The editors and contributors severally have asserted their right under the Copyright, Designs and Patents Act 1988 to be identified as Authors of this work. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. While every care has been taken to ensure the accuracy of this work, no responsibility for loss or damage occasioned to any person acting or refraining from action as a result of any statement in it can be accepted by the authors, editors or publishers. All UK Government legislation and other public sector information used in the work is Crown Copyright ©. All House of Lords and House of Commons information used in the work is Parliamentary Copyright ©. This information is reused under the terms of the Open Government Licence v3.0 (http://www. nationalarchives.gov.uk/doc/open-government-licence/version/3) except where otherwise stated. All Eur-lex material used in the work is © European Union, http://eur-lex.europa.eu/, 1998–2017. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. ISBN: HB: 978-1-50991-071-7 ePDF: 978-1-50991-069-4 ePub: 978-1-50991-070-0 Library of Congress Cataloging-in-Publication Data Names: Vogenauer, Stefan, 1968- editor.  |  Weatherill, Stephen, 1961- editor. Title: General principles of law / edited by Stefan Vogenauer and Stephen Weatherill. Description: Oxford [UK] ; Portland, oregon : Hart Publishing, 2017.  |  Series: Studies of the Oxford Institute of European and comparative law ; volume 23  |  Includes bibliographical references and index. Identifiers: LCCN 2017004608 (print)  |  LCCN 2017004793 (ebook)  |  ISBN 9781509910717 (hardback : alk. paper)  |  ISBN 9781509910700 (Epub) Subjects: LCSH: Law—European Union countries.  |  Comparative law—European Union countries.  |  International and municipal law—European Union countries.  |  Proportionality in law— European Union countries.  |  Customary law—European Union countries.  |  Legal certainty—European Union countries.  |  Court of Justice (Court of Justice of the European Union) Classification: LCC KJE969 .G46 2017 (print)  |  LCC KJE969 (ebook)  |  DDC 349.24—dc23 LC record available at https://lccn.loc.gov/2017004608 Typeset by Compuscript Ltd, Shannon To find out more about our authors and books visit www.hartpublishing.co.uk. Here you will find extracts, author information, details of forthcoming events and the option to sign up for our newsletters.

PREFACE

In September 2015 we, in our capacities as Director and Deputy Director of the Oxford Law Faculty’s Institute of European and Comparative Law, held a two-day conference in Oxford at which speakers and participants investigated the nature, purpose and development of general principles of law. This book, containing papers presented in Oxford and refined in the light of our deliberations, serves as the fruit of that conference. For the Institute of European and Comparative Law, the subject matter could not have been more appropriate to our mission. General principles stand at the intersection between EU law and comparative law. They offer a perfect opportunity for scholars and practitioners of EU law and comparative law to come together and exchange views. That is precisely what we were able to enjoy at the Oxford conference. But there was more to our conference than intellectual energy alone. The conference was held on the occasion of the twentieth anniversary of the Institute of European and Comparative Law. It was an occasion at which we were able to celebrate the contribution made by the Institute to enriching the Oxford Law Faculty and marking it as not only a home to first-class common law scholarship but also as a place dedicated to pursuit of a prominent place among the family of European law schools. This is achieved by the Institute’s contributions to the four-year undergraduate law degree at Oxford, on which students spend their third year in continental Europe (in France, Germany, Spain, Italy or the Netherlands), by our regular seminars and conferences, by acting as host to visitors from across the world, but also by our long and growing list of books published in the series Studies of the Oxford Institute of European and Comparative Law. This is the 23rd volume appearing in that series. The several Directors of the Institute have all played their part in these successes—Basil Markesinis, Stephen Weatherill, Mark Freedland and Stefan Vogenauer—and the twentieth anniversary conference allowed us to recognise and celebrate a fresh transition in the Directorship, from Stefan Vogenauer to John Cartwright. It also allowed us to thank the Institute’s peerlessly brilliant Administrator, Jenny Dix. We are immensely grateful to the many friends of the Institute who helped to make such a success of our celebratory conference in September 2015 and we thank especially those who have also contributed to this book. We hope and believe it is a fitting tribute to the excellence of the work undertaken in the fields of European and comparative law in Oxford and so too we present it as a promise of much more to come in that vein into the future. Stefan Vogenauer and Stephen Weatherill

vi 

CONTENTS

Preface�������������������������������������������������������������������������������������������������������������������������v List of Contributors��������������������������������������������������������������������������������������������������� xi

1. Introduction��������������������������������������������������������������������������������������������������������1 2. General Principles of International Law������������������������������������������������������������5 Catherine Redgwell 3. From Myth to Reality: The EU’s ‘New Legal Order’ and the Place of General Principles Within It�������������������������������������������������21 Stephen Weatherill

Part I: General Principles of European Union Law 4. Theorising General Principles of EU Law in Perspective: High Expectations, Modest Means and the Court of Justice��������������������������41 Urška Šadl and Joxerramon Bengoetxea 5. Is Legality a Principle of EU Law?��������������������������������������������������������������������53 Alexander Somek 6. General Principles and the Many Faces of Coherence: Between Law and Ideology in the European Union����������������������������������������77 Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson 7. General Principles and Customary Law in the EU Legal Order�������������������105 Samantha Besson

Part II: Proportionality 8. Proportionality������������������������������������������������������������������������������������������������133 Alison L Young and Gráinne de Búrca 9. Proportionality and Judicial Review: A UK Historical Perspective�������������������������������������������������������������������������������������������������������145 Paul Craig

viii 

Contents

10. The Evolution of the Principle of Proportionality in EU Law—Towards an Anticipative Understanding?������������������������������167 Juliane Kokott and Christoph Sobotta 11. Proportionality and the Margin of Appreciation: Strasbourg and London��������������������������������������������������������������������������������179 Philip Sales 12. Origins and Presentation of the Proportionality Principle in French Case Law�������������������������������������������������������������������������������������������195 Yoan Sanchez 13. An Italian Perspective on the Principle of Proportionality������������������������221 Giuseppe Martinico and Marta Simoncini 14. The Application of Proportionality in Denmark in the Light of European Legal Integration������������������������������������������������������������243 Helle Krunke

Part III: Private Autonomy and Protection of the Weaker Party 15. Private Autonomy and Protection of the Weaker Party������������������������������255 Stephen Weatherill, Stefan Vogenauer and Petra Weingerl 16. Private Autonomy and the Protection of the Weaker Party: Historical��������������������������������������������������������������������������������������������269 Hector L MacQueen and Stephen Bogle 17. Personal Freedom and the Protection of the Weak through the Lens of Contract: Jurisprudential Overview���������������������������297 Dori Kimel 18. Private Autonomy, Weak Parties and Private Law: Views from Law and Economics������������������������������������������������������������������307 Fernando Gómez and Mireia Artigot 19. Fairness at a Time of Perplexity: The Civil Law Principle of Fairness in the Court of Justice of the European Union��������������������������������������������������������������������������������������������329 Daniela Caruso 20. Discrimination on Grounds of Obesity������������������������������������������������������355 Niilo Jääskinen

Contents

 ix

21. Discrimination on the Grounds of Sexual Orientation and Gender Identity���������������������������������������������������������������������������������������������365 Alina Tryfonidou * * * Concluding Remarks 22. Judging General Principles���������������������������������������������������������������������������397 David Edward

Index�����������������������������������������������������������������������������������������������������������������������411

x 

LIST OF CONTRIBUTORS

Joxerramon Bengoetxea teaches Jurisprudence at the University of the Basque Country and Comparative Legal Cultures at the International Institute for the Sociology of Law, Oñati. Samantha Besson is Professor of Public International Law and European Law at the University of Fribourg. Stephen Bogle is a Lecturer in Private Law at the University of Glasgow. Daniela Caruso is Professor of Law and Jean Monnet Chair at Boston University School of Law. Paul Craig is Professor of English law at St John’s College, Oxford. Gráinne de Búrca is Florence Ellinwood Allen Professor of Law at New York University. Sir David Edward is Emeritus Professor at the University of Edinburgh and former Judge at the European Court of Justice. Mireia Artigot Golobardes is Visiting professor at the Law School of Pompeu Fabra University and Affiliated professor at the Economics Department of Pompeu Fabra University. Fernando Gomez is Professor of Civil Law and Law and Economics at Universitat Pompeu Fabra, Barcelona. Xavier Groussot is professor of EU Law at Lund University, Faculty of Law. He is also pro dean and the Director of the Master Programme in European ­Business Law. In 2009, he became a guest professor at Université Panthéon-Assas (Paris II) and he is also guest Professor at the University of Reykjavik and ­University of Nagoya. Jörgen Hettne is associate professor of EU Law at Lund University, Department of Business Law, School of Economics. He is also a senior researcher at the Swedish Institute for European Policy Studies (SIEPS). Niilo Jääskinen is a Justice of the Supreme Administrative Court of Finland and former Advocate General of the CJEU. Dori Kimel is Reader in Legal Philosophy at the University of Oxford.

xii 

List of Contributors

Juliane Kokott is Advocate General at the Court of Justice of the European Union and a former Professor of Law at several Universities. Helle Krunke is Professor of Constitutional Law at the University of Copenhagen. Hector MacQueen is Professor of Private Law at the University of Edinburgh and a Scottish Law Commissioner. Giuseppe Martinico is Associate Professor of Comparative Public Law at ­Scuola Sant’Anna, Pisa and Research Fellow at the Centre for Studies on Federalism, Turin. Gunnar Thor Petursson is professor at the University of Reykjavík. He is also a visiting lecturer at the Université Panthéon-Assas (Paris II) and University of Lund, Sweden, since 2009, and a private legal adviser for the public sector and interest groups. He has been the chairman of the Icelandic appellate body for Post- and Telecommunication affairs, and an Ad-hoc College Member for the EFTA Surveillance Authority since 2014. Catherine Redgwell is Chichele Professor of Public International Law at the ­University of Oxford. Urška Šadl is Professor of Law at the European University Institute in Florence, Italy. Philip Sales is a Lord Justice of Appeal in the Court of Appeal of England and Wales. Yoan Sanchez is a PhD candidate in Comparative Public Law at the Universities of Orléans and Strasbourg, France. Marta Simoncini is FWO post-doctoral Fellow at the University of Antwerp and King‘s College London. Christoph Sobotta is Legal Secretary at the Court of Justice of the European Union. Alexander Somek is Professor of Legal Philosophy at the University of Vienna and Global Affiliated Professor of Law at the University of Iowa. Alina Tryfonidou is Associate Professor in EU Law at the University of Reading. Stefan Vogenauer is Director of the Max Planck Institute for European Legal ­History, Frankfurt. Stephen Weatherill is Jacques Delors Professor of European Law at the University of Oxford. Petra Weingerl is a DPhil candidate in Law at the University of Oxford. Alison Young is a Professor of Public Law at the University of Oxford and a Fellow of Hertford College.

1 Introduction I.  The Challenge of General Principles of Law The very idea of ‘general principles of law’ has an immediate allure. It seems to promise a systematic structuring of a legal system beyond the intricacies of individual and detailed rules. It suggests a means to enhance the coherence of the system. It offers an opportunity to fill gaps: resort to general principles may provide a way to resolve disputes in the absence of any clearly right answer discernible from existing texts. Yet equally the notion of ‘general principles of law’ raises disquiet. What are they, where do they come from, and who decides on their content and application? Most prominent of all in this sea of troubles, the question is whether the calculatedly neutral claims to the virtues of general principles as instruments apt to achieve coherence and to fill gaps in fact conceal a device to assert judicial power to—in short—make law. Asserting the existence of general principles is inevitably to embark on a normatively charged quest. Accordingly the deduction and use of general principles in the particular context of judicial decision-making invites appreciation of the challenge of legitimacy. This constitutes an important thematic concern of this book. The two particular general principles that are the subject of examination by several authors in this book—proportionality and the protection of the weaker party—provide especially vivid insights into the broader challenges of (in short) legitimacy which are explored at a more general level in the other contributions to the book. Plainly the two principles have their differences—proportionality is mainly (though not exclusively) associated with restraining the exercise of legislative and administrative power by public authorities, while the protection of the weaker party is mainly (though not exclusively) associated with restraining the autonomy of private parties who enjoy superior economic power. But both share a concern to locate limits to the freedom of action of the affected actors, be they public bodies or private parties, and, crucially, both demand that difficult questions be asked about the basis, predictability and extent of those limits—and about the judicial role in their development and application.

2 

Introduction

II. The European Challenge of General Principles of Law If general principles of law serve to make a legal system more coherent and to seek to fill gaps, then their development in the context of the European Union is worthy of particular attention. The EU’s legal order is based on international treaties, initially those that entered into force in the 1950s, the founding treaties of Paris and Rome, renovated over time, most recently with effect from 2009 by the Treaty of Lisbon. Even though the Court of Justice has enjoyed describing the treaties as a constitutional charter and even though that Court’s rulings have done much to endow the EU legal order with a constitutional character, directly engaging individuals, which is distant from and more sophisticated than orthodox international treaty law, the fact remains that the superstructure of the EU legal order is treaty-based and relatively skeletal. There is plenty of room to make it more coherent—there are plenty of gaps. And the Court has eagerly had resort to general principles of law as one of its techniques to achieve that inflation of the size and shape of the EU legal order. That in turn raises precisely the questions of legitimacy that demand attention in any inquiry into the elaboration of general principles of law, but with added force given that what is at stake is a transnational rather than a national system of law. Some of the Court’s most spectacular and contested rulings have involved its self-set mission to ‘constitutionalise’ the treaties, and general principles have formed and continue to form, even in the light of the rise to prominence of the EU’s Charter of Fundamental Rights which acquired binding force in 2009, a central element in that narrative. The EU treaties are not silent on the matter. Article 340(2) TFEU refers to ‘the general principles common to the laws of the Member States’ as a basis for shaping EU law, which immediately suggests a direct attachment between the EU’s understanding of general principles and that found at national level. However, the nature of this relationship is not spelled out at all in Article 340, and in any event that provision governs only the particular case of the non-contractual liability of the EU’s own institutions. The real interest transcends the Treaty’s hesitant and sectorspecific embrace of general principles of law. The real interest lies in the Court’s embrace of general principles as a source of law in the EU legal order in circumstances which go far beyond any explicit Treaty mandate. The general principles common to the laws of the Member States are one obvious source of inspiration; so too are general principles that may be discovered at international level. This would appear to require the Court to develop a careful comparative preparatory methodology, engaging thorough and textbook-style analysis. This, however, has not at all been a feature of the Court’s case law, at least not openly. This striking judicial disinclination to advertise a thorough inductive methodology does not make it less interesting to look at the interaction of EU law and comparative law in this particular field. In fact, the reverse is true. Those working together in elaborating general

Introduction

 3

principles of EU law tend to be responsive to input from national laws, and the laws of the Member States have no choice but to be responsive to the general principles developed at EU level. This is a rich field, and one too in which a dynamic relationship between the ingenuity of the Court of Justice and the process of treaty revision can be tracked. These are thematically central concerns of this book.

III. The Comparative Challenge of General Principles of Law European Union law is a wonderfully rich place to look for insights into the elucidation and functions of general principles of law. International law more generally also offers an intriguing terrain, moreover one on which arise methodologically similar questions about how such general principles of law are to be discovered. But for comparative lawyers too there is huge appeal in examining the phenomenon of general principles. The interaction between international law, EU law and national laws is interesting; so too the comparison of international and EU approaches with those chosen at national level. But it is perfectly possible (also) to leave the peculiar characteristics of the international and transnational domains to one side, and to prefer instead to examine the treatment of general principles within orthodox national legal orders. How have general principles developed, and why? Why are there general principles in one legal system that are not recognised in others? Why is what is acknowledged in one jurisdiction as a ‘general principle of law’ happily spelled out as a legal ‘rule’ in the legislation or case law of another jurisdiction? Are there in fact general principles of law that are not ‘developed’, ‘found’ or ‘constructed’ by the courts but are, as French lawyers would call it, ‘enshrined’ in legislation? Are there any legal systems that do not have any general principles at all? Is it even conceivable in theory that such a legal system exists? However, perhaps the most important question is one that resembles the focal point of the debate at the level of the European Union: what does the rise of general principles (and sometimes their fall, or at least their containment) reveal about the issues of legitimacy which are so thematically central to any account of the evolution of any legal system? And to what extent do—and should—jurisdictions show a sensitivity to attitudes found elsewhere? Here too, when one explores the judicial choices that have been made over time, one acquires a sense of teasing at the outer edges of expertise. Here too the thematic challenges associated with not simply comparative law as an academic discipline but also comparative law as a basis for judicial reasoning loom large. This book does not provide the answer to all these questions. It does not even attempt to address all the issues raised in this short introduction. Yet we hope that it will help to stimulate thinking on one of the most interesting legal issues in both European and comparative law.

4 

2 General Principles of International Law CATHERINE REDGWELL

I. Introduction While there is no dispute that general principles are a recognised source of ­international law,1 it is only a slight exaggeration to state that there is agreement on little else regarding their ascertainment, content and function. As Sir Robert ­Jennings and Sir Arthur Watts have observed, with some restraint, ‘[t]he meaning of … [general principles] has been the subject of much discussion’.2 In part this is because discussion of general principles is wrapped up in more general arguments about the nature of the sources of international law and of law-making, and the appropriate role for international courts.3 It is also a reflection of the delphic and unfortunate wording of Article 38 of the Statute of the International Court of Justice (ICJ)4 which recognises general principles as one of the sources of international law which is to be applied by the Court in resolving disputes before it.5 It provides guidance on the sources of international law to be applied by the Court as follows: The Court, whose function if to decide in accordance with international law such ­disputes as are submitted to it, shall apply (a) international conventions, whether general or particular, establishing rules expressly recognized by the contesting States; 1  For a useful overview of the literature, see M Kohen and B Schramm, ‘General Principles of Law’ available at oxfordbibliographies.com (last updated 26 August 2013). 2  R Jennings and A Watts, Oppenheim’s International Law, 9th edn, vol 1 (Oxford, Oxford University Press, 2008) 36 (with numerous additional supporting references at n 1). 3  See generally J d’Aspremont and S Besson (eds), The Sources of International Law (Oxford, Oxford University Press, forthcoming); S Besson, ‘Theorizing the Sources of International Law’ in S Besson and J Tasioulas (eds), The Philosophy of International Law (Oxford, Oxford University Press, 2010); and H Thirlway, The Sources of International Law (Oxford, Oxford University Press, 2014). 4 While this contribution will focus on the international court with plenary (but not compulsory) jurisdiction—the ICJ—reference will be made to other international courts and tribunals as appropriate. 5  The ICJ has not attempted to articulate a general theory of sources, which is unsurprising in the light of its ‘function to decide the particular disputes before it, not to elaborate general theories or

6 

Catherine Redgwell (b) international custom, as evidence of general practice accepted as law; (c) the general principles of law recognized by civilized nations; (d) subject to the provisions of Article 59, judicial decisions and the teachings of the most highly qualified publicists of the various nations, as subsidiary means for the determination of the rules of law.6

Apart from the preamble, this wording is identical to the Statute of the Permanent Court of International Justice (PCIJ) of the League of Nations, directing the Court’s predecessor as to the applicable law. As Hernández notes, ‘Article 38 serves as a focal point with some “elementary authority” through which the identification of sources proceeds in international law’ with its inclusion in the Statute of the PCIJ ‘regarded from the outset as a standard reference point as to the sources of international law’.7 Its influence extends beyond the ICJ as other tribunals, albeit clearly not bound by the Statute, have viewed Article 38 as declaratory of the sources of international law.8 For example, arbitral tribunals settling investment disputes under the auspices of the International Centre for Settlement of Investment Disputes (ICSID) are empowered to apply, inter alia, ‘international law’9 which ‘should be understood in the sense given to it by Article 38(1) of the Statute of the [ICJ], allowance being made for the fact that Article 38 was designed to apply to inter-State disputes’.10 In practice Article 38 is frequently invoked as the starting point for a discussion of the general sources of international law.11 Accordingly it forms the departure here for discussion of the role and function of general principles of law in the international legal order.

to decide questions that are not in issue’: M Mendelson, ‘The International Court of Justice and the Sources of International Law’ in V Lowe and M Fitzmaurice (eds), Fifty Years of the International Court of Justice: Essays in Honour of Sir Robert Jennings (Cambridge, Cambridge University Press, 1996) 64. 6 Statute of the International Court of Justice, annexed to the Charter of the United Nations 1 United Nations Treaty Series XVI (emphasis added). 7 G Hernández, The International Court of Justice and the Judicial Function (Oxford, Oxford ­University Press, 2014) 31. 8  For critical analysis of such declaratory function, see Besson and d’Aspremont (n 3). 9  Art 42(1) of the 1965 Washington Convention establishing ICSID 575 United Nations Treaty Series 159. 10  Report of the Executive Directors of the International Bank for Reconstruction and Development, para 40, available at icsid.worldbank.org. Subsequent case law has confirmed this understanding with ICSID tribunals frequently invoking Art 38: see eg the annulment decision of 5 February 2002 in Wena Hotels Ltd v Arab Republic of Egypt, ICSID Case No ARB/98/4, (2002) 41 International Legal Materials 933, 941–2, paras 37–46 and the award of 22 May 2007 in Enron Corporation and Ponderosa Assets, LP v Argentina, ICSID Case No ARB/01/3, para 257, available at italaw.com. On the use by investment arbitration tribunals of the general principles of international law on treaties and other sources, see OK Fauchald, ‘The Legal Reasoning of ICSID Tribunals—An Empirical Analysis’ (2008) 19(2) European Journal of International Law 301–6. 11  It is not, however, an end point because, inter alia, Art 38 of the Statute of the ICJ is (only) a direction to the Court regarding the law to be applied in adjudicating the legal disputes before it: see eg J d’Aspremont, Formalism and the Sources of International Law (Oxford, Oxford University Press, 2011) 149.

General Principles of International Law

 7

II.  The Function of Article 38(1)(c) What was the intended function of this ‘rather vague category’12 of general principles in Article 38(1)(c)? One is to fill gaps in treaty and customary ­ ­international law in ‘response to the need to ensure the completeness of the law’.13 ‘Completeness’ is ‘a normative structure adequate for the needs of the society to which it applies’,14 which, it is argued, was a stage that international law had achieved by 1920 when the Statute of the PCIJ was being drafted. ­Consequently ‘the PCIJ could always give an answer to a legal dispute, precisely because ­international law was regarded as always providing an answer’.15 More particularly, ­Article 38 was designed to avoid the possibility of a non liquet16—its ‘first purpose’ according to some members of the Committee of Jurists charged with drafting Article 38 in 1920.17 As Pellet has observed, while certainly not agreeing on the meaning of the expression ‘general principles of law recognized by civilized nations’, [the Committee] were all in agreement that (i) the first purpose of para 3 was to avoid a non liquet; (ii) without giving to the Court the possibility to legislate. Moreover, they were more concerned with finding an acceptable formula for States than with doctrinal theoretical views.18

A widely accepted view of the purpose of this provision is that it was intended ‘to authorise the Court to apply the general principles of municipal jurisprudence, insofar as they are applicable to relations of states’.19 The use of ‘authorise’ is telling here, for there was a clear concern expressed at the drafting stage (and subsequently) regarding the appropriate judicial role in ascertaining the sources

12  RR Churchill and AV Lowe, The Law of the Sea, 3rd edn (Manchester, Manchester University Press, 1999) 12. 13 A Pellet, ‘Article 38’ in A Zimmermann, C Tomuschat, K Oellers-Frahm and CJ Tams (eds), The Statute of the International Court of Justice: A Commentary, 2nd edn (Oxford, Oxford ­University Press, 2012) 832, para 250. But see Hernández who considers that ‘general principles are more relevant to the systematization of international law than evidence per se of the completeness of the ­international legal order’ (n 7, 262). This point is returned to further below. 14 V Lowe, ‘The Politics of Lawmaking: Are the Method and Character of Norm Creation Changing?’ in M Beyers (ed), The Role of Law in International Politics (Oxford, Oxford University Press, 2000) 211. 15  ibid 211 (emphasis in original). Thus ‘the framing of the PCIJ Statute was undoubtedly a very significant step in the development of the international legal system’ (ibid 212). 16  H Lauterpacht, The Development of International Law by the International Court (London, ­Stevens & Sons, 1958) 166 (‘an ultimate safeguard against the possibility of a non liquet’). He further observes that this safeguard function is ‘unaffected by the relative infrequency of or lack of articulation in its use’ (ibid). 17 See Procès-Verbaux of the Proceedings of the Advisory Committee of Jurists (1920) 318 and 338 (Descamps), 311 (Loder), 312–13 (La Pradelle), 307 and 317 (Hagerup). 18  Pellet (n 13) 832, para 250 (references omitted). 19 Jennings and Watts (n 2) 37. This view is cited with approval by eg J Crawford, Brownlie’s ­Principles of Public International Law, 8th edn (Oxford, Oxford University Press, 2012) 34–35.

8 

Catherine Redgwell

of international law. It is expressive of the view that law-making is the purview (largely of) states, with the Court’s role one of identification, interpretation and application.20 Consequently, as Pellet observes above, Article 38 was intended to prevent the Court from exercising a legislative function.21 This is reflected in the Procès-Verbaux of the Committee of Jurists where one of the architects of the compromise ultimately embedded in Article 38 explained that its wording was going ‘as far as they felt they could on the subject of the liberty to give the judge’,22 and that no further advance was politically feasible. It thus also reflected the desire to achieve a compromise acceptable to states rather than, say, the development of a doctrinal theory of sources.23 By and large this reticence has also been evident in the Court24 where in the exercise of its judicial function the ICJ has been ‘essentially conservative, deciding the case rather than developing the law’: It has seen its task as being to produce a fair and objective reasoned solution to the particular dispute in front of it, constructed in such a way as to maximize the prospects of its judgment being complied with, rather than as being to seize the opportunity to shape or develop the law.25

The extent of this reticence with respect to the identification and application of general principles is explored further below. A final point on the background to the drafting of Article 38 is the tension between positivists and natural lawyers in the Committee of Jurists in the framing of Article 38(1)(c). As Crawford recounts, there was initially no consensus in the Committee, with the Belgian member (Descamps) preferring a natural law approach (‘the rules of international law recognized by the legal conscience of civilised peoples’) while the US member (Root) viewed as unlikely acceptance by governments of a Court authorised to rely ‘on subjective concepts linked with ­principles of justice’.26 The former was indeed viewed by some members of

20  It is of course readily acknowledged that this is a very blurred line: on the international judicial function and law-making, see further Hernández (n 7). 21 Pellet (n 13) 832, para 250; see also Procès-Verbaux 296 (La Pradelle), 309 (Root), 314 (Ricci-Busatti), 316 (Phillimore) and 319 (Hagerup). Yet one of the ‘fathers of international law’ who drew inspiration from Art 38(1)(c), Sir Hersch Lauterpacht, explicitly considers ‘judicial legislation through application of general principles of law’ although he does observe that ‘it would be a mistake to assume that this has been their typical application’ (n 16, 166, emphasis added). 22  Procès-Verbaux, 284 (Lord Phillimore). 23 For an overview of diverse doctrinal views drawing on the Art 38 travaux, see B Vitanyi, ‘Les positions doctrinales concernant le sens de la notion de “principes généraux de droit reconnus par les nations civilises”’ (1982) 86 Revue Générale de Droit International Public 48. 24  See Mendelson (n 5) 64. 25  F Berman, ‘The International Court of Justice as an “Agent” of Legal Development?’ in CJ Tams and J Sloan (eds), The Development of International Law by the International Court of Justice (Oxford, Oxford University Press, 2013) 11–12. 26  Crawford (n 19) 34. Bin Cheng identifies five different positions amongst the ten members of the Committee of Jurists: B Cheng, General Principles of Law as Applied by International Courts and Tribunals (Cambridge, Cambridge University Press, 1953) 10–14.

General Principles of International Law

 9

the Committee as seeming to ‘dangerously … open the door to subjectivity’.27 ­Grounding general principles in the municipal law of states was thus meant as a guarantee that those principles correspond ‘to the dictates of the legal conscience of civilised nations’ whilst at the same time ‘legalizing’ the inherent natural law elements of Article 38(1)(c) through incorporation into the legal systems of States—the essence of the compromise achieved.28

III.  Which ‘General Principles’? So much controversy and uncertainty has surrounded this provision that ­Zimmerman et al, in their magisterial commentary on the Statute of the ICJ, observe that ‘international lawyers have never agreed on the definition of ­general principles mentioned in Article 38’.29 Does it refer to general principles of ­municipal law,30 general principles of international law31 or general principles of legal systems generally?32 Despite its ‘somewhat shadowy operation’,33 a widely accepted view of Article 38(1)(c) is that general principles may be derived not only from municipal law but also from international law.34 Such principles are distinct from principles derived from treaties or from customary international law (Article 38(1)(a) and

27  Pellet (n 13) 832–33, para 251. Perhaps unconsciously providing a flavour of the debates which occurred in the Committee, Mendelson summarily dismisses natural law in discussion of the controversy over the meaning of Art 38(1)(c) because it ‘has no real place in a positivist society’ (n 5, 79). Ellis, on the other hand, detects ‘an approach tinged with natural law thinking’ in the general principles jurisprudence of the PCIJ and ICJ: J Ellis, ‘General Principles and Comparative Law’ (2011) 22 European Journal of International Law 949, 954. 28  Pellet (n 13) 834, para 256. 29  ibid, 832. 30 This is not to be confused with the question of the relationship between international and national law: see generally E Denza, ‘The Relationship between International and National Law’, in MD Evans (ed), International Law, 4th edn (Oxford, Oxford University Press, 2014) 411 and JE Nijman and A Nollkaemper (eds), New Perspectives on the Divide Between National and International Law (Oxford, Oxford University Press, 2007). 31  Simma and Alston have encouraged the use of general principles derived from international law in the human rights context, arguing that such expansion ‘would (and should) then percolate down into domestic fora, instead of being elevated from the domestic level to that of international law by way of analogy’: B Simma and P Alston, ‘The Sources of Human Rights Law: Custom, Jus Cogens and General Principles’ (1992) 12 Australian Yearbook of International Law 82, 102. 32  Mendelson (n 5) 79, in the third instance with particular reference to rules of judicial and arbitral procedure (eg res judicata, estoppel and the equality of parties). 33  M Shahabuddeen, ‘Municipal Law Reasoning in International Law’ in Lowe and Fitzmaurice (n 5) 92. 34  See eg Mosler who identifies three categories: (i) principles originating in domestic or international law; (ii) principles not restricted to any specific legal order; and (iii) principles of legal logic. H Mosler, ‘General Principles of Law’ in R Bernhardt (ed), Encyclopedia of Public International Law, vol 7 (Amsterdam, Elsevier, 1984) 90, 95 and 97–105 (examples).

10 

Catherine Redgwell

(b)), which are sources comprising both rules and principles.35 Admittedly this is far from clear from the language of Article 38. Greater clarity is found in the wording of the 1998 Rome Statute of the International Criminal Court, Article 21(1)(b) of which refers to ‘applicable treaties and the principles and rules of international law, including the established principles of the international law of armed conflict’, while Article 21(1)(c) refers to ‘general principles of law derived by the Court from national laws of legal systems of the world’.36 This wording was designed to clarify and limit judicial application of general principles to those derived from national law, underscoring the importance of domestic criminal law and procedure in the development of international criminal law.37 There is in any event a blurry boundary between principles of general international law broadly understood and ‘general principles of law’ in the Article 38(1)(c) sense. There may also be a difference in function and methodology: if we are speaking of the latter, ie a source of international law under the Statute to be applied by the Court, then as Ellis points out there may be important legal and political reasons for wishing to ensure the consent of a large and varied body of states to an international rule. In a heterogenous society defined by significant power imbalances, in which law-making processes can be described as democratic only in a very loose sense of that word, one has good reason to be wary of general principles as a source of law. At the same time, this source arguably has a very important role to play both in the settlement of individual disputes and in the development of international law.38

This blurry boundary is amply illustrated by the entries in the Max Planck ­Encyclopedia of Public International Law on ‘General International Law (Principles, Rules and Standards)’ by Wolfrum39 and on ‘General Principles of Law’ by Gaja.40 The former addresses principles ‘as a binding legal statement which describes obligations of conduct or obligations to achieve an objective. Generally speaking, 35  Which the Court appears to treat as synonymous: Mendelson (n 5) 80. The Court stated so explicitly in Delimitation of the Maritime Boundary in the Gulf of Maine Area (1984) ICJ Rep 246, 288–90. 36 2187 United Nations Treaty Series 3, in force 1 July 2002. Art 21 is also explicitly hierarchical, unlike Art 38 of the Statute of the ICJ. See also Art 7(1) and (2) of the Convention relative to the Establishment of an International Prize Court (1908) 2 American Journal of International Law Supplement 174. 37  Additionally, Arts 22–33 set forth general principles of criminal law, which were included in the Statute ‘to delimit in great detail any possible exercise of judicial discretion’: WA Schabas, An Introduction to the International Criminal Court, 4th edn (Cambridge, Cambridge University Press, 2011) 71. For discussion of the difficulties of extracting general principles of international criminal law, see Raimondo, who notes, inter alia, that ‘the criminalization of conduct by general principles of law may be a perilous judicial activity as it may jeopardise the nullum crimen nulla poena sine lege principle’: FO Raimondo, General Principles of Law in the Decisions of International Criminal Courts and Tribunals (Leiden, Martinus Nijhoff, 2008) 2. 38  Ellis (n 27) 954 (reference omitted). 39  Max Planck Encyclopedia of Public International Law, available online at opil.ouplaw.com (entry last updated December 2010). Judge Wolfrum has been a judge on the International Tribunal for the Law of the Sea since 1996. 40  ibid (entry last updated May 2013). Judge Gaja has been on the International Court of Justice since 2012.

General Principles of International Law

 11

such obligations are—and this distinguishes principles from rules—described in abstract rather than concrete terms ready for direct application’.41 Their purposes are various, including (i) systemising legal norms (eg the ‘umbrella’ of the principle of sustainable development,42 or the common heritage of mankind principle), or (ii) serving ‘as a tool in the interpretation, application, and, in particular, (iii) in progressive development of international law’.43 At least in theory, Wolfrum argues, these general principles are distinguishable from those in the Article 38(1)(c) sense44 which constitute an ‘international legal source’.45 In practice, as he acknowledges, the demarcation between principles as a legal source and those as a means for the systematisation, interpretation and progressive development of international law is difficult to maintain. As a formal source of international law, and where principles are derived from national law, their ‘transfer’ to the international plane creates an autonomous source of law, along with treaty and customary law. Under the Statute of the International Criminal Court, this source is explicitly subordinate to treaty and custom; in contrast, general principles of law under Article 38(1)(c) of the ICJ Statute are a third source of international law, independent of treaty and custom, and are not supplementary or subsidiary to these other sources.46 Indeed, the emphasis noted above on the safety net function of Article 38(1)(c), as a residual ‘gap-filler’, risks allocating to general principles merely a ‘reserve’ function which is arguably at odds with the stated orthodoxy that the sources of international law articulated in Article 38 are non-hierarchical,47 not to mention overlooking that general principles are in fact integral to the application of treaty and custom.

IV.  Judicial Practice Whatever the debates in legal doctrine, given that the relevant wording of Article 38 is found in the Statute of the PCIJ as well as in the current Statute of the ICJ, it might be reasonable to suppose that the case law of the Court has served to clarify

41 Wolfrum, Encyclopedia

(n 39) para B.1.6 (emphasis omitted). generally T Stephens, ‘Sustainability Discourses in International Courts: What Place for Global Justice?’ in D French (ed), Global Justice and Sustainable Development (Leiden, Martinus Nijhoff, 2010) 39, 51–56. 43 Wolfrum, Encyclopedia (n 39) paras B.1.7 and B.1.20. See also the two separate opinions by Judge Cancado Trindade where he notes the ‘influence’ of one of the constituent principles of sustainable development, inter-generational equity, but in exceedingly cautious language (‘forms part of conventional wisdom’; ‘marks presence’): Pulp Mills on the River Uruguay (Argentina v Uruguay) (2010) ICJ Rep, para 122, and Whaling in the Antarctic (Australia v Japan; New Zealand intervening) (2014) ICJ Rep, para 47. 44  ibid, para B.1.7. 45  ibid, para B.1.20. 46  A point emphasised by Jennings and Watts (n 2) 39–40, and in contrast with para 38(1)(d). 47  This tension is evident in the views expressed eg by Lauterpacht (n 16) 165. 42 See

12 

Catherine Redgwell

the meaning of Article 38(1)(c). Not so: while there has been much consideration by the Court of treaty and customary international law as sources of international law, there has been very little further guidance by the Court regarding general principles of law, which are seldom applied in the cases before it. Moreover, in examining the case law, an immediate difficulty is the tendency of the Court not to stipulate expressly which source is being relied upon (other than where there is the obvious need to identify the treaty source for the principle being relied upon).48 The Court may thus assert that ‘X’ is a ‘well-recognised principle of international law’ or a similarly vague phrase without specifying whether it is referring to a principle of customary international law, a general principle of law, or indeed a combination thereof.49 This process of judicial reasoning, which Mendelson refers to as another ‘gap-filler’, is not expressly mentioned in Article 38(1):50 Even if there is no provision of treaty or customary law directly in point, the Court has been adept at drawing logical deductions, reasoning by analogy or rejecting analogies, discovering implied terms, taking a teleological viewpoint, and so on. Consequently, the need to look for other means of filling gaps is greatly reduced.51

Indeed, such is the prevalence of this approach that Raimondo suggests that an important function of general principles is to reinforce legal reasoning, ie the invocation of general principles in addition to legal rules in order to confirm decisions grounded in such rules.52 In fact there is no instance to date of the ICJ—or its predecessor the PCIJ— basing a decision ‘entirely and directly’ on general principles53 to which they

48  Lauterpacht further observes that where the Court (referring principally to the PCIJ) has actually applied general principles of law, it has not done so eo nomine and by express reference to Art 38 (n 16, 167). 49  Mendelson (n 5) 64. The assertion that ‘The Court has not often applied general principles of law’ must be understood in this context; moreover, Raimondo has pointed out that ‘the absence of explicit reference to comparative law research in the judgments and advisory opinions of the PCIJ and the ICJ does not necessarily imply that they never took account of the comparative legal research offered by parties to the proceedings. This absence does not mean that the ICJ ignores the significance of examining the common denominator of national legal systems’ (n 37, 58, reference omitted). However, this subterranean comparative practice is by definition difficult to detect. 50  Mendelson (n 5) 80. The fact it is unstated may of course simply be a reflection of the fact that it is a very common judicial practice across courts and tribunals at all levels. 51  ibid 81 (references omitted). This reticence has sometimes attracted criticism from within the Court itself. In the Advisory Opinion on the Legality of the Threat or Use of Nuclear Weapons (1996) ICJ Rep 226, the Court did not draw on general principles of international humanitarian law as a basis for decision, an omission referred to by Judge Higgins in her Dissenting Opinion as a failure of judicial process: ibid 584–85. 52  Raimondo (n 37) 7: however, ‘recognition of the principles in question in the domestic sphere does not add to the Court’s duty to apply them as general principles of international law; it only reinforces the “feeling” that such principles are inherently binding’. 53  Thirlway (n 3) 93; Wolfrum, Encyclopedia (n 39) para B.2(a)27 (judgments and advisory opinions); Mendelson (n 5) 79. See further L Erades, ‘General Principles of Law in the Permanent Court of International Justice and in the International Court of Justice’ in M Fitzmaurice and C Flinterman (eds), Interactions between International and Municipal Law (The Hague, TMC Asser Institute, 1993) 134–65; and Cheng (n 26).

General Principles of International Law

 13

have referred ‘with an extreme parsimony’.54 In his authoritative commentary on ­Article 38, Professor Alain Pellet notes that in only four judgments since the PCIJ was first established in 1922 has there been express reference to Article 38(1)(c) in the case law of the ICJ (and the particular principle contended for was rejected in each case).55 There are numerous examples, however, of vague references to ­general principles without explicitly grounding them in Article 38(1)(c).56 Moreover, parties in their pleadings before the Court rarely invoke extensive comparative law arguments in support of a particular general principle. An oftcited exception is the Right of Passage over Indian Territory case where Portugal argued that general principles of law supported its right to passage from the coast to its enclaves of territory. The Pleadings contained detailed analysis of ‘rights of way of necessity’ from municipal law (64 different national laws are cited).57 The majority judgment does not address this argument, though in his Separate Opinion Judge Wellington Koo considered that the underlying principle of recognition of a right of passage was the same, whether of an international enclave or of enclaved land owned by a private individual.58 Another example is the Tunisia– Libya maritime boundary case, where Malta sought unsuccessfully to intervene on the basis of having an interest which might be affected by the decision, as ­provided for in Article 62 of the ICJ Statute—its first invocation before the Court. It relied, inter alia, on a comparative law study of the conditions and modalities of ­intervention in judicial proceedings before various national courts.59 54 

Pellet (n 13) 833, para 253. see Right of Passage case (1960) ICJ Rep 6, 43; South West Africa cases (1966) ICJ Rep 6, 47 para 91; North Sea Continental Shelf case (1969) ICJ Rep 3, 21 para 17; Avena case (2004) ICJ Rep 12, 61 para 127. This number increases slightly if separate and dissenting opinions are accounted for: see eg nn 56 and 58 below. General principles were invoked by the PCIJ, but without express reference to Art 38: see eg Mavrommatis Palestine Concessions (Judgment) (1925) PCIJ Ser A No 5, 30 (contract law principles) and Factory at Chorzów (Germany v Poland) (Merits) (1929) PCIJ Ser A No 17, 29 (the obligation to make reparation for breach of law). 56 eg Corfu Channel case (1949) ICJ Rep 4, 18 (‘This indirect evidence is admitted in all systems of law, and its use is recognized by international decisions’); Barcelona Traction case (1970) ICJ Rep 3, 37 para 50 (‘It is to rules generally accepted by municipal legal systems which recognize the limited company whose capital is represented by shares, and not to the municipal law of a particular State, that international law refers’); Difference Relating to Immunity from Legal Process of a Rapporteur of the Commission on Human Rights case (Advisory Opinion) (1999) ICJ Rep 62, 88 para 63 (It is a ‘generally recognized principle of procedural law’ that ‘questions of immunity are … preliminary issues which must be expeditiously decided in limine litis’); LaGrand (Germany v United States) (2001) ICJ Rep 466, 503 para 103 (the principle that parties to a case must abstain from any measure capable of exercising a prejudicial effect in regard of the execution of a decision to be given, relying on Electricity Company of Sofia and Bulgaria Order of 5 December 1939, PCIJ Series A/B No 79, 199); Sovereignty over Pedra Branca/Pulau Batu Puteh, Middle Rocks and South Ledge (Malaysia/Singapore) case (2008) ICJ Rep 12, 31 para 45 (‘It is a general principle of law, confirmed by the jurisprudence of this Court, that a party which advances a point of fact in support of its claim must establish that fact’). 57  Right of Passage case (1960), Pleadings vol I, 714ff and 858ff. See also the oral pleadings of Mr P Lalive d’Espinay, ibid, vol IV, 516ff. Another example is the Belgian memorial in the Barcelona Traction case (1970), Pleadings vol I, 136–37. 58  (1960) ICJ Rep 54, 66–67, paras 26–27. 59  Case Concerning the Continental Shelf (Tunisia/Libyan Arab Jamahiriya) (1981) Pleadings vol IV, 336ff (M Lalive) and 459ff (‘Etude de théorie générale de la procédure et de droit comparé en vue de l’interprétation de l’article 62 du Statut de la Cour Internationale de Justice’). See also JM Ruda, ‘Intervention before the International Court of Justice’ in Lowe and Fitzmaurice (n 5) 411. 55  ibid:

14 

Catherine Redgwell

Occasionally judges themselves may engage in an exercise of comparison, as Judge Simma did in his Separate Opinion in the Oil Platforms case in exploring the general principle of joint and several liability in tort law:60 I have engaged in some research in comparative law to see whether anything resembling a ‘general principle of law’ within the meaning of Article 38, paragraph 1 (c), of the Statute of the Court can be developed from solutions arrived at in domestic law to come to terms with the problem of multiple tortfeasors. I submit that we find ourselves here in what I would call a textbook situation calling for such an exercise in legal analogy. To state its result forthwith: research into various common law jurisdictions as well as French, Swiss and German tort law indicates that the question has been taken up and solved by these legal systems with a consistency that is striking.61

Following brief analysis of these jurisdictions he concludes: On the basis of the (admittedly modest) study of comparative tort law thus provided, I venture to conclude that the principle of joint-and-several responsibility common to the jurisdictions that I have considered can properly be regarded as a ‘general principle of law’ within the meaning of Article 38, paragraph 1 (c), of the Court’s Statute.62

Some eminent authors have argued that, where derived from domestic law ­analogies, the development of general principles of international law requires the application of a comparative law approach,63 while others question the ready availability of material and cite the enormity of the task.64 Formally, individual judges of the Court are not required to be comparativists—Article 2 of the ICJ Statute stipulates the qualifications of judges with emphasis upon ‘competence in international law’65—yet judges themselves have occasionally observed that requirements for the composition of the Court as a body ensure that key legal

60  Case Concerning Oil Platforms (Islamic Republic of Iran v United States of America) (2003) ICJ Rep 161, 354–57, paras 66–74. 61  ibid 354, para 66. 62  ibid 357, para 74. 63 See eg M Sørensen, ‘Principles de droit international public—Cours général’ (1960-III) 101 Hague Recueil des Cours de l’Academie de Droit International Public 1, 23; M Virally, ‘The Sources of International Law’ in M Sørensen (ed), Manual of Public International Law (New York, St Martin’s Press, 1968) 146. See also Ellis (n 27) 950, lamenting that: ‘The vast scholarship on comparative law seems to have had little influence on most international jurists contemplating general principles of international law.’ 64  Pellet (n 13) 770. 65  Art 2 of the Statute of the ICJ provides in full: ‘The Court shall be composed of a body of independent judges, elected regardless of their nationality from among persons of high moral character, who possess the qualifications required in their respective countries for appointment to the highest judicial offices, or are jurisconsults of recognized competence in international law.’ See also Art 9 of the Statute, which requires electors to bear in mind not only the individual qualifications required ‘but also that in the body as a whole the representation of the main forms of civilization and of the principal legal systems of the world should be assured’. On the appointment and qualification requirements for judges on international courts and tribunals, including the ICJ, see R Mackenzie, CPR Romano, Y Shany and P Sands, Manual on International Courts and Tribunals (Oxford, Oxford University Press, 2010).

General Principles of International Law

 15

families are represented,66 facilitating an intuitive comparative approach.67 Pellet has also argued that ICJ judges need not be comparative lawyers because of the methodology employed in deriving general principles: [A]ll modern domestic laws can be gathered into a few families or systems of law which, insofar as general principles are concerned, are coherent enough to be considered as ‘legal systems’,68 and, since only very general rules are to be taken into consideration in any event, it is enough to ascertain that such principles are present in any (or some) of the laws belonging to these various systems.69

V.  The Methodology for Deriving General Principles Even if we accept this ‘representative legal system’ approach70 what methodology should judges employ in choosing what is exemplary and relevant for such purposes? How should general principles be ascertained? Ellis argues this should be done ‘by applying principles of law common to the major legal systems of the world … [that] are suitable for transposition into the international legal system’.71 This anchors the principles derived with some notion of (implied) state consent.72 The most frequent use of general principles derives from analogies with domestic law

66 See eg the observations by Judge Levi Carneiro on ‘the invocation of “general principles of ordinary international law”’ in his dissenting opinion in the Anglo-Iranian Oil Company case (Preliminary Objections) (1952) ICJ Rep 93, 161, para 14: ‘It is inevitable that every one of us in this Court should retain some trace of his legal education and his former legal activities in his country of origin. This is inevitable, and even justified, because in its composition the Court is to be representative of “the main forms of civilization and of the principal legal systems of the world” (Statute, Article 9), and the Court is to apply “the general principles of law recognized by civilized nations” (Statute, Article 38 (1) (c).’ 67  Pellet has defended identification of general principles as an intuitive exercise in so far as judges of the ICJ are meant as a body to represent ‘the main forms of civilisation and the principal legal systems of the world’ (n 13, 840 para 266); see also discussion by Raimondo (n 37) 57; for critique, Ellis (n 27) 965 (‘One of the reasons for treating general principles with suspicion is that, too often, the legal reasoning used to identify them simply involves elevating legal rules and concepts with which individual judges are familiar from their own legal education and practice to the level of universal truths, sometimes without any reference to a source at all’). 68  His footnote: ‘Ie mainly, civil (or continental) law and common law, from which probably all contemporary municipal laws borrow part of their rules; to this should certainly be added nowadays, at least in some fields, the Islamic system and the specific characters deriving from adherence to socialist doctrines. Cf further David, R/Jauffret-Spinosi, C, Les grands systèmes de droit contemporain (2002).’ 69  Pellet (n 13) 837, para 263. 70  For discussion of the flaws in the commonality or representativeness approach, see Ellis (n 27) 956ff. 71  ibid, 954. 72 Thus addressing one of the critiques of general principles, namely judicial trespass on state ­sovereignty and international law-making.

16 

Catherine Redgwell

concerning rules of procedure, evidence and jurisdiction—ie those areas where state practice is less likely to evolve the applicable rules through treaty or custom. As Lord Phillimore, the author of the compromise text ultimately adopted by the Committee of Jurists in the 1920s, observed: ‘[T]he general principles referred to in [Article 38(1)(c)] were these which were accepted by all nations in foro domestico, such as certain principles of procedure, the principle of good faith, and the principle of res judicata, etc.’73 Additionally, as Judge McNair famously observed in his Separate Opinion in the South West Africa case, ‘International law has recruited, and continues to recruit, many of its rules and institutions from private systems of law’, with the concept of the trust having attained the status of a general principle of law recognised by civilised nations.74 Caution must be exercised, however, in describing this process of recruitment or transposition. It is a process of ‘abstracting the principles from the various municipal rules’, which Jennings and Watts describe as ‘the essentially inductive process of discerning a principle … found to underlie the particular rules in many national legal systems’; ‘the necessary comparative approach to this task will tend to discount the national differences of detail or procedure and to isolate the basic uniform principle which is common to all’.75 The first step, then, is to identify the principle common to representative municipal legal systems (ie those belonging to the main legal systems of the world), and secondly, to distil ‘the essence of the principle’.76 Ellis suggests there is often added a third step, ‘namely, modifying the principle to suit the particularities of international law’.77 Indeed, Jennings and Watts caution that ‘general principles of ­municipal jurisprudence’

73 

Procès-Verbaux Annex No 3, 3. International Status of South West Africa case (1950) ICJ Rep 128, 148. See generally H Lauterpacht, Private Law Sources and Analogies of International Law (London, Hamden Publishing, 1927). More recently Sand has written of ‘vertical transplants’ from national to international law in the context of international environmental law drawing on J Wiener, ‘Something Borrowed for Something Blue: Legal Transplants and the Evolution of Global Environmental Law’ (2001) 27 Ecology Law Quarterly 1295, 1305: PH Sand, ‘The Rise of Public Trusteeship in International Environmental Law’, Global Trust Working Paper Series 03/2013, available at globaltrust.tau.ac.il/publications. 75  Jennings and Watts (n 2) 37, n 3. See also H Waldock, ‘General Course on Public International Law’ (1962-II) 106 Hague Recueil des Cours de l’Academie de Droit International Public 65, where he notes that ‘it is … a question of finding in the private law institutions indications of legal policy and principles appropriate to the solution of the international problem in hand. It is not the concrete manifestations of the principle in different national systems—which are anyhow likely to vary—but the general concept of law underlying them that the international judge is entitled to apply under paragraph (c) [of Article 38].’ 76  Ellis (n 27) 954. Raimondo points to a two-stage process: (i) the vertical move of abstracting legal rules from national legal systems and deriving therefrom the underlying legal principle; and (ii) the horizontal move of comparing national legal systems to verify that the generality of states recognise the principle thus abstracted. The latter requires recognition by the main legal families of the world. Raimondo (n 37) 2–3. See also O Elias and C Lim, ‘General Principles of Law, “Soft” Law and the ­Identification of international law’ (1997) 28 Netherlands International Law Review 3. 77  Ellis (n 27) 954, echoing the ‘discounting’ step suggested by Jennings and Watts in the passage cited at n 75. 74 

General Principles of International Law

 17

are in the municipal sphere applied against a background of national laws and procedures. Unless there is some sufficient counterpart to them in the international sphere, or sufficient allowance is made for them in abstracting the principles from the various municipal rules, the operation of the principles as a source of particular rules of international law will be distorted.78

Charlesworth and Chinkin, in their seminal work The Boundaries of International Law, raise a further serious concern with the use of general principles of domestic law in the international legal system [which] is that they will simply transpose the problems of the former into the latter. … Whether the national system is based on a religious, capitalist or socialist ideology, the legal system reinforces the power structures of the male political elite.79

What is not suggested in international legal doctrine and jurisprudence is that any particular national law articulation constitutes a general principle of law. As the Permanent Court of Arbitration observed in the Norwegian Shipowners’ Claims case, a reference to equity in the Special Agreement between the parties meant ‘general principles of justice as distinguished from any particular system of jurisprudence of the municipal law of the State’.80 Similarly, in his Separate Opinion in the Case Concerning Certain Phosphate Lands in Nauru (Preliminary Objections), Judge Shahabuddeen pointed out that:81 A United Nations Trusteeship must not, of course, be confused with a trust as understood in any specific system of municipal law; but, used with discretion, the principles relating to the latter are not unhelpful in elucidating the nature of the former. As Judge McNair said, in relation to Mandates, it ‘is primarily from the principles of the trust that help can be obtained on the side of private law’.82

78  Jennings and Watts (n 2) 37 (references omitted). Ellis analyses at length why ‘attempts to distil rules to their essence are probably misguided’ (n 27, 967). There is also the risk that such processes of modification, discounting or ‘denationalisation’ of the principle may render it so vague and general as to be of little assistance to the Court. 79 H Charlesworth and C Chinkin, The Boundaries of International Law: A Feminist Analysis (Manchester, Manchester University Press, 2000) 79. 80  Norwegian Shipowners’ Claims (Norway v USA) Permanent Court of Arbitration (1922) 1 RIAA 307, 331. See discussion in A Gourgourinis, Equity and Equitable Principles in the World Trade Organization (London, Routledge Publishing, 2016) 18. See generally M Akehurst, ‘Equity and General Principles of Law’ (1976) 25 ICLQ 801; V Lowe, ‘The Role of Equity in International Law’ (1988–89) 12 Australian Year Book of International Law 73. However, ‘the routes of entry of equity’ in international law may be via other sources of international law, eg through treaty provision or custom: Case Concerning Maritime Delimitation in the Area between Greenland and Jan Mayen (Denmark v Norway), Separate Opinion of Judge Weeramantry, (1993) ICJ Rep 211, paras 74–102. 81  (1992) ICJ Rep 270, 285. He states that ‘[i]f domestic analogies are to be considered, the most likely area lies within the broad principles of the law of trust in English law and of cognate institutions in other systems’ (ibid). 82  ibid. He cites as authority the South West Africa case (1950) 151 and the Namibia case (1971) ICJ Rep 16, 214 (Judge de Castro). Judge Shahabuddeen goes on to cite Halsbury’s as authority for the applicable rule in English trust law regarding the joint and several liability of several trustees for breach of trust (ibid).

18 

Catherine Redgwell

This reference to Judge McNair recalls his admonition in the Court’s 1950 ­Advisory Opinion in the South West Africa case83 not to import ‘lock, stock and barrel’ ­private law institutions into international law. In his view this would be inconsistent with the application of general principles of law which requires that international tribunals have ‘regard to any features or terminology which are reminiscent of the rules and institutions of private law as an indication of policy and principles rather than as directly importing these rules and institutions’.84 Sometimes the Court will eschew reference to rules and principles altogether and merely refer to a ‘concept’, the most notorious case being the concept of sustainable development in the Gabcikovo-Nagymaros case where the Court interpreted a bilateral treaty for the construction and operation of a dam in the light of subsequent developments in international environmental law. In particular it considered the emergence of the concept of sustainable development, observing that ‘the Treaty is not static, and is open to adapt to emerging norms of international law’.85 The approach of the Court was thus grounded in the evolutionary language of the treaty text itself, and it stopped short of recognising sustainable development as a principle of customary international law or as a general principle of law. In contrast, in his Separate Opinion Judge Weeramantry went further, commencing with an historical overview of its significance in Sri Lankan domestic law, before turning to the international law context.86

VI. Conclusion It was noted at the outset that discussion of the role of general principles in ­international law inevitably gives rise to questions about the nature of public 83 

South West Africa (1950) 128. 148. This quote from Judge McNair is one of—if not the—most frequently cited passage from the ICJ on general principles, but (a) it was in a Separate Opinion and (b) in the judgment itself, the Court declined to apply the concept of mandate as a general principle. See further the summary in G Fitzmaurice, The Law and Procedure of the International Court of Justice, vol 1 (Cambridge, Grotius Publications, 1986) 10–11. Jennings and Watts also question the distinction drawn by McNair between rules and principles in this regard, on the basis that at least in this connection the applicable rules and principles are probably not essentially different (n 2, 316, n 7). Clearly, however, they share his view of the appropriate methodology. 85  Case concerning the Gabcikovo/Nagymaros Project (Hungary/Slovakia) (1996) ICJ Rep 76–80. See also Iron Rhine Arbitration (Belgium/Netherlands) (2005) XXVII RIAA 35 para 80; Pulp Mills on the River Uruguay (Argentina v Uruguay) (2010) ICJ Rep 14 para 205; and generally, E Bjorge, The Evolutionary Interpretation of Treaties (Oxford, Oxford University Press, 2014). See also n 43 above (inter-generational equity). 86  (1996) ICJ Rep 88; for recent, less ambiguous, invocation, see the Kishenganga Arbitration (Partial Award), PCA (2013) para 452. For analysis of its interpretative function as an ‘interstitial norm’, see AV Lowe, ‘Sustainable Development and Unsustainable Arguments’ in AE Boyle and D Freestone (eds), International Law and Sustainable Development (Oxford, Oxford University Press, 1999) 234; and on Judge Weeramantry, see D French, ‘The Heroic Undertaking? The Separate and Dissenting Opinions of Judge Weeramantry during his Time on the Bench of the International Court of Justice’ (2006) 11 Asian Yearbook of International Law 35. 84  ibid,

General Principles of International Law

 19

international law and its relationship with municipal law, its ‘completeness’, and the appropriate role for international courts and tribunals in the identification, interpretation and application of international law. There is no doubt that in the 1920s the function of general principles was seen principally as gap-filling, and to avoid a non liquet, the very antithesis of the ‘completeness’ of international law.87 Yet in the Nuclear Weapons Advisory Opinion, it was precisely the ‘failure of judicial process’ in not exploring general principles of international humanitarian law that led Judge Higgins to criticise the majority for delivering a non liquet in paragraph 2E of the dispositif.88 Her view—and one clearly hoped for by the Committee of Jurists drafting the Statute of the PCIJ in the 1920s in guarding against it—is that the concept of non liquet forms no part of the Court’s jurisprudence:89 while ‘there is [not] always a clear and specific legal rule readily applicable to every international situation … every international situation is capable of being determined as a matter of law’.90 It is clear that, however much there may be room for improvement methodologically in the identification of general principles,91 and whether derived from municipal or international law, judicial reasoning continues to play a central role in their identification, interpretation and application.92

87 ‘Non liquet is a declaration of incompetence’—‘an acknowledgement that the matter should be regulated by law, but that there is no identifiable rule that can be applied to the facts before the ­tribunal’: Lowe (n 14) 210. See further J Stone, ‘Non Liquet and the Function of Law in the I­ nternational ­Community’ (1959) XXXV British Year Book of International Law 124. 88 Higgins, Nuclear Weapons 583, para 2; 584, para 7; 590, para 30. 89  ibid, 591, para 36. 90 ibid, 592, para 39, citing Jennings and Watts (n 2) 12–13 (emphasis in original). Thus, she observes, ‘At no point in its Opinion does the Court engage in the task that is surely at the heart of the question asked: the systematic application of the relevant law to the use or threat of nuclear weapons. It reaches its conclusions without the benefit of detailed analysis. An essential step in the judicial process—that of legal reasoning—has been omitted’ (ibid, 584, para 9). 91 As is already evident from the preceding analysis, from the perspective of comparative law ­scholarship Ellis considers ‘[t]he prevailing international law methodology for identifying principles in municipal law and transforming them into general principles of law is highly unsatisfactory’ (n 27, 970). 92  Similarly, Raimondo (n 37) 259.

20 

3 From Myth to Reality: The EU’s ‘New Legal Order’ and the Place of General Principles Within It STEPHEN WEATHERILL

I. Introduction No ruling of the Court of Justice is more significant than those that it delivered in February 1963 in Van Gend en Loos and in July 1964 in Costa v ENEL. The purpose of this chapter is to reflect on the astonishing chutzpah which embellishes those two landmark judgments and, most of all, to argue that the Court’s perceived and sometimes contested creativity in more recent times, especially but not only in the development of general principles, is the direct consequence of the fundamentally important choices it made in these two cases over fifty years ago. The conversion of EU law from an order founded on an international treaty into one which also insists on the protection of individual rights before national courts generated a momentum. The Court was induced to develop general principles in order to fill gaps in the fabric of the EU legal order precisely because its rulings in Van Gend en Loos and in Costa v ENEL exposed the pressing need for a pattern of legal protection of the individual which was denser and more intricate than would ordinarily be demanded under international law.

II.  Van Gend en Loos: The Principle of Direct Effect The EU treaties do not stipulate how, if at all, national courts should apply rules sourced at EU level. The Court has intervened. It treats EU law as capable of direct effect, which means that rules of sufficient clarity and precision fall to be applied on an everyday basis by the national courts of the Member States. This was first

22 

Stephen Weatherill

established in 1963, only a few years after the entry into force of the Treaty of Rome, in the landmark ruling in Van Gend en Loos.1 The dispute arose as a result of the imposition of a customs duty on the importation of a chemical, ureaformaldehyde, from Germany into the Netherlands. This was a straightforward breach of EU law governing the free movement of goods: the relevant provision today is Article 30 TFEU. The constitutional question concerned how, if at all, this breach should be the subject of sanction at national level. Van Gend en Loos claimed reimbursement of the sum before the courts of the Netherlands. A preliminary reference was made by the Tariefcommissie, a Dutch administrative tribunal having final jurisdiction in revenue cases, asking whether the Treaty could assist a private litigant before a national court. The Belgian government proposed to the Court of Justice that it should rule that the question of how EU law is treated within national legal orders is a matter of national constitutional law, not EU law. This would align EU law with international treaties generally. It would also mean inter alia that its effects at national level would accordingly vary among the Member States. The Dutch government pressed on the Court that a violation of the Treaty by a Member State should be the subject of control exercised by the Commission pursuant to the infringement procedure created by the Treaty which is today Article 258 TFEU. It should not, it was argued, also be apt to be brought before the Court via the preliminary reference procedure. Advocate General Roemer expressed a similar view in his Opinion in the case. This approach is to insist on the separation of EU and national procedures and on the alignment of EU and international law. How different EU law would look had this conservative attitude found favour with the Court in Van Gend en Loos. But it did not. The Court was in revolutionary mood. The Court in Van Gend en Loos found EU law capable of direct effect, which means it may create legally enforceable rights before national courts. So how EU law should be handled at national level was the subjection of direction pursuant to EU law. It was not left in the hands of national law. This finding derives from the nature, not the text, of the Treaty. In its ruling the Court carefully sought to place distance between EU law (as it would become) and orthodox international Treaty law. It did this by emphasising the Treaty of Rome’s objective. This, it announced, focuses on the project of economic integration, ‘the functioning of which is of direct concern to interested parties in the Community’, which ‘implies that this Treaty is more than an agreement which merely creates mutual obligations between the contracting states’. It relied too on ‘the preamble to the Treaty which refers not only to governments but to peoples’ and on the ‘establishment of institutions endowed with sovereign rights, the exercise of which affects Member States and also their citizens’. It added that the nationals of the Member States ‘cooperate in the functioning of this Community through the intermediary of the European Parliament and the Economic and

1 

Case 26/62 Van Gend en Loos [1963] ECR 1.

From Myth to Reality

 23

Social Committee’. Moreover, the preliminary reference procedure ‘confirms that the states have acknowledged that Community law has an authority which can be invoked by their nationals before those courts and tribunals’. Some of these reasons have more force than others. The existence of the preliminary reference procedure does indeed suggest a more intense relationship between the Court of Justice and national courts than one would find in orthodox international treaty regimes, and it is a plausible basis for extrapolating a claim that EU law is designed to assert a correspondingly deeper penetration into national legal orders than would be characteristic of a conventional Treaty regime. By contrast, the reference to the Economic and Social Committee, a body pitifully devoid of profile or legitimacy, is shallow, feeble and opportunistic. But overall the point is that the Court draws from the structure and content of the Treaty itself a claim that this is more than an arrangement struck between states which operates solely on the international plane. This then led the Court to its transformative paragraph which merits citation in full: The conclusion to be drawn from this is that the Community constitutes a new legal order of international law for the benefit of which the states have limited their sovereign rights, albeit within limited fields, and the subjects of which comprise not only Member States but also their nationals. Independently of the legislation of Member States, Community law therefore not only imposes obligations on individuals but is also intended to confer upon them rights which become part of their legal heritage. These rights arise not only where they are expressly granted by the Treaty, but also by reason of obligations which the Treaty imposes in a clearly defined way upon individuals as well as upon the Member States and upon the institutions of the Community.

EU law, then, is capable of direct effect. And the Court not only found the unconditional provision forbidding customs duties which was at stake in Van Gend en Loos to be directly effective, it subsequently went much further and extended direct effect even to provisions which might have appeared at first glance to be too conditional in nature to attract direct effect. Direct effect quickly emerged as normal, not exceptional. Reyners v Belgian State serves as a well-known and relatively early example of this trend.2 A Dutch national resident in Belgium held a Belgian legal qualification. He was refused admission as an avocat in Belgium because rules restricted access to that profession to Belgians. He argued that he had a right under EU law to establish himself in Belgium—today this is the province of Article 49 TFEU. The response was that this Treaty provision could not be applied directly by national courts because its implementation was dependent on the subsequent adoption of specific Directives regulating the professions. But this restrictive approach found no favour before the Court of Justice. The Court observed that Reyners was the victim of discrimination on grounds of nationality: he was being treated less favourably than would have been an identically qualified 2 

Case 2/74 [1974] ECR 631.

24 

Stephen Weatherill

Belgian. The subsequent adoption of secondary legislation facilitates free movement and adds supporting detail, but the rule of equal treatment on the basis of nationality demanded by the Treaty is itself enforceable by a private individual. It is directly effective. This is an approach which promotes judicial condemnation of obstructive national rules at the expense of the discretion of the political process to choose the pace of liberalisation of the market by choosing among possible legislative solutions. The landmark ruling in Cassis de Dijon, which concerned goods not people, is of this type too.3 The Court declared that (what is now) Article 34 TFEU applies to non-discriminatory national technical standards governing the composition of products and there is no need to await the attention of the legislative process in order to push forward the project of building an internal market for the EU. In similar vein, but in the field of corporate mobility, the Court in Centros found that the refusal of the Danish authorities to register a branch of Centros Ltd, a UKregistered company, in Denmark on the basis that it did not comply with Danish rules concerning, in particular, the paying-up of minimum capital for a company was a violation of EU free movement law.4 Interpreting the Treaty in this way to protect free movement by corporate entities, it noted with supreme disdain that it was ‘of little consequence’ that company law was not harmonised across the EU.5 Some Treaty provisions are capable of direct effect against private parties too—ie they are capable of horizontal direct effect. This is true of Articles 101 and 102 TFEU, the Treaty competition rules that forbid particular types of anticompetitive practices and abusive conduct by an economically dominant firm. The Court’s predilection for an extended understanding of the penetration of the national legal order by EU rules in the name of direct effect is especially vividly illustrated by Defrenne v SABENA.6 Although (what is now) Article 157 TFEU directs that ‘the principle of equal pay for male and female workers for equal work’ shall be ensured by each Member State, the Court decided in Defrenne that it is directly effective even in proceedings between individuals. So this core element of EU social policy can be pursued and promoted through private litigation before national courts. Directives, by contrast, appear to be different. According to Article 288 TFEU, the Directive, as an instrument of EU secondary legislation, requires implementation at national level. Since the Member State chooses the applicable form and methods for converting the Directive into domestic law, there would appear to be no unconditional rule at EU level. Nevertheless, the Court has found Directives capable of direct effect in circumstances where enforcement against public authorities is at stake, under a logic which insists that the non-implementing state shall not be able to skip free of the consequences of its default.7 In this way national judges go some

3 

Case 120/78 Rewe-Zentrale AG v Bundesmonopolverwaltung für Branntwein [1979] ECR 649. Case C-212/97 Centros [1999] ECR I-1459. Para 28. 6  Case 43/75 Defrenne v SABENA [1976] ECR 455. 7  eg Case 152/84 Marshall [1986] ECR 723. 4  5 

From Myth to Reality

 25

way to correcting the failings of the political or administrative process within the Member States: direct effect makes obligations undertaken pursuant to Directives harder for states to evade. But the Court limits this logic to enforcement against the public authorities of the defaulting state. It has consistently ruled against the horizontal direct effect of Directives—they may not impose obligations directly on private parties in proceedings before national courts.8 The gap caused by the Court’s refusal to accept that Directives are capable of horizontal direct effect is real and unfortunate, but the overall message is captured by what the Court has accepted, not what it has refused. Its acceptance and development of EU law’s capacity to attract direct effect makes EU law much harder to evade than orthodox international law. It institutes a system of dual vigilance, whereby the enforcement mechanisms foreseen explicitly by the Treaty, most prominently the infringement procedure initiated by the Commission which is set out in Article 258 TFEU, is supplemented by the role of national courts which are required to apply directly effective provisions of EU law in order to hold defaulting parties to the obligations specified in the Treaties and supporting secondary legislation. But this understanding of EU law is not simply instrumental. It is more than a means to strengthen its policing. It is in addition a discourse of individual rights which serves as a fiercely important legitimating feature of EU law.9

III.  Costa v ENEL: The Principle of Supremacy (or, Perhaps Better, Primacy) The EU treaties have nothing to say about the question of whether EU or national law prevails in the event of conflict. Here too the Court has intervened. The Court has long insisted that EU law is supreme. This means that it applies in preference to national law in the event of conflict. This was first established in the early years of the development of the EU in Costa v ENEL.10 The dispute concerned a trivially small amount of money. But it allowed the Court to assert the principle of central and enduring importance in EU law that it prevails over national law in the event of conflict.

8  eg Case C-91/92 Faccini Dori v Recreb [1994] ECR I-3325; Case C-201/02 Wells [2004] ECR I-723; Case C-192/94 El Corte Inglés v Cristina Blasquez Rivero [1996] ECR I-1281; Case C-97/96 Verband Deutscher Daihatsu Händler [1997] ECR I-6843; Case C-80/06 Carp Snc v Ecorad [2007] ECR I-4473; Case C-555/07 Kücükdeveci [2010] ECR I-365; Case C-508/14 Česky telekomunikačni judgment of 6 October 2015. 9  eg P Pescatore, ‘The Doctrine of Direct Effect: An Infant Disease of Community Law’ (1983) 8 European Law Review 155; D Halberstam, ‘The Bride of Messina: Constitutionalism and Democracy in Europe’ (2005) 30 European Law Review 775; D Chalmers and L Barroso, ‘What Van Gend en Loos Stands For’ (2014) 12 International Journal of Constitutional Law 105. 10  Case 6/64 [1964] ECR 585.

26 

Stephen Weatherill

The reasoning in Costa v ENEL has much in common with that in Van Gend en Loos. The Court carefully constructed its claim on the foundation of its understanding of the nature of the system established by the (then) EEC Treaty. The Treaty, it declared in Costa v ENEL, ‘has created its own legal system which, on the entry into force of the Treaty, became an integral part of the legal systems of the Member States and which their courts are bound to apply’. It explicitly pitched this as a ‘contrast with ordinary international treaties’. It certainly is. The orthodox approach to international treaties holds that their effects within national legal orders, if any, derive from the choices made by national constitutional law. In Costa v ENEL the Court continued: By creating a Community of unlimited duration, having its own institutions, its own personality, its own legal capacity and capacity of representation on the international plane and, more particularly, real powers stemming from a limitation of sovereignty or a transfer of powers from the States to the Community, the Member States have limited their sovereign rights, albeit within limited fields, and have thus created a body of law which binds both their nationals and themselves.

This then led the Court to claim that EU law must enjoy supremacy over national law in the event of conflict. It drew on ‘the terms and the spirit of the Treaty’. Given acceptance of obligations by the Member States on a basis of reciprocity, EU law cannot be blocked by subsequent domestic laws ‘without jeopardizing the attainment of the objectives of the Treaty’. The Court picked out a number of provisions that explicitly grant Member States the right to act unilaterally: otherwise, it reasoned, no such permission exists, for then ‘[t]he obligations undertaken under the Treaty establishing the Community would not be unconditional, but merely contingent’. So it concluded: It follows from all these observations that the law stemming from the Treaty, an independent source of law, could not, because of its special and original nature, be overridden by domestic legal provisions, however framed, without being deprived of its character as Community law and without the legal basis of the Community itself being called into question. The transfer by the States from their domestic legal system to the Community legal system of the rights and obligations arising under the Treaty carries with it a permanent limitation of their sovereign rights, against which a subsequent unilateral act incompatible with the concept of the Community cannot prevail.

The Court’s claim that EU law has and must have priority over ‘domestic legal provisions, however framed’ is astonishingly bold. But it has stuck remorselessly to this logic. The Court’s stance that means in principle that any measure of Union law, however trivial, overrides even national constitutionally protected rights, including fundamental rights. And the Court has insisted that it really means this! Internationale Handelsgesellschaft provided its first confirmation.11 It was alleged that

11 

Case 11/70 [1970] ECR 1125.

From Myth to Reality

 27

the EU’s regulation of the market for cereals contravened German rules dedicated to the protection of property rights. The Court declared firmly that to permit what we know today as EU law to be subjected to review pursuant to conditions of validity set by national law would damage its ‘uniformity and efficacy’.12 It is therefore only against the standards of EU law, not national law, that the validity of EU acts falls to be measured. This ruling carries impeccable logical force. If it were possible for national legal orders to place selected matters above the demands of EU law, then the inevitable consequence would be patterns of variation in the scope and effect of EU law across the twenty-eight Member States. This is exactly what the Court in its ruling in Costa v ENEL wished to treat as incompatible with and therefore disallowed by the nature of the legal order established by the Treaty. Over forty years later the Court thumped home its commitment to EU law’s supremacy even in the face of apparently conflicting national constitutional law. This occurred in the landmark ruling in Stefano Melloni.13 The Court was asked whether a Member State (in casu, Spain) could make the execution of a European arrest warrant subject to conditions which were intended to secure compliance with fundamental rights recognised by the Spanish constitution which were by contrast not permitted under the applicable EU measure.14 The reasoning deployed in Costa v ENEL and in Internationale Handelsgesellschaft would plainly generate a negative conclusion: Spanish courts could not add domestic constraints to the application of EU law. There was, however, a twist. Article 53 of the EU’s Charter of Fundamental Rights, which had been granted binding with effect from 2009 as a result of the entry into force of the Treaty of Lisbon, declares that nothing in the Charter shall be interpreted as restricting human rights and fundamental freedoms recognised by Union law, international law and international agreements and by Member States’ constitutions. The question in Stefano Melloni was whether this serves to moderate the clean lines of supremacy by creating space for human rights and fundamental freedoms recognised by Member States’ constitutions even in circumstances where they conflict with the demands of EU law. The Court’s answer was clear: nothing has changed! The Court ruled that that national law, even national constitutional law, may not be used as a basis for the review of the validity of EU acts, for otherwise this would ‘undermine the principle of the primacy of EU law’.15 The EU measure could be tested only against the standards set by EU law. And, an overt declaration of continuity, the Court cited Internationale Handelsgesellschaft, decided over forty years earlier.16 Both Internationale Handelsgesellschaft and Stefano Melloni are borne along by a wave of logic. Of course EU law must prevail over national law, including national constitutional law, for otherwise the very spirit of the EU legal order as a uniform

12 

Para 3. Case C-399/11 judgment of 26 February 2013. EU Framework Decision 2002/584 [2002] OJ L190/1. 15  Para 58. 16  Para 59. 13  14 

28 

Stephen Weatherill

system would be violated. And yet there are values other than uniformity that should animate European legal culture. Protection of fundamental rights stands high on any list of European values. And the Court does not neglect this. Its point is not to deny that fundamental rights are of central importance in EU law: instead its point is to deny that national conceptions of fundamental rights may override EU law and instead to locate at EU level the commitment to protection of fundamental rights which must accompany the claim to EU law’s supremacy if it is to deserve to be treated as legitimate. The Court, and latterly the Treaty and the Charter, have been energetic in promoting respect for fundamental rights as foundational values within the EU legal order. No measure adopted by the EU will survive judicial scrutiny where it fails to comply with the conditions of legality imposed by EU law, of which compliance with fundamental rights is one. So, for example, in Digital Rights Ireland the Court condemned Directive 2006/24 on data retention as ‘a wide-ranging and particularly serious interference’ with Articles 7 and 8 of the Charter on privacy and data protection.17 Given its sweeping breadth unsupported by adequate safeguards, the measure could not stand. EU law purports to be fundamental rights-compliant. One might question whether this manifests a sincere commitment by the EU and its Court to values beyond the market or whether instead it is provoked by shrewd awareness that a failure to enrich the EU legal order in this way will probably induce national courts to assert the need to maintain protection of the type mandated by national legal orders: this would amount to a rebellion against the claim to supremacy which the Court has no coercive powers to put down. Both narratives probably contain strains of truth. Supremacy does not mean that the offending national rule is exterminated. EU law does not require that a national court declare the rule invalid. Instead the national judge’s job is to ‘disapply that rule’.18 Primacy rather than supremacy, is the word normally preferred in the Court’s judgments, primauté being the word in French. Use of ‘supremacy’ is probably now so embedded in English language writing that it cannot be displaced. But in fact that is a pity. It is not simply a matter of semantics. There is a good reason to favour use of the term primacy over supremacy in order to capture the notion that EU law controls the applicability but not the validity of national law.19 The national law, since it is not invalid for want of conformity with EU law, may still remain in place and it could be applied in, for example, a purely domestic context with no connection to EU law—although this would doubtless lead to a plea of discriminatory treatment by the purely domestic litigant deprived of the judicial protection available to an EU law-fuelled

17 

Joined Cases C-293/12 and C-594/12 Digital Rights Ireland et al judgment of 8 April 2014. Eg Cases C-10/97 to C-22/97 Ministero delle Finanze v IN.CO.GE [1998] ECR I-6307 para 21; also Case C-314/08 Krzysztof Filipciak [2009] ECR I-11049. 19  See M Claes, ‘The Primacy of EU Law in European and National Law’ in A Arnull and D Chalmers (eds), The Oxford Handbook of European Union Law (Oxford, Oxford University Press, 2015) ch 8, 182–85. 18 

From Myth to Reality

 29

litigant. This is more than sophistry: it asserts preservation of the jurisdiction of the Member State, rather than of the Court of Justice, to decide on the constitutional validity of national law and so too it asserts a separation between the EU legal order and those of the Member States.

IV.  What Really Is ‘New’? The sheer audacity of the Court’s claims in the foundational rulings of Van Gend en Loos and Costa v ENEL should be appreciated. How could this possibly be the new legal order which the Court hailed? It simply was not! It was in fact a perfectly familiar legal order—it was one created by the ratification of an international treaty by (at the time) six Member States. The EU, as an international organisation, is created, defined and limited by its treaties, and this root is most conspicuous at times of treaty revision, when the orthodox patterns of agreement, signature and ratification pursuant to domestic constitutional stipulations surround the process. The novelty of Van Gend en Loos and Costa v ENEL therefore lies somewhere else. In its two foundationally important decisions of the early 1960s the Court was not attempting to deny that EU law is international law. Instead it was attempting to portray EU law (as it has now become) also as something extra—something new. The novelty lies in the Court’s claims that EU shall be treated by national courts as supreme and as directly effective. These concepts are not unknown outside EU law. International law is supreme over national law in the sense that no state may plead provisions of its domestic legal order as a basis for excusing infraction of obligations existing on the international plane.20 Moreover, international treaties may be regarded as self-executing: they may, depending on the perceived intentions of their parties, be treated as apt for application by courts of participating states. So the extent of the novelty engaged by distancing EU law from orthodox international law deserves a cautious assessment.21 But there is calculated distancing: Van Gend en Loos and Costa v ENEL did contain within them genuine novelty. The orthodox approach under international law is to leave it to domestic law to decide how, if at all, international obligations shall be applied by domestic courts. So international law will not excuse a state violation on the basis that it was mandated by domestic law, but this is a principle applicable on the international plane. International law makes no demands of national judges in such circumstances. Similarly the embrace of an EU treaty as self-executing is a choice made as a matter of national law. International law does not at all dictate that treaty provisions shall

20 

See now Art 27 of the Vienna Convention of the Law of Treaties. See eg D Wyatt, ‘New Legal Order, or Old?’ (1982) 7 European Law Review 147; O Spiermann, ‘The Other Side of the Story: An Unpopular Essay on the Making of the European Community Legal Order’ (1999) 10 European Journal of International Law 763. 21 

30 

Stephen Weatherill

be applied by national courts. It concerns itself with obligations undertaken on the international plane. By direct and deliberate contrast in Van Gend en Loos and in Costa v ENEL the Court of Justice was making exactly these extra demands. It was stipulating how (and why) national judges should apply EU law. It was insisting that EU law embeds requirements about its reception into national legal orders— requirements that are carried through by national judges. The Court was dictating not simply obligations on the international plane but also obligations that directly engage individuals and which fall to be vindicated in proceedings before national courts. This novelty has been embroidered over time in at least three distinct ways. First, the duty of consistent interpretation requires a national court called on to interpret national law ‘to do so, as far as possible, in the light of the wording and the purpose of the Directive in order to achieve the result pursued by the latter’,22 so where a Member State has not implemented a Directive at all, or has done so inadequately, the Directive still acquires a voice in relevant national judicial proceedings. Second, the insistence that remedies available at national level shall be effective, now codified in Article 19(1) TEU and Article 47 of the Charter but the subject of development by the Court in case law beginning as far back as 1976,23 serves to prevent national law operating to ‘frustrate’ the application of EU law,24 and accordingly robs national systems of autonomy in matters of procedure and remedies.25 Third, pertaining to one particular remedy, the Court has fashioned a doctrine of state liability to compensate those suffering loss caused by a sufficiently serious infraction of EU law which is required as a matter of EU law but handled by national courts. This was famously shaped first of all in relation to failure to implement a Directive26 and subsequently has been elaborated to cover violation of primary law,27 including by private parties in the field of competition law28 and even in cases where judicial malpractice is the basis for state responsibility.29 All of this involves reliance on EU law to make—increasingly specific and concrete—demands of national judiciaries. This is a process of constitutionalisation which the Court initially situated within the framework of the obligation of

22  eg Case C-106/89 Marleasing SA v CIA [1990] ECR I-4135 para 8. See also eg Case C-62/00 Marks and Spencer [2002] ECR I-6325; Case C-42/11 Lopes Da Silva Jorge judgment of 5 September 2012; Cases C-397/01 to C-403/01 Pfeiffer [2004] ECR I-8835; Case C-212/04 Adeneler [2006] ECR I-6057. 23  Case 45/76 Comet [1976] ECR 2043. See subsequently eg Case C-432/05 Unibet v Justitiekanslern [2007] ECR I-2271. 24  eg Case C-119/05 Lucchini [2007] ECR I-6199 para 59. 25  See eg M Bobek, ‘Why There is No Principle of Procedural Autonomy of the Member States’ in B de Witte and H Micklitz (eds), The European Court of Justice and Autonomy of the Member States (Cambridge, Intersentia, 2011). 26  Cases C-6/90 and C-9/90 Francovich and Others v Italian State [1991] ECR I-5357. 27  Joined Cases C-46/93 and C-48/93 Brasserie du Pêcheur v Germany and R v Secretary of State for Transport, ex parte Factortame Ltd [1996] ECR I-1029. 28  Case C-453/99 Courage v Crehan [2001] ECR I-6297. 29  Case C-224/01 Köbler [2003] ECR I-10239; Case C-160/14 Da Silva e Brito judgment of 9 September 2015.

From Myth to Reality

 31

sincere co-operation today found in Article 4(3) TEU, which itself has interesting connections with functionally similar requirements in international treaty law,30 but which more recently it has airily declared to be ‘inherent in the system of the Treaty’.31 This conforms to the concern, visible as far back as Van Gend en Loos over fifty years ago, to root the special character of EU law in the structure and purpose of the Treaty itself, not in any particular textual provisions. There is room to argue about whether the Court’s vision of when and why EU law is apt to determine the outcome of proceedings before national courts is consistent.32 In this vein the most striking twist contrasts the Court’s scrupulous concern to reject the capacity of unimplemented Directives to impose obligations on private parties in national proceedings with other devices such as the obligation of consistent interpretation which tend to subvert that protection of the individual.33 But the point of present concern is that this was all part of a conscious judicial strategy directed from Luxembourg to loosen EU law from the limits of public international law and, in particular, to assert an intimate interconnection between EU law and national legal orders which is missing from international law’s sharply contrasting readiness to keep distinct, on the one hand, obligations undertaken on the international plane and, on the other, implications within national legal orders. The Court’s recent Opinion 2/13, in which it refused to agree that the planned arrangements for the EU’s accession to the ECHR were compatible with EU law, is rich in its depiction of the distinctive character of EU law.34 And it celebrates the fifty-year history of the Court’s key case law by asserting that it has repeatedly held that ‘the founding treaties of the EU, unlike ordinary international treaties, established a new legal order, possessing its own institutions, for the benefit of which the Member States thereof have limited their sovereign rights, in ever wider fields, and the subjects of which comprise not only those States but also their nationals’ and of course the Opinion faithfully cites both Van Gend and Loos and Costa v ENEL.35 It proceeded, setting the scene for its investigations of the threshold conditions that needed to be met (and which were not met) before the EU could accede to the ECHR, that ‘the EU has a new kind of legal order, the nature of which is peculiar to the EU, its own constitutional framework and founding principles,

30  cf G De Baere and T Rose, ‘EU Loyalty as Good Faith’ (2015) 64 International and Comparative Law Quarterly 829. 31  Cases C-6/90 and C-9/90 n 00 above, para 35; Cases C-397/01 to C-403/01 n 00 above, para 113. 32  eg M Dougan, ‘When Worlds collide! Competing Visions of the Relationship between Direct Effect and Supremacy’ (2007) 44 Common Market Law Review 931; K Lenaerts and T Corthaut, ‘Of Birds and Hedges: The Role of Primacy in Invoking Norms of EU Law’ (2006) 31 European Law Review 287. 33  T Mörsdorf, ‘Unmittelbare Anwendung von EG-Richtlinien zwischen Privaten in der Rechtsprechung des EuGH’ [2009] Europarecht 219; P Craig, ‘The Legal Effect of Directives: Policy, Rules and Exceptions’ (2009) 34 European Law Review 349. See also S Weatherill, Law and Values in the European Union (Oxford, Oxford University Press, 2016) ch 4.9. 34  18 December 2014. 35  Para 157 (emphasis added).

32 

Stephen Weatherill

a particularly sophisticated institutional structure and a full set of legal rules to ensure its operation’.36 The rise of general principles, binding on the EU and on Member States when they act within the scope of EU law, is exactly in line with this narrative. It was a necessary part of the Court’s concern to make good its claim to be crafting a new legal order with a full and reliable set of rules designed to ensure judicial protection of the individual. The Court could not feasibly expect national judges to apply EU law as directly effective and supreme over conflicting national law if the character of EU law was not endowed with principles that secured the protection of the individual against public and private power exercised in the name of EU law. So the Court’s findings in Van Gend en Loos and Costa v ENEL, supplemented by Internationale Handelsgesellschaft, which were based on the object and purpose of the system and which elucidated the principles of direct effect and supremacy, necessitated further more ambitious findings about the nature and content of EU law too. In this way were nurtured the general principles of the EU legal order. So EU law is new. Its understanding of the relationship between itself and national law is not the same as the normal understanding of the impact of international treaties on national legal orders. Instead EU law makes very concrete demands of national legal orders, and as such it asserts an integration of EU law within national law and everyday national judicial practice in a manner that is distant from the orthodox expectations of international law.

V.  And Yet … Pause once more before fully embracing the discourse of ‘novelty’ attached to EU law. Article 31 of the Vienna Convention on the Law of Treaties states that: ‘A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.’ Is this not what the Court of Justice did in Van Gend en Loos and Costa v ENEL? Is it in fact the case the Court has always been interpreting the EU treaties in line with the conventional interpretative tools afforded by public international law? Reliance on ‘object and purpose’ is precisely the animating theme of both Van Gend en Loos and Costa v ENEL. True, the Court does not explicitly adopt the Vienna Convention as the basis for its interpretative choices. That would have defeated its calculated strategy to distance EU law from the international law of treaties. In any event the Convention post-dates the heroic cases decided in Luxembourg in the early 1960s and,

36 

Para 158.

From Myth to Reality

 33

moreover, the EU is not a party to it.37 However, both Van Gend en Loos and Costa v ENEL are fully and thoroughly rooted in a claim about the object and purpose of the (at the time) EEC Treaty which corresponds to the interpretative approach today mandated by Article 31 of the Vienna Convention. The Court’s claim about EU law’s capacity to dictate how it shall be applied within domestic legal orders is built on and inspired by the object and purpose of the Treaty of Rome, most of all its identification of the system’s concern not only with the rights and obligations of states but also those of individuals. So in fact the Court’s interpretative methodology is not so new at all. The Court’s path in fact amounts to an ambitious reimagining of the EU legal order as one which also embeds the protection of the individual according to particular legal precepts deep in national judicial practice. The constitutionalised legal order is sufficiently powerful to be difficult to evade when it is not to the shortterm taste of public or private actors and it is granted the necessary qualities that constrain the exercise of public (and sometimes private) power that entitle it to deserve respect as an expression of the rule of law, respect for which is written into Article 2 TEU. The Court has in the same vein moved beyond its original depiction of a ‘new legal order’ of international law38 to describe the Treaty as a ‘basic constitutional charter’.39 Opinion 2/13 treats the EU as a new legal order, ‘the nature of which is peculiar to the EU’ and it insists too on the need to protect the treaties as the ‘basic constitutional charter’.40 The Court’s engagement with general principles is a direct result of what it chose to do over fifty years ago when it used international law’s interpretative methodology, in particular by relying on its perception of the object and purpose of the Treaty of Rome, to distance EU law’s relationship with national law from the normally understood relationship between national and international law. The core of this is to emphasise the place of individual rights within an EU legal order that makes significant demands of national legal orders. My argument here is not that all of this is for the best in the best of all possible judicial worlds. There remain plenty of room for critical engagement both with the substance of the Court’s choices and with the factors that explain them. To identify the rise of general principles in the EU legal order as a consequence of the momentum supplied by the foundational rulings in Van Gend en Loos and Costa v ENEL does not mean that one has to agree with the Court’s view of the object and purpose of the Treaty of Rome. And it certainly does not mean that

37  The Court does rely on the Vienna Convention when it deals with the interpretation of international Treaties between the EU and the Member States (eg Joined Cases C‑464/13 and C‑465/13 Europäische Schule München v Silvana Oberto, Barbara O’Leary judgment of 11 March 2015) and between the EU and its Member States and third countries (eg Case C-613/12 Helm Düngemittel v HZA Krefeld judgment of 6 February 2014; Case C-386/08 Brita GmbH v HZA Hamburg-Hafen [2010] ECR I-1289; Opinion 1/91 on the draft EEA Agreement [1991] ECR I-6079). 38  Case 26/62 note 1 above. 39  Case 294/83 Parti Ecologiste Les Verts v European Parliament [1986] ECR 1339. 40  Note 34 above, paras 158 and (citing Case 294/83 note 39 above) para 163.

34 

Stephen Weatherill

one has to agree with the interpretative choices that underpin every single ruling of the Court.41 In particular the Court sometimes allows itself to be dazzled by its own constitutionalising rhetoric. The Court on occasion takes too much on its own shoulders and too much on EU law’s generally. The rulings that tend to generate the most anxiety are those which are perceived to go too far in constraining national autonomy and/or to go too far in addressing problems through judicial means in preference to legislative solutions. Reyners, Cassis de Dijon and Centros, mentioned above,42 deserve to be assessed both from the perspective of the deepened intrusion they assert into national autonomy in the name of Court-driven free movement law and the consequent diminished scope required for legislative action at EU level. I believe they are fully justified in that vein: they promote an integrated market based on regulatory diversity disciplined by the justificatory requirements imposed by the free movement rules in preference to the promotion of centralisation and homogeneity. Sometimes, however, the Court stands more plausibly accused of over-inflating its role and illegitimately disturbing the vertical distributions of competences between the EU and its Member States and the horizontal distribution of powers between the Court and the EU’s political institutions. It escapes the scope of this chapter’s ambition to explore all the manifestations of this trend. Suffice to note that, in the interpretation of primary law, one would inquire into rulings such as Mary Carpenter,43 the run of cases that began in early in 2011 with Ruiz Zambrano,44 and Viking Line and Laval,45 and, in the interpretation of secondary legislation Sturgeon,46 Simone Leitner47 and West Tankers Inc.48 There is a risk here of collecting a colourful but atypical bag of oddities, but there is a common thread: the Court stands accused of sometimes promoting over-centralisation (by tugging the vertical distribution of competences in favour of the EU and away from the national level) and over-judicialisation (by locating too much problem-solving and policy-making capacity through provisions that are applied by courts at the expense of the powers conferred on the EU’s political institutions). Just because there is a gap in legal protection does not mean it should always be filled by EU law in general or by the Court in Luxembourg

41  See eg G Conway, The Limits of Legal Reasoning and the European Court of Justice (Cambridge, Cambridge University Press, 2012); G Beck, The Legal Reasoning of the Court of Justice of the EU (Oxford, Hart Publishing, 2013); E Paunio, Legal Certainty in Multilingual EU Law (Aldershot, Ashgate, 2013); S Sankari, European Court of Justice Legal Reasoning in Context (Groningen, Europa Law Publishing, 2013). For a helpful review of recent work in this vein, see J Bengoetxea, ‘Text and Telos in the European Court of Justice’ (2015) 11 European Constitutional Law Review 184; also, reviewing Beck’s book but commenting more broadly, M Bobek, Review Article (2014) 39 European Law Review 418. 42  Case 2/74 note 2 above, Case 120/78 note 3 above, Case C-212/97 note 4 above, respectively. 43  Case C-60/00 [2002] ECR I-6279. 44  Case C-34/09 [2011] ECR I-1177. See also especially Case C-256/11 Dereci [2011] ECR I-11315. 45  Case C-438/05 International Transport Workers’ Federation v Viking Line ABP [2007] ECR I-10779; Case C-341/05 Laval un Partneri [2007] ECR I-11767. 46  Joined Cases C-402 and C-432/07 [2009] ECR I-10923. 47  Case C-168/00 [2002] ECR I-2631. 48  Case C-185/07 [2009] ECR I-663; also Case C-536/13 Gazprom judgment of 13 May 2015.

From Myth to Reality

 35

in particular.49 This paper’s concern is with general principles of law. In that field Mangold stands tall and nervous in any excoriation of the Court’s alleged overconstitutionalisation of EU law.50 The Court in that case examined ‘various international instruments and … the constitutional traditions common to the Member States’,51 and from these seeds it extracted a general principle of law forbidding discrimination on grounds of age which moreover applied to disputes between private parties. But the seeds were in truth not as fertile as the Court claimed, and the judgment has attracted much dismayed comment for its unexpected injection of EU law into both national and private autonomy.52 If this were commonplace there would be cause for alarm. But it is not. The Court is usually attentive to the limits of its role, both where it touches national autonomy and where it affects the EU’s legislative process. In Audiolux the Court refused to extract a general principle addressing the protection of minority shareholders from the existing pattern of rights and obligations created by the several Directives that harmonise company law. What was lacking was ‘the general, comprehensive character which is otherwise naturally inherent in general principles of law’.53 Advocate General Trstenjak’s Opinion in the case provides a very good explanation of the virtues of judicial caution. The Court should respect the superior democratic legitimation and expertise which belongs to the political institutions in matters that demand a choice between competing political and social interests. She warned too against judicial over-ambition that might harm ‘protection of legitimate expectations and legal certainty’.54 Subsequently in Sonia Chacon Navas the Court refused to accept that non-discrimination on grounds of sickness counts as a general principle of EU law.55 In the same vein the Court in Kaltoft was willing to accept that discrimination in the workplace on the ground of obesity might fall within the scope of discriminatory conduct on the basis of disability which is prohibited by Directive 2000/78 establishing a general framework for equal treatment in employment and occupation,56 but it refused to accede to the much more ambitious claim that EU law contains a general principle forbidding discrimination on the basis of obesity.57 49  cf sources at note 41 above, and on free movement law in particular, see P Caro de Sousa, The European Fundamental Freedoms: A Contextual Approach (Oxford, Oxford University Press, 2015) and N NicShuibhne, The Coherence of EU Free Movement Law: Constitutional Responsibility and the Court of Justice (Oxford, Hart Publishing, 2015). Clearly the issues addressed in compressed manner in this paragraph (and which are largely confined to the case for the prosecution) deserve book-length exploration in their own right. 50  Case C-144/04 Mangold [2005] ECR I-9981. 51  Para 74. 52 For assessment of the issues and the literature, see M Dougan, ‘In Defence of Mangold?’ in A Arnull, C Barnard, M Dougan and E Spaventa (eds), A Constitutional Order of States? Essays in EU Law in Honour of Alan Dashwood (Oxford, Hart Publishing, 2011). 53  Case C-101/08 Audiolux [2009] ECR I-9823, para 42. 54  Para 111. 55  Case C-13/05 Sonia Chacon Navas [2006] ECR I-6467. 56  [2000] OJ L303/16. 57  Case C-354/13 judgment of 18 December 2014.

36 

Stephen Weatherill

The general principles of EU law are no judicial free-for-all, notwithstanding the shock of Mangold. Probably the Court will today choose to confine their development within the parameters set by the Charter of Fundamental Rights, which was endowed with binding legal effect as a result of the entry into force of the amending Treaty of Lisbon in 2009. In exactly this spirit the Court in 2010 in Seda Kücükdeveci v Swedex GmbH & Co KG treated discrimination on grounds of age in the context of a dispute between private parties, the very issue at stake in Mangold, as a general principle of EU law, to which Directive 2000/78 gives specific expression, but it added too reliance on Article 21 of the Charter which prohibits discrimination based on age.58 General principles of law breathe extra life into the EU legal order as a means to improve the protection of the individual. They are exactly in line with the animating logic of Van Gend en Loos and Costa v ENEL. They are of the same blood as those foundational judgments. To identify the rise of general principles in the EU legal order as a consequence of the momentum supplied by the foundational rulings in Van Gend en Loos and Costa v ENEL should not allow the inquiry to miss the interest in inquiring why Member States, academic scholars of the early years and, the most powerful gatekeepers of all, national courts did not rebel against the Court’s pretensions and especially its assertion, distant from the orthodoxy of international treaty law, that EU law determines how national law shall treat and apply this particular treaty.59 It was the national courts in particular who converted the Court’s evasion of the conventional discourse of international treaties in order to create a founding myth of juristic novelty within the EU into reality. EU law in fact became new because it migrated from Luxembourg to national judges who across the territory of the EU shared an overall readiness to apply EU law on the (directly effective and supreme) terms dictated by the Court in Luxembourg. True, national courts did not and do not accept the radical claim of the Court of Justice that the authority of EU law within national legal orders is the consequence of the entry into force of the Treaty itself. They instead—in different ways60—find routes under national constitutional law along which EU law travels in order to reach its destination within national legal orders. This is a major reason why national courts need to be persuaded of the adequacy of legal protection secured within the EU system. Their anxieties have focused in particular on protection of fundamental

58 

Case C-555/07 [2010] ECR I-365. it suited them in so far as they noticed, it suited them and they did notice, and it suited them but they got to the destination through a different route (dictated by national law) from that laid out by the Court. I try to tell this story in S Weatherill, Law and Values in the European Union (Oxford, Oxford University Press, 2016) ch 4. 60  See eg M Claes, The National Court’s Mandate in the European Constitution (Oxford, Hart Publishing, 2005); M. Everson and J. Eisner, The Making of a European Constitution: Judges and Law beyond Constitutive Power (London, Routledge Cavendish, 2007). 59  Respectively:

From Myth to Reality

 37

rights, the scope and limits of EU competence and preservation of national and constitutional identity.61 The track record shows that national courts are open to be persuaded and that they do treat EU law as supreme and direct effective. There remains, in sum, plenty of room for critical engagement with the Court’s choices. But the point for present purposes is to say that the Court’s predilection to go beyond the text, established so early in the development of EU law, is not revolutionary as a matter of interpretative methodology applicable to international Treaties—even if the result, to dictate in detail how individual rights shall be enforced at national level and to supplement the pattern with general principles of law that affect the individual, definitely is.

VI. Conclusion The Court has identified EU law as supreme—it applies in preference to national law in the event of conflict—and as directly effective, which means that rules of sufficient clarity and precision fall to be applied on an everyday basis by the national courts of the Member States. And the Court of Justice and the national courts are organically tied together in a co-operative relationship by a procedure that is foreseen explicitly in the EU treaties, the preliminary reference procedure which permits, and in some circumstances requires, awkward points of EU law raised at national level to be transmitted to the Court in Luxembourg for authoritative resolution. Crucially, national courts have accepted the invitation to dance to this tune, even if they treat themselves and their own domestic constitutional orders, rather than the Court of Justice, as the conductor of the orchestra. The net result is that EU law exerts a much more powerful influence at national level than does orthodox public international law. EU law functions as if arranging the vertical relationship between tiers of governance in a federal system (of sorts). The identification of general principles follows directly from these major choices. The existence of general principles of law in the EU legal order flows unavoidably from the choices made by the Court in Van Gend en Loos and Costa v ENEL, even if detailed questions about their proper scope and impact remain open and contested. Only a legal order embroidered by general principles could match the breadth of the Court’s vision of the new legal order which directly concerns individuals and not states alone, and only such an enriched legal order could deserve to command

61  See eg M Payandeh, ‘Constitutional Review of EU Law after Honeywell: Contextualising the Relationship between the German Constitutional Court and the EU Court of Justice’ (2011) 48 Common Market Law Review 9. The latest chapter in a story of dialogue which by definition cannot be finished short of a legal revolution is Case C-62/14 Gauweiler judgment of 16 June 2015.

38 

Stephen Weatherill

the respect of properly suspicious national courts. And national courts have accepted the Court’s chutzpah in directing how EU law shall be applied within national legal orders, even if they do not go so far as to accept the Court’s reasons why that should be so. So myth—the ‘new legal order’ proclaimed over fifty years ago—became reality.

Part I

General Principles of European Union Law

40 

4 Theorising General Principles of EU Law in Perspective: High Expectations, Modest Means and the Court of Justice URŠKA ŠADL AND JOXERRAMON BENGOETXEA

The aim of our contribution is to introduce and discuss three papers presented at the first panel of the Conference that dealt with selected theoretical questions surrounding the general principles of EU law. We discuss the papers in the context of the judicial decision-making of the Court of Justice of the European Union (hereinafter the Court). More concretely, we engage with three concerns and challenges that the papers pose to the nature, role and function of the general principles: first, the underused potential of EU customary law as a source of general principles of EU law; second, the role of the Court in the general endeavour of the juristic field or community to elaborate a theory of general principles of EU law, and to bind structural and substantive principles of EU law into a coherent whole; and third, the particular nature of legality as a general principle of EU law. Before we proceed with the discussion, we introduce the main arguments of the papers: we inquire into how they contribute to an understanding of the operation of general principles in the context of legal reasoning, as it is used in concrete decision-making contexts, and how they complement each other.

I.  Theorising General Principles of EU Law in the Context of Judicial Decision-Making Principles point in a general direction or orientation but they do not dictate any specific answers. In a sense they help provide structure and architecture, systematising the vast body of norms into a coherent whole. They reduce legal dissonance. The Court most often deals with the general principles of EU law in disputes arising at a Member State (national or domestic) level, due to an alleged ­incompatibility

42 

Urška Šadl and Joxerramon Bengoetxea

of national legislation with EU law, or in proceedings between the undertakings and the Commission, and individuals claiming rights or complaining about the breach of their rights. Thus conceptual and theoretical discourse about general principles of EU law is probably best analysed in the context of their application. Take the situations below as examples, which are by no means exceptional. In Denmark, Mr K, a child-minder, is dismissed from his job due to his employer’s financial difficulties. Prior to the dismissal his employer enquires about the progress that Mr K is making with a weight-loss programme. This leads Mr K to believe that he is being dismissed on grounds of obesity, and he decides to sue the employer for unjustified dismissal. The national court sends a preliminary reference to the Court of Justice asking whether discrimination on grounds of obesity is a general principle, and if so, whether it is prohibited under EU law.1 In Poland, Mr M, a court enforcer, collects VAT from the compulsory sale of real property. When the local tax authority demands the payment of the collected tax, Mr M brings an action before the administrative court, arguing that national legislation, which holds him accountable and liable for the payment, is contrary to EU law and the principle of proportionality. The national court stays the proceedings, addressing a preliminary reference to the Court of Justice: is national legislation in accordance with the VAT Directive and is the principle of proportionality a general principle of EU law?2 In Italy, wine producers initiate judicial proceedings because they are no longer allowed to use the word ‘Tocai’ in the term ‘Tocai friulano’ or ‘Tocai italico’ for certain wines. This prohibition is based on the exchange of letters on Tocai annexed to the EC–Hungary Agreement on wines, ratified prior to accession. The national court must rule whether the right to property includes protection of intellectual property in designations of origin of wines and its exercise, and whether it precludes this prohibition? The interpretation of the right to property must take into account multiple legal sources, as the right to property is a general principle of EU law, spelled out in Article 1 of Protocol No 1 to the European Convention on Human Rights and incorporated in Article 17 of the Charter of Fundamental Rights. It refers the question to the Court of Justice.3 In the first case the Court is asked to find or to create, by extension, a general principle, in the second case to decide whether the existing principle, proportionality, was breached, and in the third case to define the scope of protection of an existing principle (to redefine the object of protection). In all cases the Court must take a stance on the body of law that governs the actual situation. This body of law that must be observed in the interpretation and the application of the EU treaties (Article 19(1) TEU) includes a set of broader principles. Some of these

1 

Case C-354/13 Fag og Arbejde v Kommunernes Landsforening, ECLI:EU:C:2014:2463. Case C-499/13 Marian Macikowski v Dyrektor Izby Skarbowej w Gdańsku, ECLI:EU:C:2015:201. 3  Case C-347/03 Regione autonoma Friuli-Venezia Giulia and Agenzia regionale per lo sviluppo rurale v Ministero delle Politiche Agricole e Forestali, ECLI:EU:C:2005:285. 2 

Theorising General Principles of EU Law in Perspective

 43

principles can operate as normative statements of legal reasoning, leading to a ­particular solution to the problem case at hand, other principles make that law into a ­coherent whole, which can be uniformly interpreted and applied in the twenty-eight Member States and in all the legal systems of the EU. At the same time, the Court must ensure that the coherence of the whole system is not compromised, and that the rights of entrepreneurs, overweight people and the like in Italy, Denmark and Poland and all other Member States are protected. In Dworkinian terms, the Court is asked to weave a seamless web of rules and principles which would reflect the fundamental values of the legal system. Preferably, we could add, in due time, and not abusively. An already difficult task is further complicated by the fact that the Court needs to weave into this web the principles of numerous legal orders, meaningful, if not common, to them all. The working of this so-called creative or interpretative process might not be entirely easy to understand, even for the Court, but the process, if genuinely, sincerely and transparently carried out may reflect the fundamental values of the Union. In other words, the Court must uphold the ideal of law as a rational system. It is an ambitious goal, which the Court must accomplish with modest means in terms of judicial method, time, personnel, knowledge and administration.

II.  Familiar Concerns and New Challenges Samantha Besson, Alexander Somek and Xavier Groussot, J Hettne and GT Petursson present distinct and original accounts of the Court’s use and non-use of principles and their role in EU law. On the one hand, the contributions on the conceptual and theoretical aspects of general principles are as diverse as the topic itself, ranging from the doctrine of the sources of EU law, decisionism as its main feature and the consequences thereof, and the duty of the Court as a constitutional court in elaborating a coherent legal system. On the other hand, the contributions revisit the same or at least very similar and well-known themes: the often-criticised decision-making practice of the Court, and the peculiar character of the EU legal system, with its distinct sources. With regard to the latter, the contributions complement each other. Somek’s contribution answers Besson’s inquiry into the possible reasons for the very limited role that European customary law has played in EU law and the Court’s practice. Groussot et al might want to take into account the role of unwritten practices identified by Besson when assessing the meaning of mutual trust as an emerging principle of EU law. Somek’s analysis of legality could be fine-tuned even further by distinguishing between substantive and structural general principles of EU law. Besson unveils perhaps the most underplayed source of EU law, ie EU c­ ustomary law, to discuss the role of general principles in EU law. As opposed to national and

44 

Urška Šadl and Joxerramon Bengoetxea

international law, EU law increasingly relies on the general principles of law but, surprisingly, does not mention customary law amongst its sources. ­Customary law, however, is a de facto source of many general principles of law. Customary law is also identified in the same manner and following the same process as general principles of law. Accordingly, it should therefore attract more scholarly attention, and feature more prominently in the decision-making of the Court and the European institutions. Besson convincingly argues how, from the viewpoint of the accepted doctrines of sources of international or European law, none of the arguments against a more pronounced role of customary law are truly persuasive: neither the supranational character of EU law, nor its democratic legitimacy, nor the complex vertical and horizontal division of powers in the EU, nor the jurisprudential nature of EU law, with a pronounced role of the Court. Legality as a general principle of EU law would seem to be in sharp contrast with the notion of customary law as a source. Somek inspects the credentials of EU law ‘as law’, asking: is legality a principle of EU law? Does EU law exhibit elementary features of legality? What weight does legality have? (Maybe, partly and little.) He furthermore argues that EU law exhibits a peculiar legality, which enters the picture only to mitigate the entrepreneurship of the Court. The latter is nothing but a consequence of the so-called original sin of EU law, which was conceived in a way that required clarification by means of decisions (hence the term decisionism) by the Court through the mechanism of preliminary references. To know EU law is thus to know EU law as determined in judicial decisions, something that perturbs a continental European lawyer more than it does other jurists. The Court took full liberty of that feature of EU law, at the expense of the principle of legality, as it has conventionally been interpreted, ie as linked to certainty and plain meaning. This principle is now strained even further given the economic and constitutional crises. Finally, Groussot et al broach the often implicit and ill-defined connection between the general principles of law and coherence to discuss the relationship among different types of general principles: structural and substantive. When the Court develops substantive principles, such as the principle of good administration or mutual trust it at the same time touches upon structural principles, such as the principle of effectiveness, which are embodied in them. The Court (or a court that takes its constitutional responsibility seriously) should construct a coherent legal order, striking a balance between substantive and structural principles, and provide answers that are specific enough to efficiently resolve individual conflicts at hand and simultaneously articulate the content of the general principle for the future. Perhaps this is already a provisional reply to the objection brought from the perspective of legality raised by Somek. The analysis of the principles of good administration and mutual trust shows that the Court has not fully succeeded in this task. Here the charge would not be normative or existential—what should the Court do?—as in Somek, but rather one of efficiency and quality of what it does or does not achieve.

Theorising General Principles of EU Law in Perspective

 45

Would it be possible, by drawing on the three papers, to tell the Court what to do, more concretely, in the above cases? If the Court creates (or finds) that prohibition of discrimination on grounds of body weight is a general principle, that ­intellectual property can be classified as property under Article 17 of the Charter and that proportionality allows the Member States to collect tax from court enforcers, it might potentially disrupt the rights of traders, employers and employees, as well as public officials in twenty-eight national states. Some of those might already prohibit discrimination on grounds of obesity, others might impose even stricter rules on public officials. For the former the Court’s ruling will seem intuitively correct. For the latter the ruling might be too soft. The criticism of the Court, to be defensible, will have to come from the perspective of EU law, or the laws of the Member States generally speaking, and not from a single Member State. Could it come from EU customary law? In principle, customary EU law might provide an answer that could be more legitimate (more communitarian) than a comparative study of national legal systems or a simple general reliance on the general principles common to the Member States (which was considered an interpretive sin in Mangold).4 What would legality require? The Court was called to decide, to interpret EU law in the way, which would uphold the principles of EU law as law (cf Fuller’s list).5 The Court was cautious and deferential in its responses. But was it overly cautious, not giving enough guidance for the future cases? Too casuistic perhaps? Or were the Court’s tactics to reserve for itself the possibility to expand the principle at will in a subsequent case?

III.  The Sui Generis (Autonomy) Thesis and EU Customary Law From Somek’s contribution we can answer the question why the absence of customary EU law is not (so) surprising. It would be possible to argue that EU customary law was conceptually impossible not (only) because of the claim to supranationality, democratic legitimacy, the complex power structure and the pronounced role of the Court but due to the idea that EU law should be real law and distinct from international law.6 Substantial efforts went into upholding the new legal order of international law (the sui generis thesis), of which the absence of reference to custom was a logical consequence. The opinio iuris that could have identified the unwritten sources of law was decidedly communitarian in spirit

4 

Case C-144/04 Werner Mangold v Rüdiger Helm, ECLI:EU:C:2005:709. Somek, see chapter 5 in this volume. 6  JHH Weiler, ‘The Transformation of Europe’ (1991) 100 Yale Law Journal 2403–83. 5 

46 

Urška Šadl and Joxerramon Bengoetxea

and ambition,7 and also due to a scholarly self-interest in creating a separate and autonomous legal discipline. Namely, the idea of EU law as real law, post World War II, was a reaction to the fiasco of international law as institutionalised in the League of Nations, and to the failure of national courts as guardians of unwritten law (particularly in G ­ ermany). If anything, the European Community was seeking to prevent the return to interstate (or worse, nation state) practices as the ultimate measure of what was valid law. It replaced them by the Treaty and the new commitments of the Member States, which the Court turned into a federal-like structure,8 creating a common market that had a clear harmonisation and uniformisation objective driven by Community legislation, almost a normative substitution of the state in certain sectors. To compensate, the new legal system promised peace and prosperity. It was to be a cultural and spiritual force (Hallstein’s speech cited in Somek’s ­contribution).9 The decision in Dassonville10 followed the same path, prohibiting discriminatory and/or restrictive practices of the Member States without reference to whether they were practised inter-state and by all traders (lex mercatoria), as not permitted under Community law. Any form of state protectionism in the common market of goods was unacceptable. Moreover, in this legal context, general principles of law might have been ­(perhaps wrongly) perceived as less confrontational. Even if they came from non-positive sources of law, such as custom, and were therefore vague, abstract and hopelessly indeterminate, they still played an overall unifying role and laid an inherent claim to validity and legitimacy, unlike the codification of interstate practice in public international law, which was more factual, more akin to ­realpolitik. This raises another interesting discussion as to the legitimating force of a type of jurisprudence that is comparative in nature and attuned to the practice of individual Member States. Nevertheless, customary law in the EU is related to institutional practice, and even though it might eventually feed into positive law and get codified in the ­treaties, it will still be relevant, at least in the interim. Just think about the Luxembourg compromise, the summits of the ‘European Council’ before this organ was brought into the treaties as an institution, or afterwards, if we take the odd formation of the Heads of Government or State meeting in the Council framework, nowhere to be found in the treaties, but still adopting important international law

7  H Schepel and R Wesseling, ‘The Legal Community: Judges, Lawyers, Officials and Clerks in the Writing of Europe’ (1997) 3 European Law Journal 165. See also A Vauchez, ‘The Transnational Politics of Judicialization. Van Gend En Loos and the Making of EU Polity’ (2010) 16 European Law Journal 1; M Rasmussen, ‘From Costa V Enel to the Treaties of Rome—A Brief History of a Legal Revolution’ in LMP Maduro and L Azoulai (eds), The Past and Future of EU Law: The Classics of EU Law Revisited on the 50th Anniversary of the Rome Treaty (Oxford, Hart Publishing, 2010). 8  The term coined by E Stein, ‘Lawyers, Judges, and the Making of a Transnational Constitution’ (1981) 75 American Journal of International Law 1. 9  Somek, see chapter 5 in this volume. 10  Case 8/74 Procureur du Roi v Benoît and Gustave Dassonville, ECLI:EU:C:1974:82.

Theorising General Principles of EU Law in Perspective

 47

instruments that have a bearing on the very treaties. The most recent example of this is the March 2016 decision concerning the conditions of membership of the UK in the EU with a view to the Brexit campaign. Comitology is another good example of customary institutional practices, or constitutional conventions ­playing a key role in EU law-making.

IV.  General Principles as a Structuring Device in Need of a (Coherent) Structure: The Case of Human Rights As Groussot et al observe, in the beginning the coherence of the EU legal order was policy-based. It rested comfortably on the accepted premise that the goal of the Community was to build a common market for goods and services. All legislation had the same policy goal in mind, hence very little need for the hierarchy of legal sources of secondary law or for the systematisation of the guiding substantive principles. Then, human rights as general principles of EU law introduced a so-called irritant to the policy-based coherence of the EU legal order. They furthered substantive principles. Functional coherence of policies that promoted the common market, the single market or the internal market was replaced with substantive coherence, produced through these principles of procedural and substantive justice. When the Court in Opinion 2/13 today speaks of the structured network of principles, the authors rightly ask, what kind of structure does it have in mind? How are human rights as general principles, as norms recognised in the Charter and international treaties and not least in the state constitutions that form part of the common constitutional traditions (another source of EU general principles), found, interpreted and systematised to produce a coherent legal framework, within which concrete cases can be decided? In other words, what is the organising principle (of coherence), from which the Court should build this structure, if it is to take seriously its role as a constitutional court? On the one hand, the mere plurality of legal sources can be disconcerting. As Groussot et al show, when it comes to the principle of good administration, it is almost a matter of coincidence which sources the Court will use, and on what legal ground.11 On the other hand, and quite intuitively, there should always be room for one more. Especially so, if the extra one could help explain or contextualise the seemingly inconsistent practice of the Court, or act as the organising principle that could structure the general principles into a coherent system. The argument presented by Besson, that the customary approach to EU fundamental rights best accounts for specificities in practice, seems, at first glance, to be one

11 

Contribution by Groussot et al, see chapter 6 in this volume.

48 

Urška Šadl and Joxerramon Bengoetxea

such u ­ nderlying and potent device. The argument is persuasive, especially as it cuts the Gordian knot of the multiplicity of almost casually interacting sources, exposed by Groussot et al. First, as Besson submits, the function of EU fundamental rights may not be compared to either national or international human rights (Articles 51 and 53 of the Charter). Rather, they are transnational, and grow out of inter-state practice, but not exclusively from state administrations, since other actors and players generate horizontal normative relations. Second, the customary account is respectful of the idea that the protection of fundamental rights in the EU cannot be a ground for a new competence and thus avoids charges of a competence creep into the human rights duties of the Member States. Third, the customary account of EU fundamental rights justifies and explains why EU fundamental rights are rarely interpreted in conformity with international human rights: since EU rights grew from within Member States’ practice as an integrated human rights regime, they are not concerned with the minimal external guarantees that apply directly to those regimes. And fourth, EU fundamental rights as customary law work as unwritten constitutional constraints on EU primary law, because they are concurrent to this law, rather than internal to the latter, similarly as constitutional customary law works as the supra-constitutional constraint to national constitutional law. The recommendation to return to a more comparative fundamental rights reasoning, perhaps a consensus-seeking reasoning, as reflected in the Court’s early decisions might solve Groussot et al’s conundrum and contribute to a more consistent practice of the Court with respect to fundamental rights. The return to customary law would solve the problem of coherence and consistency of the Court’s practice. But would it solve the more principled problem of the balance of different types of principles (of structure and substance)? Perhaps, if it also seeks consensus.

V.  Continuously Renegotiated Legality Finally and perhaps most critically, does EU law possess the necessary credentials to be called law as we know it? Legality is closely tied to legal certainty and predictability, which is inter alia played out in the interpretation and the application of rules to individual fact situations. The Court’s creative interpretations (or, in Somek’s words, divinations) clearly go against it. EU law cannot truly sanction them for structural reasons, one among them being the intrinsic need for clarification of indeterminate law through judicial decisions instituted into the system by the preliminary reference procedure. This, Somek forcefully argues, brings the law from the usual and acceptable weak indeterminacy into a state of strong indeterminacy, where nearly anything goes. It is a state of actively constituted indeterminacy.12 12 

Somek, see chapter 5 in this volume.

Theorising General Principles of EU Law in Perspective

 49

In Somek’s view the Court has compromised the principle of legality with its entrepreneurial decision-making, making good use of the preliminary reference procedure. The compromised principle has then been restored through political action and the subsequent acceptance by the Member States. Thus legality can only be accidental to the EU legal order, a sort of side-effect.13 Defrenne14 and Barber15 are two cases in point. However, we could conceive of an alternative diagnosis. While Defrenne can certainly be and has been read as an example of judicial policy-making,16 it can also be read as an instance of principled incrementalism, ie a step-by-step judicial decision-making that strives for a workable balance between societal and legal concerns. All courts consciously or unconsciously balance the demands of the individual case against the demands of the whole body of law, in particular legal coherence and consistency.17 Courts use strategies that let them preserve the authority to interpret legal norms in the long run, and at the same time preserve individual rights. In the case of international courts the balancing is additionally strained by the increased political pressure from various states with dissimilar legal systems, conflicting political interests, varying degrees of international commitment and, of course, the absence of a central enforcement mechanism that makes international courts particularly vulnerable and dependent on co-operation with powerful political actors. The legality that emerges is renegotiated legality through renegotiated authority of what courts can reasonably and legitimately do. Doctrinally speaking, Defrenne was an extension of the principle of direct effect to the so-called horizontal situations. It was one case in a series of cases in which the Court slowly and step by step (incrementally) elaborated the doctrine of direct effect. The Court’s limitation of temporal effects had a stabilising effect on the system, and a rebalancing function. It served legal certainty. As is well known, the Court used the separation strategy and the narrow doctrinal application strategy. First, it decoupled the principle of horizontal direct effect from the remedy, both on the level or language and in terms of its effects in practice. The temporal limitation was a reconciliatory gesture of the Court, and it did not escape criticism. While some gave the Court credit for accepting ‘the responsibility to mould constitutional doctrine in order to make more acceptable the practical effects of ­judicial decisions’,18 others found this type of ‘amnesty’ unacceptable.19 Second,

13 ibid. 14 

Case 43-75 Defrenne v Société anonyme belge de navigation aérienne Sabena, ECLI:EU:C:1976:56 C-262/88 Douglas Harvey Barber v Guardian Royal Exchange Assurance Group, ECLI:EU:C:1990:209. 16  Most strongly in H Rasmussen, On Law and Policy in the European Court of Justice: A Comparative Study in Judicial Policymaking (Leiden, Martinus Nijhoff, 1986). 17  U Šadl, ‘The Role of Effet Utile in Preserving the Continuity and Authority of European Union Law: Evidence from the Citation Web of the Pre-Accession Case Law of the Court of Justice of the EU’ (2015) 8 European Journal of Legal Studies 18. 18  Rasmussen (n 16) 438ff. 19  A comment by C Philip [1976] Revue trimestrielle de droit européen 529–35. 15 Case

50 

Urška Šadl and Joxerramon Bengoetxea

the Court did not address the entire spectrum of situations that could/would lead to discrimination but only addressed a narrow aspect of discrimination: direct and overt discrimination, which could be determined by ‘legal means’.20 This further tapered the doctrinal scope and potential of the judgment. Following this account, it could be argued, legality was not suspended, but renegotiated. Legal certainty in the form of legitimate expectations was preserved on the level of the (un)imposed remedy. Legal continuity and coherence of the body of remedies developed by the Court to protect the individual were preserved on the level of principle, safeguarding commitments made in Van Gend. The gradual acceptance of horizontal effect was then further negotiated through a series of more and less permissive rulings. Direct effect is not the only example of renegotiated legality. Examples of such subsequent acceptance of the Court’s authority to frame the fundamental doctrines of EU law are primacy, human rights protection as general principles of EU law, the principle of liability of the Member States for thea of Community law, or the principle of institutional balance (standing of the European Parliament). While some were accepted tacitly, others were explicitly written into the treaties or declarations.21 Subsequent acceptance does not have to mean that legality has been compromised and re-established on the level of formal rules, and that every such acceptance weakens the system. Legality is a principle, and as such does not impose specific actions and interpretations. It is thus difficult to make it stand as the sole benchmark for assessing the credential of a legal system as law properly understood. Instead, as Groussot et al seem to suggest, it is much more important that the structure of principles, to which the legal systems responds, is coherent. In more general terms, when courts settle questions raised in our examples in the introduction, which are novel and might touch upon the most fundamental rules, their authority to decide these questions can unavoidably only be accepted after the questions have arisen and the decision has been given; or, in Hart’s famous phrase: all that succeeds is success.22 The success of courts hinges on the acceptance of their rulings by the political actors. They increase the chances of success using different strategies. The Court, it was argued, could enhance the authority of law and upgrade its own powers by establishing its doctrines gradually,23 mitigating legal i­nnovation

20 

Defrenne, para 18. For example, the status of the European Parliament as a privileged applicant was formalised in the Maastricht Treaty, in what is now Art 263 TFEU, and the protection of human rights culminated in the Charter. Declaration 17 (Declaration concerning primacy) to the Treaties, annexed to the Final Act of the Intergovernmental Conference which adopted the Treaty of Lisbon ([2012] OJ C326 0001– 0390), explicitly refers to ‘the settled case law of the Court of Justice’. 22  HLA Hart, The Concept of Law, 3rd edn (Oxford, Oxford University Press, 2012) 153. 23  A-M Burley and W Mattli, ‘Europe before the Court: A Political Theory of Legal Integration’ (1993) 47 International Organization 41, 69. 21 

Theorising General Principles of EU Law in Perspective

 51

either by delaying the full practical effects of established principles,24 or by reiterating formulas that lift the level of discussion away from the facts and the consequences of concrete cases and decisions to the long-term common goals and interests of integration.25 When discussing legality we should thus discuss whether these strategies and techniques are in themselves illegitimate and what could replace them, or whether they are acceptable as inherent to every judicial and administrative decision-making. Legality is renegotiated through hard cases. They might make bad law but they do make law.

VI. Conclusion To conclude, the contributions of Somek, Besson and Groussot et al remind us that we have not solved the old challenges of the doctrine of the sources of EU law, and its main features, such as legality. They attest to the fact that our expectations and the Court’s practice are not aligned, especially with regard to the Court’s duties as a constitutional court in elaborating a coherent legal system. At the same time, they provide benchmarks and novel solutions. They bring new and exciting arguments to the table. They deserve to be read carefully and assessed critically. After all, general principles will evolve and revolve around the use of legal reasoning by Courts and jurists in concrete cases or contexts of application, where legality and the rule of law is not a given, but rather a process that oscillates between the requirements of legal certainty and the quest for coherence, over time and over cases.

24  ‘[I]n the first case that comes before it, the Court will establish the doctrine as a general principle but suggest that it is subject to various qualifications; the Court may even find some reason why it should not be applied to the particular facts of the case. The principle, however, is now established. If there are not too many protests, it will be re-affirmed in later cases; the qualifications can then be whittled away and the full extent of the doctrine revealed.’ TC Hartley, The Foundations of E ­ uropean Union Law: An Introduction to the Constitutional and Administrative Law of the European Union, 7th edn (Oxford, Oxford University Press, 2010) 74. 25  Burley and Mattli (n 23) 68–69.

52 

5 Is Legality a Principle of EU Law? ALEXANDER SOMEK

I. Introduction EU law claims to belong to the genus of law. This claim is implicit in every piece of legislation and in every judicial decision. Hence, raising the question of whether legality—ie being law—is a principle of EU law is like asking of a particular horse whether being a horse is part of what it is. What does such a seemingly paradoxical question put into question? Horses have hooves, nostrils and long tails, but if being a horse is not part of a horse, then the respective horse is not what it purports to be. It is something that perhaps merely wishes to pass as a horse, whereas in fact—deep down—it is something different, possibly some mechanical artifice or a donkey. Similarly, if legality accounts for what makes some organised human activity into law, then a legal system that fails to exhibit elementary features of legality cannot be really law. It is merely wearing vestiges thereof. The term ‘legality’ is used in a variety of ways.1 The best-known Anglo-­ American usage is associated with Lon Fuller’s reconstruction of the ‘inner’ or

1 Legality can stand for what it has stood for ever since Kant introduced the notion to legal ­ hilosophy. In this understanding, legality designates the quality of action to be merely and simply in p conformity with law. You act legally as long as you obey norms ‘externally’. That is to say that you act without endorsing as correct, or obtaining guidance from, the norms that you happen not to transgress. The absence of endorsement is manifest in two forms of indifference, namely, on the one hand, in the irrelevance of the moral merit of laws for observance and, on the other, in allowing conduct to be actually motivated by any reason or attitude that is sufficient to give rise to obedience. Thus understood, legality is a second-personal concept. See, generally, S Darwall, The Second-Personal Standpoint: Morality, Respect, and Accountability (Cambridge, MA, Harvard University Press, 2006). It points to what you do not owe to an authority that regulates your conduct. You do not have to conform by having your conduct guided by the norm, let alone have it guided on the basis of your rational insight into its merits. Expressed in Razian parlance, this means that the law does not expect you to comply out of respect for the exclusionary reasons for action that constitute rules. See J Raz, Between Authority and Interpretation: On the Theory of Law and Practical Reason (Oxford, Oxford University Press, 2009) 144. The attitude towards law is not an issue. Even if one acted with the intention of doing something illegal, one still acted legally if the conduct turned out not to be proscribed. Kantian legality goes beyond, but

54 

Alexander Somek

‘internal’ morality of law. The precepts of this morality are addressed to people making it their business to ‘subject human conduct to the governance of rules’. Fuller refers to these precepts also as ‘principles of legality’.2 Since these principles basically draw out some commendable features of the rule of law they represent what the law can be if it lives up to this standard of political morality. ‘Legality’ designates all or some of the virtues of the rule of law. Not surprisingly, Fuller’s list gives us the ordinary trivia: 1. 2. 3. 4. 5. 6. 7.

The law is composed of rules and not of a motley array of ad hoc decisions. The rules are publicly known. The rules are known in advance of action and there is no retroactive legislation. Rules are comprehensible. Rules do not contradict one another and are not internally contradictory. Rules do not demand the impossible. Rules are not changed so frequently that taking guidance from them becomes impossible. 8. There is congruence between the announced rules and their administration. The following analysis engages with EU law by asking whether legality, thus understood, accounts for what it truly is.3 The question suggests itself in light of the notoriously rather surprising divinations of law arrived at by the Court of Justice of the European Union (CJEU). They must give rise to doubts concerning the publicness and non-retroactivity of EU law. Moreover, in important

also extends to, the attitude of being uninvolved with regard to the purposes pursued by the l­aw-giver. The legal subject is free to disregard the fact that the legal system may well be a system of plans. On this point, see SJ Shapiro, Legality (Cambridge, MA, Harvard University Press, 2011). This attitude has been famously elaborated—and extended along the horizontal axis towards others—in Oakeshott’s account of the rule of law and of the civic relation. See M Oakeshott, ‘The Rule of Law’ in On History and Other Essays (Indianapolis, Liberty Fund, 1999) 129–79; On Human Conduct (Oxford, Oxford University Press, 1975) 128–29. 2  See LL Fuller, The Morality of Law, 2nd edn (New Haven, Yale University Press, 1969) 197–98. For a recent and most perceptive work on Fuller, see K Rundle, Forms Liberate: Reclaiming the Jurisprudence of Lon L Fuller (Oxford, Hart Publishing, 2012). By contrast, American diehard Hartians, such as Coleman and Shapiro, have whittled down ‘legality’ to a property that can be possessed by abstract entities. It is the property of being law. See JL Coleman, The Practice of Principle: In Defence of a Pragmatist Approach to Legal Theory (Oxford, Oxford University Press, 2001) 84; SJ Shapiro, Legality (Cambridge, MA, Harvard University Press, 2011) 7. Legality means the quality of whatever belongs to the set of entities called ‘law’. The major difference to Fuller’s understanding is that the property ‘legality’ would also extend to norms that fall short of the standards of the rule of law. 3  For the author’s earlier forays into this field, see his ‘The Emancipation of Legal Dissonance’ in H Koch et al (eds), Europe: The New Legal Realism (Copenhagen, Djøf Forlag, 2011) 679–713; ‘Inexplicable Law: Legality’s Adventure in Europe’ (2006) 15 Journal of Transnational Law and Contemporary Problems 627–53. For a different perspective that transmutes Sir John Fortescue’s distinction between governing (gubernaculum) and adjudication (iurisdictio) into a distinction between law that does not abide by higher law and law that does, see D Kochenov, ‘EU Law without the Rule of Law: Is the ­Veneration of Autonomy Worth It?’ (2015) 34 Yearbook of European Law 74–96.

Is Legality a Principle of EU Law?

 55

cases, there appears too little congruence between rules as announced and their administration in practice.4 The question presupposes, hence, where we have to look, and what we have to look for, in order to determine the significance of EU law from a rule-of-law ­perspective.5 Since EU law serves for the most part as an engine for the transformation of the laws of its Member States, this significance is most evocatively revealed in how it makes itself authoritatively known vis-à-vis national laws. It is posited, thus, that one can look deeply into the heart of EU law if one understands what the preliminary reference procedure entails for the legality of EU law. The following contribution would like to explain that—not least owing to the existence of the preliminary reference procedure—in cases where EU law appears to be doubtful, it can be known only by virtue of a decision. Decisions involve choices. Viewing knowledge of EU law as dependent on choices is consistent with conceiving of EU law as indeterminate on the level of principles, for only under this condition can EU law be ‘known’ in choices concerning its substance. Selecting the principles that are right for Europe is a matter of political entrepreneurship. Exercising entrepreneurial agency is the dominion of the same Court that mitigates the impact of its own innovative decision-making. It is only as a result of these mitigation efforts that Fullerian legality enters the picture. More precisely, it is a consequence of the rather peculiar legality of EU law that ordinary legality plays only a limited role. Returning to the analogy used above, it is fair to say that EU law is, hence, like a horse for which being a horse is quite inessential. The essence of law, ordinarily understood, is accidental to EU law. It is essential to EU law to accord to ordinary legality only an ancillary role. The following account of the relation between EU law and legality seeks to offer an alternative to two myths held dear by Europeanists: the belief in the ‘constitutionalizsation’ of EU law and the thesis that such a ‘constitutionalisation’ was derivative of the teleological interpretation of the EC Treaty.

4  See, for example, the case that is usually referred to as the ‘curious Mangold case’: Case C-144/04 Werner Mangold v Rüdiger Helm [2005] ECR I-9982. 5 Owing to its well-known common law parochialism, the Anglo-American use of ‘legality’ is entirely oblivious to the specific meaning that the term has long attained in the Franco-German constitutional tradition. For a useful introduction, see L Besselink, F Pennings and S Prechal, ­‘Introduction: Legality in Multiple Legal Orders’ in The Eclipse of Legality in Europe (Leiden, Wolters Kluwer 2011) 3–10. There, the principle of legality stands for the requirement that acts of public authority be based on an act of parliament, more precisely, on a law passed with the consent of the representatives of the people. Viewed from one angle, this understanding of legality is narrower than Fuller’s, for it seeks to have virtues such as predictability or generality guaranteed by means of parliamentary legislation; from another angle, however, the Franco-German notion of legality sweeps more broadly since it demands more than what one finds on Fuller’s list, namely participation by the people’s representatives or the constitution of administrative powers on the basis of laws. Since Franco-German ‘legality’, nonetheless, partakes for the most part of Fullerian virtues, the discussion will proceed as though its subject were equivalent to the continental principle of ‘legality’. It needs to be conceded, however, that an analysis focusing on the fate of Franco-German ‘legality’ in a multilevel system would have to address the diminishing influence of democratically elected legislatures.

56 

Alexander Somek

II. Impudence? A. Hallstein At first glance, even raising the question of whether EU law embraces ‘legality’ as one of its principles must appear to be cheeky. No less a figure than Walter ­Hallstein, the first president of the European Commission, famously proclaimed that in contrast to previous political and military attempts at European integration, the (then) European Community was entirely committed to creating unity through law. The European Community was a legal community, a Rechtsgemeinschaft that is, and as such a creature of law. Here are Hallstein’s words auf Deutsch: Die Gemeinschaft ist eine Schöpfung des Rechts. Das ist das entscheidend Neue, was sie gegenüber früheren Versuchen auszeichnet, Europa zu einigen. Nicht Gewalt, nicht Unterwerfung ist als Mittel eingesetzt, sondern eine geistige, eine kulturelle Kraft, das Recht. Die Majestät des Rechts soll schaffen, was Blut und Eisen in Jahrhunderten nicht vermochten. Denn nur die selbstgewollte Einheit hat Aussicht auf Bestand, und Rechtsgleichheit und -einheit sind untrennbar miteinander verbunden. Keine Rechtsordnung ohne Gleichheit vor dem Gesetz, Gleichheit aber bedeutet Einheit. Auf dieser Einsicht beruht der Vertrag von Rom, und darum schafft er eine Friedensordnung par excellence.6

This is what Hallstein claims: no longer are violence or subjugation the means for the unification of Europe, but a ‘spiritual’ and ‘cultural force’, namely the law. The ‘majesty of law’ is supposed to accomplish what ‘blood and iron’ have failed to do. Only unity that is freely endorsed by means of legal arrangements is likely to be long lasting. Even if one distrusts Hallstein’s rather overblown characterisations of the power of law, it is fair to say that EU law works legally. There is no better proof than the abundance of litigation concerning the validity of EU legal acts. Indeed, in Les Verts7 the CJEU proclaimed that the Community was based on the rule of law because there is always a review of the legality available for measures adopted by its institutions. This demonstrates that the observance of legal standards is not taken lightly in the Union. More importantly, the CJEU has consistently served as a reliable partner of individuals seeking legal protection against parochial nation states. It must therefore appear to be a bit of a stretch, to say the least, to put into question, even if only hypothetically, that an organization that has become famous for the protection of the rights of its citizens may not embrace legality as one of its foundational principles.

6  W Hallstein, Der unvollendete Bundesstaat. Europäische Erfahrungen und Erkenntnisse (Stuttgart, Econ Verlag, 1969) 33. 7  Case C-294/83 Les Verts v Parliament [1986] ECR 1339, para 23.

Is Legality a Principle of EU Law?

 57

B.  A Matter of Degree Yet, the impression of impudence disappears once we admit of the possibility that there can be law even if the relevant institutional practices do not meet all of the conditions mentioned on Fuller’s list. Indeed, Fuller himself implicitly admitted of the fact that the embodiment of legality may be a matter of degree.8 In this view, achieving legality comprises both a morality of duty and a morality of aspiration, not least since the principles of legality do not merely require forbearance.9 Interestingly, serving well the interests that individuals regard to be protected as a matter of right can be paired with a certain propensity to underachieve from the vantage point of legality. At certain critical moments, the protection of fundamental interests may require some boldness, as not least the story of the Warren Court has shown.10 A court carrying out frequent leaps towards an ‘unknown destination’11 does not necessarily leave the ambit of legality, at least as long as it can claim to resolve conflicts that have emerged in its own jurisprudence.12 But even if the court indeed ‘legislates’ it may do so in the service of a principle that is of greater weight than observing Fuller’s list. Racial equality may be a case in point.13 It follows that the question of whether a system is committed to legality does not necessarily have to be answered in the notorious ‘all or nothing’ fashion. Possibly, the weight given to legality is relatively weak in certain quarters. Legality, ordinarily understood, may readily yield to other principles or, using Dworkin’s original language, yield to other, relatively more important policies. The pursuit of an integration objective by means of judicial decision-making may thus give rise to a rather peculiar form of legality.

III.  European Decisionism A.  Adjudication as Administration The virtues of the preliminary reference procedure are well established. It helps to clarify what would otherwise remain opaque and vulnerable to multifarious

8  The only clear line that Fuller draws concerns legal rules that are kept secret. They cannot pass as ‘law’ in his opinion. 9  See Fuller (n 2) 42. 10  See MJ Horwitz, The Warren Court and the Pursuit of Justice (New York, Hill and Wang, 1999). 11  See JHH Weiler, ‘Journey to an Unknown Destination: A Retrospective and Prospective of the European Court of Justice in the Arena of Political Integration’ (1993) 31 Journal of Common Market Studies 417–446. 12  See, generally, M Eisenberg, The Nature of the Common Law (Cambridge, MA, Harvard University Press, 1999). 13 See Brown v Board of Education, 347 US 483 (1954).

58 

Alexander Somek

expositions. In its judgments, the Court lays down rules governing future cases.14 Precedents provide guidance and unify. Clarification and unification serve the ends of legality. The CJEU renders clear what would otherwise remain shrouded in mystery. As a result of repeated clarifications, the preliminary reference procedure enhances the predictability of law. In contrast to having the law clarified by a highest appellate court, the CJEU approach is less intrusive of Member State sovereignty, while at the same time more focused on the general clarification of laws. This sets the procedure apart from proceedings in which a court is charged with deciding cases. Despite these widely recognised virtues, it is important to recognise what the existence of a preliminary reference procedure says about the law of which it is to serve as the wellspring of insight. Its existence suggests that were it not for this procedure the law could not be known at all. It posits that divinations on the part of the CJEU are essential to knowing EU law. The difference between the CJEU and an appellate system should be duly noted. A supreme court is instrumental in clarifying and consolidating the law. It is, however, in principle, only a primus inter pares, for there is no single court in which the legal system entrusts the prerogative of knowing the law with authority for all others. A supreme court is, when it comes to this task, merely the last in a sequence of contenders. It is not, that is, the one institution on which the legal system confers pre-emptive cognitive authority.15 In principle, the technique of knowing the law by virtue of decisions is not tied to the type of law that EU law represents. This technique could be attached to any other type of law. Even if one imagined a legal system that contained only the prohibition of murder and added to it a procedure enabling references concerning its meaning, people may feel encouraged to inquire what this prohibition might mean on top of what it is ordinarily taken to mean (eg ‘Does it mean that government has to provide food as long as it is in a position to do so?’) The presence of a provision that promises clarification can easily create the incentive for pushing the envelope, even in face of the risk that the institution providing the answer might say ‘no’.

14  See Case 283/81 Srl CILFIT and Lanificio di Gavardo SpA v Ministry of Health, [1982] ECR 3415, para 14. 15  Granted, there are high courts everywhere but their dispensation of clarification is the by-product of deciding final appeals. The basic supposition is that the law can be known, even though this may be difficult and tricky in singular cases. By contrast, the preliminary reference procedure, in which the judiciary is specifically given the task of clarifying the law not only apropos the decision of disputes and by permitting references from all courts, represents a decisive break with the standard self-idealisation according to which the law can be known once it has been adopted and promulgated. For a related observation concerning a ‘null proposition’ according to which legal norms ‘impose no meaningful substantive constraints’ in the exercise of a court’s review authority, see A Stone-Sweet, ‘The Juridical Coup d’État and the Problem of Authority: CILFIT and Foto-Frost’ in M Poiares Maduro and L Azoulai (eds), The Past and Future of EU Law: The Classics of EU Law Revisited on the 50th Anniversary of the Rome Treaty (Oxford, Hart Publishing, 2010) 201–10 at 208.

Is Legality a Principle of EU Law?

 59

This incentive effect will take hold, however, only if the institution answering these questions is disposed to perform innovative moves. If preliminary references ended up on the desks of traditional legalists, their effect would likely remain unspectacular. In the eyes of traditionalists, the legal system does not act. It merely reacts. It is not, in particular, an agent of innovation and change. If properly activated the law reacts appropriately, ie in the manner in which it has been programmed to act. The programme is to be found in laws (and what results from the application of laws is ‘legality’ in the Franco-German sense).16 The whole process of question and answer promises to be lively and rewarding, however, if it is based upon the premise that the programme triggering answers is not yet fully developed or perhaps not yet in place. In that case, the legal system will behave not reactively but proactively. It will begin to churn out new rules and thereby perform the task of a regulatory body. It takes policy entrepreneurs to make this happen.

B.  Two Versions of Indeterminate Law But how could an institution answer questions of law if it needs to bring about the object of its propositions? Wouldn’t this be tantamount to acting in bad faith? The apparent paradox disappears by drawing a distinction between legal ideas or principles, on the one hand, and rules designed to give them effect, on the other.17 Usually, it is taken for granted that the meaning and validity of such rules is derivative of the normative ideas to which they are supposed to give effect. For example, the belief that stealing is wrong represents such an idea or principle; various provisions on sanctions and law enforcement comprise the matching set of implementing rules. In order to design these well, however, we need to have a clear idea of what constitutes theft.18 With these two layers in mind it is fair to say that the law, taken as a whole, can exist without existing in a fully determinate state, for example, as soon as the principles are relatively clear while the implementing rules remain incomplete. It can be said, then, that the law exists in a state of weak indeterminacy. In a sense, this is the situation that defines fundamental rights jurisprudence. Its task is to tease out what fundamental rights demand vis-à-vis interfering state action. The proper means thereto is the development of various ‘tests’ c­ onstraining

16 

See above n 5. more illuminating than Dworkin’s famous writings from the 1960s on this matter is the early work by Carl Schmitt, Der Wert des Staates und die Bedeutung des Einzelnen (Tübingen, Mohr, 1914), in which Schmitt distinguishes between law, on the one hand, and the positive laws of the state that are supposed to give it effect, on the other hand. At one point in his essay (56, 80) it emerges quite clearly that what Schmitt means by law is a layer of principles, which are incomplete norms that do not comprise a method of implementation. 18  See Schmitt, ibid, 52, 71, on the state qua artefact, the sole purpose of which is to realise the law. 17  Much

60 

Alexander Somek

administrative and legislative action.19 Fundamental rights law is indeterminate in that choices have to be made regarding its proper realisation in relation to state action; but it is not one big void either. The situation, however, could also be the reverse. Then, the weak indeterminacy becomes strong. This is the case when the implementing apparatus of rules and enforcement mechanism is in place while the meaning of its underlying principal ideas is up for grabs.20 The task of the legal knower must then be to decide what the principles mean that the apparatus is designed to give effect to. This appears to have been the situation created under the circumstances that allowed EU law to outgrow its original format.

C.  An Inherent Lack of Clarity Indicates Strong Indeterminacy Historically, it is uncontested that the preliminary reference procedure, which had already antedated EU law in the Italian constitutional system, became part of the judicial system of the Treaty of Rome owing to its promise to facilitate the uniform interpretation of Community law.21 The causes for potentially divergent interpretations might have been manifold, ranging from the politically delicate quality of certain subjects to the remoteness of trade issues for the national judiciary. Even if the purpose of the procedure is to guarantee uniformity of application, its existence, nonetheless, allows everyone to presume de iure that Community law, in contrast to other types of law, is intrinsically in need of clarification. Therein lies, of course, a paradoxical, but legal, abnegation of legality. The one thing that can be legally known with certainty about EU law is that, when all conjectures have run their course, it can be known only in decisions arrived at by a designated tribunal. Necessarily, that is, the law has to be known by this tribunal, for otherwise it would, owing to its intrinsic lack of clarity, easily give way to political opportunism or parochially motivated interpretive tinkering. That the law be known in decisions is, thus understood, a requirement of the legality that is peculiar to EU law (and for which ordinary legality is accidental). An intrinsic lack of clarity implies strong indeterminacy. It must concern the meaning of principles and ideas. In the case of weak indeterminacy, clear principles provide guidance as to how to design implementing rules with the aid of means–ends rationality. The absence of implementation is an accidental lack of

19  For a highly perceptive elaboration of this point, see R Fallon, Implementing the Constitution (Cambridge, MA, Harvard University Press, 2001). 20  For example, one may be confronted with provisions on European citizenship and not know what these mean. 21  See A Boerger-De Smedt, ‘Negotiating the Foundations of European Law, 1950–57: The Legal History of the Treaties of Paris and Rome’ (2012) 21 Contemporary European History 339–56 at 351–53; D Tamm, ‘The History of the Court of Justice of the European Union Since its Origin’ in The Court of Justice and the Construction of Europe: Analyses and Perspectives on Sixty Years of Case-law (The Hague, TMC Asser Press, 2013) 9–35 at 19–20.

Is Legality a Principle of EU Law?

 61

determinacy, which can be cured by a judicial tribunal, even though a l­egislature may be somewhat better positioned when it comes to taking cognisance of r­ elevant facts. Once the indeterminacy concerns principles, however, the implementing rules are suspended in thin air. They are thus in danger of being manipulated by anyone coming to the table of interpretation with an interested agenda. Only knowing the law in the form of decisions can contain this danger. It is, therefore, entirely consistent with the existence of then Article 177 EEC Treaty to attribute indeterminacy to Union law on the level of principles. One may therefore lawfully (‘legally’) presume that Union law, in non-decided form, is inherently indeterminate and therefore in need of clarification (which contains a strong element of facere, ie doing).

D.  Strong Indeterminacy Is Not a Given If critical legal studies have taught us anything, it is that legal indeterminacy is rarely ever some semantic condition in which the law happens to exist ‘naturally’. Rather, indeterminacy is a product of a tension between opposing principles or of conflicting accounts of their meaning.22 Such a state of actively constituted indeterminacy of principles seems to match exactly the state in which EU law began, not least owing to the presence of an ambitious legal elite.23 At any rate, this is imparted by a story that historians of Union law tell about its beginnings. After the failure of the French National Assembly to ratify the European Defence Community in 1954, the EEC Treaty was officially supposed to amount to far less than the blueprint of a federal system. Actually, from the perspective of the contracting parties, it just looked like another conventional international agreement and not at all like ‘a new legal order of international law for the benefit of which the states have limited their sovereign rights’.24 Nevertheless, committed European jurists, in particular working in the so-called Groupe de redaction, were able to build into the Treaty a number of elements that had the potential to unfold a more federalising dynamic.25 The preliminary reference procedure was among those elements. While the Treaty was concluded in far more muted spirit and emphasised the role of an intergovernmental Council, the juridical elite favouring a f­ederal

22 See R Mangabeira Unger, The Critical Legal Studies Movement (Cambridge, MA, Harvard University Press, 1983); M Kelman, A Guide to Critical Legal Studies (Cambridge, MA, Harvard University Press, 1988). 23  On the following, see the very helpful hints to be found in M Rasmussen, ‘From Costa v Enel to the Treaties of Rome: A Brief History of a Legal Revolution’ in Poiares Maduro and Azoulai (n 15) 68–85 at 81–83. 24  See Case 26/62 NV Algemene Transporten Expeditie Onderneming van Gend en Loos v. Nederlandse Administratie der Belastingen [1963] ECR 1. 25  See Boerger-De Smedt (n 21) 351.

62 

Alexander Somek

Europe was able to put in something bolder. Quite remarkably, therefore, the addition of the preliminary reference procedure contributed to the ambiguities that it had been designed to resolve. Its presence helped to constitute the indeterminacy that it allowed to address. Initially, Member State governments perceived actions against Member States brought by the Commission or another Member State (under then Articles 169 and 170) as major mechanisms for dealing with treaty violations. Article 177 must originally not have seemed to be capable of giving direct effect to Community law vis-à-vis conflicting national law as part of a ‘dual vigilance’ system.26 This explains, however, why initially objections were frequently made to preliminary references that indirectly concerned conflicting Member State laws, for no small part of the governments took it for granted that the procedures of Articles 169 and 170 offered the appropriate avenues for dealing with compliance matters.27 From that angle, Article 177 was to be used in order to elucidate the meaning of regulations, but not to resolve a conflict concerning sovereign powers. At the same time, it must have appeared also plausible to use Article 177 to that end too. The ambiguity to which Article 177 gave rise, in particular with regard to direct effect and its own status in the EU Treaty system, was not resolved at all on the basis of this very same article. But this proves, of course, the point that from the outset EU law was constituted such as to require clarification by means of decisions. It was not doubt ingenious on the part of the Court to realise that it was in the position of being able to make real choices.28 It self-consciously occupied the spot from which the presumed indeterminacy could be made to feed into the dynamics of European integration.

E.  Known in the Form of Decisions European Union law is not simply known. It is known as determined in decisions. One of the major champions of ‘decisionism’ as a principle of political morality, the German philosopher Hermann Lübbe,29 explains that the validity of decisions

26 See Cases 28–30/62 Da Costa en Schaake NV, Jacob Meijer VN and Hoescht-Holland NV v ­Nederlandse Belastingadministratie [1963] ECR 31. For a wonderful introduction, see S Weatherill, Law and Integration in the European Union (Oxford, Oxford University Press, 1995). De Witte quite perceptively observes that the innovation brought about by the CJEU in the Van Gend case was that the question whether the Treaty was self-executing was decided by the supranational court and not by national courts pursuant to their respective constitutional traditions. See Bruno de Witte, ‘The ­Continuous Significance of Van Gend en Loos’ in Poiares Maduro and Azoulai (n 15) 9–15 at 10. 27  See Rasmussen (note 23) 82. 28  It was, in a sense, a ‘choice for Europe’ to paraphrase the title of Moravcsik’s famous book. A Moravcsik, The Choice for Europe: Social Purpose and State Power from Messina to Maastricht ­(London, UCL Press, 1998). 29  See H Lübbe, ‘Dezisionismus—Eine kompromittierte politische Theorie’ (1976) 55 Schweizer Monatshefte 949–60 at 950.

Is Legality a Principle of EU Law?

 63

does not hinge on their substantive correctness. What makes a decision legitimate is not the reasoning leading up to it, but rather its inherent power of clarification (which explains why decisionists detest nothing more than ‘decisions’ that do not really settle anything). Nobody understood the inherent appeal of decisions better than Carl Schmitt, at least during the period of his career when his works appeared to betray a decisionist outlook. More plainly than Lübbe, Schmitt emphasises that a decision overcomes a situation of disorder or disorientation. In this light, the chief purpose of the law is to ‘limit the unlimited, to determine the relatively indeterminate’.30 Schmitt also had a keen sense of the ex cathedra quality of decisions.31 They are valid because the decider has spoken. This means, negatively speaking, that their mode of validity signals that substantially they could have even come out of nothing. Decisions exhibit a tacit indifference to what grounds them.32 This explains why they are surrounded with a somewhat mystical air. In recent path-breaking studies, Hans Lindahl has developed a vocabulary that captures quite adequately the significance of decisions from a different angle. Lindahl complements the binary of ‘legality’ and ‘illegality’ with the third concept of ‘a-legality’.33 While illegality stands for a breach of established legality, a-legality designates the force instituting the order within which the distinction between legal and illegal can attain significance in the first place. One of Lindahl’s paradigmatic instances of a-legality are acts of founding,34 which are, in his view, ‘anomic’ and transcend the divide of selfhood and alterity.35 Acts of founding are anomic, ‘for they are not themselves the exercise of a right or competence, or in breach thereof ’,36 and they defy the distinction of self and other, for they claim to represent a first-person plural before a collective has been constituted. As acts of self-constitution, they can only be retroactively validated through endorsement (or appropriation) by the collective self they have helped to bring into existence. The normative order of law is, according to Lindahl, ‘the outcome of a process of normalization that has its inception in the abnormal. In the beginning was a-legality.’37 The ‘formless forming’38 and social magma of a-legal acts lays bare the ­‘ineradicable 30  E Cassierer, Symbol, Technik, Sprache: Aufsätze aus den Jahren 1927–1933, ed E-W Orth and JM Krois (Hamburg, Meiner, 1985) 100 (I owe this reference and the translation to Hans Lindahl). 31  See C Schmitt, On the Three Types of Juristic Thought, trans J Bendersky (New York, Praeger, 2004). 32  There is a story to tell about how the indifference towards substance spills over to procedure and jurisdiction. 33  See H Lindahl, ‘Border Crossings by Immigrants: Legality, Illegality, and Alegality’ (2008) 14 Res Publica 117–35. 34 Other examples include insurrection, secession and terrorism. See H Lindahl, Fault-Lines of Globalisation: Legal Order and the Politics of A-Legality (Oxford: Oxford University Press, 2013) 167–77. Practices of this kind are not merely illegal because they contest the collective self that is claimed to be the subject of the order that gives rise to the lines dividing legality from illegality. They reject the viewpoint of the established order. 35  See Lindahl (n 33), 126. 36 ibid. 37  ibid, 155. 38  M Loughlin, ‘A-Legality or Jus Politicum? A Critical Appraisal of Lindahl’s Fault Lines of Globalization’ (2014) 16 Etica & Politica/Ethics and Politics 965–72 at 969.

64 

Alexander Somek

core of groundlessness’ of an instituted order that accounts its ­contingency. The experience of a-legality reminds every order of its lack of necessity. Students of EU law have made frequent encounters with the miraculous quality of a-legality in the leading cases decided by the CJEU. They introduce a perspective, a point of view, and thereby absorb the indeterminacy of the original situation. For example, there appears in a ruling, without explicit and unequivocal anchor in the Treaty, the statement that ‘Union citizenship is destined to be the fundamental status of nationals of the Member States, enabling those who find themselves in the same situation to enjoy the same treatment in law irrespective of their nationality, subject to such exceptions as are expressly provided for’;39 or that ‘the right of individuals to rely on the directly effective provisions of the Treaty before national courts is only a minimum guarantee’.40 All of a sudden a decision concludes by saying that ‘interpreted in this way the provision contains nothing capable of prejudicing the fundamental human rights enshrined in the general principles of Community law and protected by the Court’,41 which is subsequently followed by ‘in fact, respect for fundamental rights forms an integral part of the general principles of Community law protected by the Court of Justice’42 and that ‘the Court has already stated, fundamental rights form an integral part of the general principles of law, the observance of which it ensures’.43 Much earlier in the case law we are confronted with the rather surprising statement that ‘all trading rules enacted by Member States which are capable of hindering, directly or indirectly, actually or potentially, intra-community trade are to be considered as measures having an effect equivalent to quantitative restrictions’.44

F.  The Myth of Teleological Interpretation This way of presenting the genius of EU law will strike its ordinary apologists as counterintuitive. Was not the whole process that Stein and Weiler reconstructed as ‘constitutionalization’45 about taking the EU out of the hands of politicians and 39 

Case C-184/99 Rudy Grzelczyk v CPAS [2001] ECR I-6193, para 31. Case 168/85 Commission v Italy [1986] ECR 2945, para 11. 41  Case 29/69 Stauder v City of Ulm [1969] ECR 419, para 7. 42 Case 11/70 Internationale Handelsgesellschaft v Einfuhr- und Vorratsstelle für Getreide und ­Futtermittel [1970] ECR 1125. 43  Case 4/73 Nold v Commission [1974] ECR 491, para 13. 44  Case 8/74 Procureur du Roi v Benoît and Gustave Dassonville [1974] ECR 837. Why this statement is surprising is beautifully explained by JHH Weiler, ‘The Constitution of the Common Market Place: Text and Context in the Evolution of the Free Movement of Goods’ in PP Craig and G de Búrca (eds), The Evolution of EU Law (Oxford, Oxford University Press, 1999) 349–75. The CJEU has done quite a bit over the last few decades to contribute to the re-enchantment of legal reason. I am borrowing from P Schlag, The Enchantment of Reason (Durham, NC, Duke University Press, 1998). 45  See E Stein, ‘Lawyers, Judges, and the Making of a Transnational Constitution’ (1981) 75 ­American Journal of International Law 1–27; JHH Weiler, The Constitution of Europe (Cambridge, Cambridge University Press, 1999). 40 

Is Legality a Principle of EU Law?

 65

rescuing it by means of the rule of ‘higher law’?46 And aren’t the major decisions by the Court merely, as they themselves indicate, exercises in teleological interpretation? Hasn’t the Court merely drawn out the consequences of goals to which the Member States had committed themselves? Isn’t, plainly speaking, all this talk of the indeterminacy of principles merely outlandish legal theory mumbo jumbo? The teleological approach to Treaty interpretations seems to fit well with what we just found with regard to the federalist agenda of a pro-European elite. Nevertheless, it is nothing short of a convenient myth.47 The myth has its origin in the Court’s self-presentation. In Van Gend we read that in order to answer a question of interpretation we need to consult the ‘spirit’ and the ‘general scheme’ of the Treaty.48 Such a statement about the task of interpretation is, in and of itself, not unusual. According to the Vienna Convention on the Law of Treaties, the teleological interpretation of international instruments is an acceptable way of drawing out their meaning.49 Admittedly, there are some judgments in which a purposive interpretation may indeed have done all the work. The EC Treaty stated clearly that there ought to be a ‘common’ market. Arguably, Cassis drew out the implication.50 If there is a common market, then any good that enters a national market legally also enters all other markets that are part of the common market. Mutual recognition and the obstacle approach to interferences seem to follow by implication.51 If they did not follow, there would be no ‘common’ market but merely a concatenation of national markets. Whether or not one finds such reasoning compelling does not affect the fact that it observes the form of teleological interpretation.

46 

See Rasmussen (n 23) 68. made my first-hand encounter with this myth when Francesca Bignami objected to my usual debunking of the CJEU jurisprudence as social engineering by pointing out to me that it was explicable in terms of teleological interpretation. I hope that in the text I have finally succeeded at formulating some form of a rebuttal. Bignami’s objection is, however, most important, for once we address it we can perceive the EU’s position on legality. For the tenacity of the myth, see E Sharpston, ‘The Shock Troops Arrive in Force: Horizontal Direct Effect of a Treaty Provision and Temporal Limitation of Judgments Join the Armoury of EC Law’ in Poiares Maduro and Azoulai (n 15) 251–64 at 257, where the purposive interpretation by the CJEU is described as unfamiliar to the British stem of the common law tradition. 48  See Case 26/62 NV Algemene Transporten Expeditie Onderneming van Gend en Loos v Nederlandse Administratie der Belastingen [1963] ECR 1; Case 6/64 Flaminio Costa v ENEL [1964] ECR 585; Cases 28–30/62 Da Costa en Schaake NV, Jacob Meijer VN and Hoescht-Holland NV v Nederlandse Belastingadministratie [1963] ECR 31; see also Case 41/74 Van Duyn v Home Office [1974] ECR 1337 49  See Art 31(1): ‘A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.’ 50  Case 120/78 Rewe-Zentrale AG v Bundesmonopolverwaltung für Branntwein (= Cassis de Dijon) [1979] ECR 649 51  For a highly perceptive and slightly revisionist analysis of the relation between Dassonville and Cassis de Dijon that explains quite persuasively why only in retrospect was Dassonville able to be seen as the precedent for Cassis, see N Bernard, ‘On the Art of Not Mixing One’s Drinks: Dassonville and Cassis de Dijon Revisited’ in Poiares Maduro and Azoulai (n 15) 456–64. Prior to Cassis, Dassonville was basically a ‘parallel imports’ case (459, 461). 47  I

66 

Alexander Somek

What one encounters, by contrast, in early cases, such as Van Gend,52 Costa53 and Defrenne II,54 is teleological interpretation in disguise. It is a disguise for what Philip Allot described already in the 1970s as the ‘laconic temerity of the European Court’.55 In order to understand what is really going on in the case law one merely needs to examine the most famous ‘argument’ from Van Gend.56 The functioning of the common market is of ‘direct concern’ to the private parties. This, the Court claims, implies that, subject to certain conditions, the freedoms guaranteed by the Treaty have to have direct effect. This is equivalent to saying that the United Nations Declaration of Human Rights has direct effect because it is of interest to all human beings. Historical hindsight, indeed, confirms the audacity of the position taken by the slim majority of judges in the European Court of Justice (ECJ).57 Several Member State governments, aided by the Advocate General, had argued that Article 12 of the Treaty was only addressed to the Member States and was not, unless the Member State legal system said so, ‘self-executing’.58 Indeed, public international law, as hitherto developed, provided backing for this view. As Eric Stein pointed out, after an Advisory Opinion had been issued by the Permanent Court of International Justice in 1928 it became communis opinio among international law scholars that any Treaty interpretation needs to rebut a general presumption against direct effect. Such a rebuttal had to be based upon ‘explicit evidence of the intention of the parties to the contrary’.59 As is well known, the CJEU did not bother to muster any ‘explicit’ evidence and rather appealed to what it took to be the nature and overall aim of the Treaty. It is easy to be led astray by the Court and to assume that this is just ‘objective teleological interpretation’, ie the Court inferring the aim from the substance

52 

See above n 48. See above n 48. See Case C-43/75 Gabrielle Defrenne v Société anonyme belge de navigation aérienne Sabena [1976] ECR 455. 55  Quoted in Sharpston (n 47) 255. 56  Without apparent irony or embarrassment are these ‘arguments’ classified as ‘well developed’, ‘elegant’ and ‘persuasive’ by P Pescatore, ‘Van Gend en Loos, 3 February 1963—A View from Within’ in Poiares Maduro and Azoulai (n 15) 3–8 at 6. The lack of merit of the other arguments, even though totally ignored by Pescatore, is quite obvious. The Treaty preamble refers to people and not only governments. Hence, it must have real effect for the citizens. This is, as Kelsen had known all along, true of any system of public international law. Pescatore explains that the sovereign rights given up by the member states were those of the nationals and ‘remain those of the nationals’ after their reaggregation at the Community level. This is utter nonsense. The power to legislation does not belong to individual citizens. 57  It was indeed by the margin of one vote (4:3) that the Court decided in favour of direct effect. The jury is still out on whether the substance of the decision reflects the position of the European Commission, influenced by Michel Gaudet as the Director of Legal services, or springs directly from the ingenuity of Robert Lecourt. For the view that perceives the influence of the Commission as essential, see Stein (n 45) 17. Pierre Pescatore is of a different opinion. See Rasmussen (n 23) 77. 58  See Pescatore (n 56) 4. 59  Stein (n 45) 459. 53  54 

Is Legality a Principle of EU Law?

 67

of the Treaty in order to thereby invest Article 12 with self-executing force.60 The Court flatly disregarded the old principle that the actual intent of the contracting parties counts. What the in fact Court did with its reference to the aim of the common market and interests of private parties was to construct the equivalent of the intent of the substance of Community law. The Court effectively attributed to the substance of Community law the intention to confer direct effect on itself. As Bruno de Witte correctly observes, the Court therewith not only reversed the traditional presumption against direct effect, it invested itself with the jurisdiction to determine the self-executing character of Community law without paying heed to national law regulating the interaction between domestic and international law.61 In other words, the Court inferred the legal form of Community law from its content. Even in retrospect this must appear miraculous, for it defies juridical common sense. If my neighbour and I conclude a covenant the point of which is to secure the tranquillity in the neighbourhood, the substance of the agreement implies that all other neighbours ought to be bound by it too. Yet, we do not have the legal power to bind them, just like the signatory Member States of the EEC merely had the power to conclude an international agreement and not to install some ‘higher law’.

G.  EU law as Natural Law: The Myth of ‘Constitutionalisation’ This same inference is repeated in Costa62 and in Defrenne.63 In Costa, the Court carried it out with greater clarity and grace. Among other appeals to the overall nature of the Community, the reasoning again focused on the substance of the right in question, namely the right to be free from discrimination on the ground of nationality, in order to conclude that if supremacy were not added to direct effect such discrimination could not be prevented because of divergent national approaches to the force of international obligations. In Defrenne, the strategy of the Court became even more obvious. After claiming that the principle of equal pay serves the dual aim of, firstly, avoiding competitive disadvantage in intraCommunity competition and, secondly, ensuring social progress, the Court concluded that it had to have direct effect in cases of direct discrimination. Otherwise these ends could not have been served. It is important to understand that teleological interpretation is indeed involved in the context of calibrating the meaning of equal pay. But ascribing to it direct effect is simply a non sequitur. The operation at work, in the case of Defrenne,

60  See B de Witte, ‘The Continuous Significance of Van Gend en Loos’ in Poiares Maduro and ­Azoulai (n 15) 9–15 at 11. 61  ibid, 10. 62  See above n 48. 63  See above n 54.

68 

Alexander Somek

is therefore more accurately described as ‘supranational self-extraction’.64 In the hands of the CJEU, the substance of an international agreement results in ­‘unpacking’ its direct effect. It is as though Community law said to itself, in an act of application, that because of how it perceives its own task it has to have more effect than has been given it by the High Contracting Parties. As we have already observed above, if the principle of supranational self-extraction were a juridical principle, it would apply, in particular, to all human rights documents. Our understanding of supranational self-extraction suffered at the hands of public international law scholars harbouring relatively superficial ideas about constitutional law. It is one of the greatest intellectual travesties that Stein and Weiler described the introduction of direct effect as part of a ‘constitutionalisation’ of EU law.65 They thereby overlooked the obvious. Attributing to some norm the rank of higher law on the ground of its substance is the epitome of a natural law argument.66 This is what the court did in the case of the fundamental freedoms guaranteed by primary law. It applied them as though they were natural law. The Court’s jurisprudence amounts to no more, and no less, than liberal natural law theory in the guise of positivistic legality.67 Stein and Weiler were taken in by the Court’s rhetoric. It explains why they were giving us at best half of the story. What they de-emphasized is that the alleged ‘constitutionalisation’ of EU law amounted to a deconstitutionalisation of established understandings with regard to how constitutions and international agreements interact. Worse still, ‘constitutionalisation’ talk is entirely oblivious to one major feature of modern constitutions. They require that forms be observed—procedure, voting rules—in order to attain certain legal effects. What the rulings by the Court amounted to was attaining higherlaw effects by unconstitutional means—and not the observance of constitutional discipline.68 In this respect, Alec Stone-Sweet is not incorrect in referring to this jurisprudence as a ‘judicial Coup D’État’, by which he means ‘a fundamental transformation in the normative foundations of a legal system through constitutional law-making by a court’.69 The new constitutional law would have been rejected by the founders as it not longer represented what they intended to be law.

64  See A Somek, Individualism: An Essay on the Authority of the European Union (Oxford, Oxford University Press, 2008) 215. 65  See above n 45. 66  Arguably, therefore, the constitutionalisation thesis bespeaks the ignorance of the natural law tradition on the part of its purveyors. Traditionally, the quality of being ‘higher law’ is the attribute of natural law. See BH Bix, Jurisprudence: Theory and Context, 7th edn (Durham, NC, Carolina Academic Press, 2015) 69. 67  Max Weber observed quite perceptively that ‘faith in legality’ (Legalitätsglaube) is a source of de facto legitimacy. See M Weber, Staatssoziologie, 2nd edn (Johannes Winckelmann, Berlin, 1988) 28. 68  I have tried my hand at critiquing the concept of “constitutionalization” in a different instalment. See my ‘Constitutionalization: Constitution-Making for Individualists’ in S Puntscher Riekmann and D. Wydra (eds), Is there a European Common Good? (Baden-Baden: Nomos, 2013) 95–119. 69  Stone-Sweet (n 15) 202.

Is Legality a Principle of EU Law?

 69

H.  Entrepreneurial Authority At the same time, the notion of ‘judicial Coup D’État’ does not capture accurately the ethos underpinning this enterprise. Evidently, there is no sudden seizure or capture of authority. One encounters, rather, diligent judicial entrepreneurship. One needs to be precise at this point. Entrepreneurs innovate and thereby rise above market conditions, such as the elasticity of prices, by which ordinary market participants operate. Entrepreneurs generate demand by producing ‘wow-effects’. Folks want what the entrepreneur has created as soon as they have seen it, eg the iPhone, the iPad, the Tesla Model 3. Historically, at any rate, exercising entrepreneurship must have crossed the mind of Michel Gaudet, then the Director of the Commission’s legal services. When confronted with the question posed in the Van Gend case, he advocated strongly that a ‘legal innovation’ be made, an innovation of the type that would effectively transform part of the EEC Treaty into a Bill of Rights for individuals; hence, direct effect and supremacy.70 Gaudet was part of a small group of pro-European elite jurists pushing integration ahead in the face of what they may have perceived to be the alternative of ‘ruin’.71 There is something decidedly authoritarian about moving ahead in this ­manner.72 First, the elite self-righteously assumes that their principals would ­confer authority to them had those principals—the European citizens—only correct insight or an opportunity to voice their views. Presumably, those citizens are also homogeneous, for it is taken for granted that everyone shares that interest. Second, the elite is plunging ahead through action, ie by means of a fait accompli. It is done. The law does not serve a constraint; rather, it provides a means of action. For all intents and purposes, the Court has repeatedly proven its readiness to do with law whatever was necessary to assert the supremacy of the Union even if this involved emulating German fundamental rights jurisprudence without availing of a constitutional source. The Court, under these circumstances, did not hesitate to embrace what friendly critics describe as a ‘rationalist human rights framework’.73 Once more, it is important to see the difference between legal ideas and the implementing rules.74 In the eyes of the elite, the written law may only imperfectly serve the ends they wish the law to accomplish. Hence, new programmes have to be

70 

See Rasmussen (n 23) 76. See ibid, 84. 72  For an elaboration of my use of ‘authoritarian’ here, see my ‘Delegation and Authority: Authoritarian Liberalism Today’ (2015) 21 European Law Journal 340–60. 73  See M Kumm, ‘Internationale Handelsgesellschaft, Nold and the New Human Rights Paradigm’ in Poiares Maduro and Azoulai (n 15) 106–18 at 110. 74  See C Schmitt, Die Diktatur: Von den Anfängen des modernen Souveränitätsgedankens bis zum proletarischen Klassenkampf, 4th edn (Berlin, Duncker & Humblot, 1978) xvi. 71 

70 

Alexander Somek

developed. This explains why the CJEU proceeds as though whatever is necessary to create ‘more Europe’ already is EU law.75

I.  Legality Enters only Later Evidently, ordinary legality is immaterial to the progress of legal innovation. It can enter only subsequently in order to mitigate its effect. It becomes accidental. This is what happened first in Defrenne II 76 and subsequently with greater clarity in Barber.77 Defrenne II represents the first attempt to render ‘integration through law’ consistent with the rule of law. It implicitly recognises the unpredictability of the Court’s decisions by limiting the application of new interpretations to future cases. Evidently, this likens the operation of courts to a legislature.78 While it is possible to make out principles guiding a decision concerning limiting the effect of an interpretation to future cases, it is difficult to apply them in practice. Barber tried to flesh these out clearly.79 First, the restriction of temporal effects to the future must be made in the ruling containing the interpretation. Second, the restriction may be only imposed if there is a risk of serious economic repercussions owing to a large number of affected legal relations that were entered upon in good faith. Third, neither private parties nor state authority must have failed to conform with community law owing to objective and significant uncertainty concerning its implications, an uncertainty to which the Member States and Union institutions may have contributed. The latter factor is also relevant to the important ruling in Köbler 80 that represents the limits set on the responsibility of Member States for judicial misapprehensions of EU law, in particular, whether or not EU law is indeed perceived to be in need of clarification by the CJEU.81 We do not have to explore these matters any further here. It is clear that doctrines like these reintroduce legality into a context where judicial decision-making is first used in an entrepreneurial manner in order to advance integration.

75  Quoting from Schmitt, ibid, xviii: ‘Denn eine weitere Eigenart der Diktatur liegt in F ­ olgendem: Weil alles berechtigt wird, was, unter dem Gesichtspunkt des konkret zur erreichenden Erfolges betrachtet, erforderlich ist, bestimmt sich bei der Diktatur der Inhalt der Ermächtigung unbedingt und ausschließlich nach Lage der Sache; daraus entsteht eine absolute Gleichheit von Aufgabe und Befugnis, Ermessen und Ermächtigung, Kommission und Autorität.’ Schmitt continues with quite perceptive observations on the role of the Commissioner. See, for that matter, also S Puntscher ­Riekmann, Die kommissarische Neuordnung Europas: Das Dispositiv der Integration (Vienna, Springer Verlag, 1998). 76  See above n 5454. 77  See Case C-262/88 Douglas Barber v Guardian Royal Exchange [1990] ECR I-1889. 78  On the relevant contemporary commentaries, see Sharpeston (n 47) 59. 79  See n 77 and, more recently, Case C-292/04 Meilicke [2007] ECR 1835. Sharpeston (n 47) 263. 80  See Case C-224/01 Gerhard Köbler v Republik Österreich [2003] ECR I-10239. 81  I have tried my hand at making sense of this ruling. See ‘Inexplicable Law: The Adventures of Legality in Europe’ (2006) 15 Journal of Transnational Law and Contemporary Problems 627–53.

Is Legality a Principle of EU Law?

 71

IV.  The Present A.  Two Questions Much ink has been spilled recently on how the European Union has pushed the boundaries of its jurisdiction in the course of managing recent crises.82 It is undoubtedly true that in the face of various crises the prospects for legality are particularly bleak. The reason is at least twofold. First, if the adjudication of questions of competence involves a high degree of technical expertise, then the reviewing body is likely to abdicate authority.83 Second, if the survival and integrity of the entity that is supposed to guarantee legality is at stake and if this survival can only be guaranteed by ‘stretching’ the entity’s powers, then legality will always come out second after taking effective action.

B.  Crowding Out Legal Expertise With regard to the first reason, the question at issue in the Outright Monetary Transactions case may serve to illustrate the point.84 Legally, the acquisition, on secondary markets, by the European Central Bank (ECB) of bonds issued by states that benefit from loans administered through the European Financial Stabilisation Mechanism or the European Stability Mechanism does not exceed the powers of the Bank if such purchases can pass muster as exercises of ‘monetary policy’. While direct purchases are prohibited by Article 123(1) TFEU for the reason that states would thus be in a position to finance their debt below market rates, the question is whether the acquisition of such instruments by the ECB does not in fact also help Member States to finance their budget independently of capital markets. If that were the case, the bond purchasing policy would violate Article 123(1) and at best amount to the pursuit of economic policy, which is not within the ambit of the ECB’s powers. Yet, that same Article does not rule out that the ECB purchase bonds on s­ econdary markets in the pursuit of monetary policy, eg with the aim of ­influencing interest rates.85 Once the ECB buys, the bond prices go up and the return falls. Indeed, the

82  See, for example, M Ruffert, ‘The European Sovereign Debt Crisis and European Union Law’ (2011) 48 Common Market Law Review 1777–806; K Tuori and K Tuori, The Eurozone Crisis: A Constitutional Analysis (Cambridge, Cambridge University Press, 2014). 83 For a similar observation, see K Tuori, European Constitutionalism (Cambridge, Cambridge ­University Press, 2015) 223. 84  For commentaries that I found helpful, see MA Wilkinson, ‘The Euro is Irreversible! … Or is it? On OMT, Austerity, and the Threat of “Grexit”’ (MS, London, 2015); A Steinbach, ‘The Legality of European Central Bank’s Sovereign Bond Purchases’ (2013) 39 Yale Journal of International Law Online 15–31. 85  See Steinbach (n 84) 22.

72 

Alexander Somek

argument can be made that the purchases in question help to correct widening bond spreads that exceed a reasonable level.86 The measure of the ECB can thus be presented as correcting a de facto undervaluation of government bonds. If the ECB consequently merely buys bonds at a price that accurately reflects actual market risks, no financing of Member States by the ECB is taking place. Viewed from this angle, the action by the ECB hence does not go beyond the mandate to pursue monetary policy and flow out into the domain of general economic policy (in the context of which transfers to a Member State are made in order to stabilise the economy).87 Whether or not either is indeed the case is the decisive issue. Generally, competence allocation issues involve a basic feature of ‘naïve action theory’ that has been analysed (not too gloriously) by Michael Thompson.88 It is manifest in explaining what we are doing by saying that we are doing something else. ‘Why are you pressing this button?’ ‘Cause I am preparing a cup of cappuccino.’

Competence allocation jurisprudence brings means–ends relations into perspective. One action is cast as the means for carrying out another action that counts as exercising the allocated legal power. Whether or not the bond purchases in question constitute a measure of monetary policy turns on the seemingly empirical question whether the ECB creates a market equilibrium that markets have not brought about. Not only must one wonder how a judicial tribunal should ever be in a position to answer the question, the question is how a judicial tribunal should ever be in a position to assess the trustworthiness or reliability of competing economic expertise on this issue. Not by accident, courts are taking cover by adopting a ‘hands off ’ approach (even the BVerfG did so in the so-called Honeywell case).89 They gladly defer to institutions that have to have power to pass judgment on whose expertise is credible enough in order to provide a basis for action. A general lesson can be drawn from this. The more reviewing bodies find themselves thrown out of their proper field of legal expertise, the more likely are to grow the powers of institutions that these bodies have jurisdiction to review. Law cannot constrain where controversial technical expertise is all there is.

C.  Saving the Polity There are many cases with which one could illustrate the second reason, but the underlying issue can be clarified also on the basis of three very simple considerations. 86 

See ibid, 23. This is what had been suspected by the BVerfG, 14 January 2014, 2 BvR 2728/13. 88  See M Thompson, Life and Action (Cambridge, MA, Harvard University Press, 2008). 89 In Honeywell, the Court stated that an ultra vires review of acts of European institutions ought to respect the power of the European Court of Justice to interpret the Treaty. See BVerfG, 6 July 2010, 2 BvR 2261/06. 87 

Is Legality a Principle of EU Law?

 73

Standard explanations for the absorptive effect of delegations and, a fortiori, of the exercise of emergency powers point to empirical factors, such as judicial reluctance to enforce relatively vague constraints when the salus publica is at stake.90 Judges cannot rely on the authority that comes from winning elections. Timing also prevents effective control. In cases of emergency, the executive branch is not only always ready to take action, it is also prone to rationalise it with reference to alleged ‘inherent powers’ or statutes that stem from a different age.91 What is more, even if some pushback by the legislature follows (which will likely demand respect for fundamental rights), eventually any post factum oversight is likely to be rather weak. Unless the executive branch committed atrocities, it is likely to benefit from creating a fait accompli.92 It is also the case that judicial review finds itself disabled by the use of different legal forms. For example, the highly intrusive Memoranda of Understanding concluded under the auspices of the European Stability Mechanism undermine de facto the principle of conferral to which the Treaty pays its lip service.93 Similar— and perhaps even more troubling—developments have occurred in the context of the reform of the preventative and corrective part of the Stability and Growth Pact, in particular with the introduction of the ‘European Semester’.94 The comprehensive review by the Commission of what amounts essentially to the complete public policy of a Member State from the perspective of fiscal stability strips all states of the last remaining nucleus of sovereignty. What is more, both the seductively clear, yet also opaque, language of the relevant regulations and the legally largely unguided use of a ‘scoreboard’ of macroeconomic indicators by the ­European Commission95 in order to identify imbalances demonstrate that their existence 90  The problem is even recognised by D Dyzenhaus, The Constitution of Law: Legality in a Time of Emergency (Cambridge, Cambridge University Press, 2006) 31, 230. 91  See EA Posner and A Vermeule, The Executive Unbound: After the Madisonian Republic (Oxford, Oxford University Press, 2010) 44. On the problem of ‘grey holes’ that purport to formulate constraints while being, at the same time, insubstantial, see Dyzenhaus (n 90) 42. 92  See Posner and Vermeule (note 91) 53. 93  See Tuori (n 83) 214–15. 94  See Art 2(a) of Regulation (EC) No 1466/97. The European Semester includes the formulation, and the surveillance of the implementation, of the broad guidelines of the economic policies of the Member States and of the Union (broad economic policy guidelines) in accordance with Art 121(2) TFEU; the formulation, and the examination of the implementation, of the employment guidelines that must be taken into account by Member States in accordance with Art 148(2) TFEU (employment guidelines); the submission and assessment of Member States' stability or convergence programmes under that Regulation; the submission and assessment of Member States’ national reform programmes supporting the Union’s strategy for growth and jobs and established in line with the broad economic guidelines, with the employment guidelines and with the general guidance to Member States issued by the Commission (the annual growth survey) and the European Council at the beginning of the annual cycle of surveillance; and surveillance to prevent and correct macroeconomic imbalances under Regulation (EU) No 1176/2011. For an introduction to the various pieces of legislation and the relevant procedures, see www.consilium.europa.eu/en/policies/european-semester/ european-semester-key-rules-and-documents/. 95  See Arts 3 and 4 of Regulation (EU) No 1176/2011 of the European Parliament and of the C ­ ouncil of 16 November 2011 on the prevention and correction of macroeconomic imbalances. See http:// ec.europa.eu/economy_finance/economic_governance/macroeconomic_imbalance_procedure/ mip_scoreboard/index_en.htm.

74 

Alexander Somek

merely opens the gate to the free play of economic expertise ridding itself of all constitutional shackles.96 Indeed, as Touri observed,97 it shows that the desired and required activity is in fact not susceptible to regulation—in Aristotle’s terms not capable of being governed—because it requires situational adjustment on the basis of ‘sound judgement’. No longer are we dealing with rule-guided conduct, but rather with strategies of management. By their very nature, these are intrusive and not easily bothered by normative constraints. They may appear in the disguise of recommendations; yet, make no mistake, they are essentially commands. But there is an even more troubling explanation why constitutional law cannot bind. Paradoxically, it does not want to. In order to realise this unwillingness on the part of the law one must only perceive how small the difference is between constitutional legality and acting under the impression of exigency. A good place to begin is McCulloch v Maryland.98 It was in this memorable opinion that Justice Marshall justified the establishment of the Bank of the United States with regard to the vast number of powers of the Federal Government. More precisely, Marshall invited imagining the totality that emerges from considering these powers as a whole: look at how big this Federal Government is and you will realise that it has to have a little extra power in order to create the Bank of the United States.99 Implied powers doctrine is all about a ‘has to have’, implying ‘therefore it has’, for without this entailment the entity concerned would be impossible or unable to sustain itself. It is as though the government said: Look at the task and take into account how great we are. We have to have the power to take up this task, too. Therefore, we have this power.

It does not take much to take this reasoning a step further in the face of challenges, for example, in the following ways: Look at this challenge and take into account how great we are. We have to have the power to address it, too. Therefore, we have it. Look at this challenge. It is a real threat. If anyone does, we have the capacity to take it on. Hence, we have to have power. Therefore, we have it.

What distinguishes the hypothetical reasons in the examples above from what Marshall claimed in McCulloch is the intervention of the normativity of distress. It simply demands that someone do something. To the perceived necessity of action one merely needs to add the greatness or comparatively greater capacity of the entity and all of a sudden powers flow and grow from one situation to the next. More precisely, and more intriguingly, the impersonality of the n ­ ormativity 96  For an attempt, however, to give them Member States a ‘heads up’, see European Commission, ‘Making the Best Use of the Flexibility Within the Existing Rules of the Stability and Growth Pact’ COM(2015) 12 final provisional. 97  See Tuori (n 83) 217. 98  17 US 316 (1819). 99  For a helpful commentary, see A Reed Amar, America’s Unwritten Constitution: The Precedents and Principles We Live By (New York, Basic Books, 2012) 26–27.

Is Legality a Principle of EU Law?

 75

of distress—someone has to do something—is transitive with regard to the justification. The agent who claims to have greater capacity than others will say that the situation requires that x be done because the situation cannot be left unaltered. Someone has to do it and the agent makes himself available100 because he is better positioned than others. Indeed, this is how authorship for powers that emerge from emergencies or delegations can present itself as entirely impersonal and anonymous. What has to be done has to be done. Constitutional legality is aware of the distinctions between tasks and challenges and between choices and manifestations of situational necessities. It is not really unhappy about moving beyond implied powers. It observes the logical progression from securing pedigree to marking the ability to protect the polity against disaster and has neither power nor desire to stop it.

V. Conclusion Do we have an answer to the question whether legality is a fundamental principle of EU law? The answer to this question is as good as the standard answer that can be given to any legal question, namely: maybe.101 It depends on how you look at it. Ordinary legality is invoked in order to mitigate that impact of entrepreneurial decision-making by the Court. It is accidental. At its core, however, the EU is different from a constitutional democracy: project-driven, elitist and with little qualms about embracing authority without authorisation. This is the legality that is peculiar to EU law.

100  Arnold Schönberg wrote in one of his letters that while he was serving in the military an officer asked him whether he really was ‘that’ Arnold Schönberg; to this question he, Schönberg, claimed to have replied that, yes, he was the man: someone had to be that and he had made himself available for it. See A Schönberg, Briefe, ed E Stein (Mainz, B Schott’s Söhne, 1958) 301 (19 September 1949). 101  M Richard Fischl and J Paul, Getting to Maybe: How to Excel on Law School Exams (Durham, NC, Carolina Academic Press, 1999).

76 

6 General Principles and the Many Faces of Coherence: Between Law and Ideology in the European Union XAVIER GROUSSOT, JÖRGEN HETTNE AND GUNNAR THOR PETURSSON

I. Introduction In the light of the constitutional legal developments by the Court of Justice of the European Union (hereinafter referred to as ‘the Court of Justice’ or ‘the CJEU’) as regards protection of fundamental rights, subsequently accepted by the authors of the EU Treaties (see Article 6.3 TEU), the Court can today be correctly described as the constitutional court of the EU. Against this backdrop, the CJEU must take its constitutional responsibility seriously and strive for a coherent development of EU law.1 It is important to keep in mind that the case law of the CJEU must be applied in all courts and tribunals across twenty-eight Member States. This means that its decisions must provide appropriate specificity for the effective resolution of the dispute in front of it and articulate workable general principles for widespread subsequent application. This is important also when the Court develops principles of administrative law which, in addition to the development of a sound culture of administration within the Union, have an impact on the application of national administrative law in situations governed by EU substantive law. But the general principles have many faces: they are not only substantive but are also structural, ie they respond to the need of the system.2 What is the dividing line between substantive and structural principles? This contribution aims at discussing this specific issue in the aftermath of ­Opinion 2/13, which is in our view essential to understanding the relationship between the various types of principles in the EU legal order.3 To do this, we will 1  See N Nic Shuibhne, The Coherence of EU Free Movement Law—Constitutional Responsibility and the European Court of Justice (Oxford, Oxford University Press, 2013), 11. 2  S Prechal, ‘Protection of Rights: How Far?’ in S Prechal and B van Roermund, The Coherence of EU Law: The Search for Unity in Divergent Concepts (Oxford, Oxford University Press, 2008) 156. 3  Opinion 2/13 of the Court (Full Court) EU:C:2014:2454.

78 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

look first at the issue of ‘substantive coherence’. This is connected to the reach and function of the morally laden principles. We will see that coherence can take many forms when it comes to general principles.  Section III will focus on the study of a specific substantive principle: good administration. This principle is now codified in Article 41 of the EU Charter of Fundamental Rights. It constitutes a perfect example to study the relationship between substantive and structural principles since this right embodies in its very core elements the (structural) principle of effectiveness. The last section will consider the issue of ‘ideological coherence’ by using Opinion 2/13 as a starting point of discussion. This Opinion relies heavily on effectiveness and mutual trust in order to justify the rejection of the accession agreement to the European Convention on Human Rights (ECHR).

II.  Framework of Substantive Coherence for General Principles A. Introduction As long ago as 1960, Pescatore wrote: ‘Principles transform the law into a coherent system.’4 Employing the vocabulary of Ronald Dworkin, Tuori states that coherence can be due either to policies or to principles sensu stricto, such as fundamental rights.5 The fact that the Treaty of Rome bypassed the formal relations among acts of secondary legislation can be explained by the prevalence of a substantive order, brought about by the policy orientation of the Treaty and the secondary legislation it authorised. All the legislative acts were originally expected to promote the establishment of the common market as the main objective of the Treaty. The order of European law was conceived of a policy-based coherence. However, Tuori also points out that subsequent developments have complicated the situation. Fundamental rights and reference to universalist values (Article 2 TEU) have complemented policy-based coherence with principle-related elements.6 Furthermore, also in state legal orders substantive coherence has complemented and partly even replaced formal principles of order. The increasing openness of national legal orders to norms of international or transnational origin has made it more and more difficult to stick to a Kelsenian hierarchical conception. For its part, policy-oriented European law, claiming direct effect, supremacy and integration, has induced confusion in the hierarchical structure of the national legal order. Substantive coherence, produced by morally laden general legal principles,

4  See P Pescatore, Introduction à la science du droit (Luxembourg, Office des Imprimés de L'État, 1960). 5  K Tuori, European Constitutionalism (Cambridge, Cambridge University Press, 2015) 46. 6 ibid.

General Principles and the Many Faces of Coherence

 79

such as those related to fundamental rights, has become ever more vital for the functioning of the national legal order. In clashes of authority between European and national law, often two different principles of order confront each other: the policy orientation of European law and the principle orientation of national law. This is one way of depicting the reservations that national constitutional courts have made to the supremacy of European law. Constitutional courts defend the principle-based foundations of national law against the policy-motivated claims of European law.7 In this context, Tuori asserts that the understanding and functioning of a legal order cannot only be considered in hierarchical terms, but also substantive terms, which is useful when studying EU law. Substantive order, coherence, is produced through legal principles permeating the norm material.8 In fact, even if factually the very reason behind the ‘success’ of general principles, (including fundamental rights) in EU law is well-known, both their existence and content remains somewhat unclear. Even their methodological framework comes across as somewhat mysterious.9 However, that does mean that they are without any force—quite the contrary.10 It is against this background that in this chapter the general principles of law—and fundamental rights in general—are considered in the light of coherence. In fact it may be looked upon ‘as an attempt to inject coherence into the polycentricity of legal sources and the pluralism of legal orders and systems’11 such as that of the EU. Interestingly, the concept itself—coherence—is multifaceted and may reflect the different functionalities of the general principles, including their authority, applicability and inter-legality. These functions/functionalities may be described as: creative coherence, normative coherence and transnational coherence.

B.  Creative Coherence The genesis of fundamental human rights protection in EU law is by now well known. The Court of Justice gradually developed and acknowledged that fundamental rights were indeed to be used as the yardstick according to which both measures adopted by the EU institutions and finally also measures of the Member States falling within the scope of EU law were to be measured. Principles do not fall from heaven and creative coherence stands for the role of fundamental rights

7 ibid. 8 ibid.

9  X Groussot and GT Petursson, ‘The EU Charter of Fundamental Rights Five Years On—The Emergence of a New Constitutional Framework’ in S de Vries, U Bernitz and S Weatherill (eds), The EU Charter of Fundamental Rights as a Binding Instrument—Five Years Old and Growing (Oxford, Hart Publishing, 2015). 10  See eg the number of recent cases of invalidation of EU norms, in light of fundamental rights, post EU Charter, Volker, Test, Digital Rights, Facebook discussed below. 11  K Tuori, Ratio and Voluntas—The Tension Between Reason and Will in Law (Farnham, Ashgate, 2011) 165 (ch 5 ‘The Law´s System—From Total to Local Coherence’) 314.

80 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

in EU law, formulated by the Court of Justice, as a tool to ‘ensure observance of the law’ (Article 19 TEU)—adding flesh to the bones of EU law.12 However, this ‘adding flesh to the bones’ has not been done in a vacuum, since the general principles of law (including the fundamental rights) in EU law find their roots in the national law of the Member States and international law, more particularly the ECHR. As will be discussed below, this varying background and (legal) basis for general principles and fundamental rights may give rise to certain problems in terms of claims of authority (hierarchy). This last point is particularly of concern when one bears in mind that the two main sources of the general principles of law, ie national (constitutional) law and international law, also reflect the two main claims of authority that EU law makes.

C.  Normative Coherence An important façade of coherence, for this chapter, is the normative coherence— more precisely the applicative process of general principles. After the entry into force of the Treaty of Lisbon the main question evolves around the ‘the constitutional interplay between unwritten and written fundamental rights within the Union legal order [and] [w]hether the Charter will eventually displace the general principles as the Court’s primary reference point; or whether the Court will continue to prefer to rely on its own case law’.13 The normative background to the substantive general principles of EU law is the key, but, as demonstrated below, a field where coherence clearly needs to be ‘injected’. In Opinion 2/13 the Court stated that EU law has ‘given rise to a structured network of principles’.14 It is true that EU law has gained its force largely through the judicial development of general principles. Furthermore, as held in Opinion 2/13, fundamental rights are ‘at the heart of that legal structure’.15 However, it may be questioned how ‘structured’ this network of principles really is. Clearly, Article 6 TEU is the core legal basis for the current fundamental rights jurisdiction of the EU. The assertion through Article 6(1) TEU that the EU Charter of Fundamental Rights lies at the same hierarchical level ‘as the Treaties’ arguably has had its effects. For example, shortly after its entry into force, the Court of Justice annulled EU secondary law for the first time with a reference to fundamental rights found in the EU Charter of Fundamental Rights (Articles 7 and 8).16 12  See Opinion of Advocate General Mazák in Case C-411/05, Félix Palacios de la Villa EU:C:2007:106, paras 85–86. According to the Advocate General, general principles of law are to be sought rather in the Platonic heaven of law than in the law books. While this might be true, it is also true that those principles do not fall down from heaven—Platonic or otherwise—either. The principle of non-­ discrimination, as it is developed in EU law, is a fine illustration. 13  See Editorial comments, ‘The Scope of Application of General Principles of Union Law: An ever expending Union’ (2010) 47 CMLRev 1589, 1595. 14  Opinion 2/13 of the Court (Full Court) EU:C:2014:2454, para 167. 15  ibid, para 169. 16  Joined Cases C-92/09 and C-93/09 Volker und Markus Schecke and Eifert, EU:C:2010:662.

General Principles and the Many Faces of Coherence

 81

Even the above reference to ‘fundamental rights’ being ‘at the heart’ of EU’s legal structure is limited to fundamental rights ‘recognised by the Charter’. In this light, it is relevant to reflect upon what is or should be the real effect of Article 6 TEU— to lay down a real hierarchical line of fundamental rights in the EU, based on their origins, or simply describing these different layers of fundamental rights (general principles) in EU law? What will be the status of the general principles of EU law that were, at least until very recently, the core norms of human rights protection in the EU. Are they, as they stem from Article 6(3) TEU, only to be ancillary to the EU Charter of Fundamental Rights or the true guiding norm (Leitnormen) of the EU legal order? Additionally, and significantly, in the wake of Opinion 2/13, the role of the ECHR as a source of fundamental human rights protection in the EU has been put into question and the Court of Justice’s relationship with the ECHR seems currently to be complicated. While part of the purpose of the EU Charter was to bring clarity and coherence into the chaos, that may currently not be the case. An apparent tension between ‘new’ and ‘old’ rights comes across in the post-Charter case law, and the level of coherence is tested. Below, a series of recent post-Charter cases will be discussed in order to demonstrate the challenges facing the applicative process of fundamental rights in EU law. As mentioned above, the Court of Justice annulled EU secondary legislation with a reference to the EU Charter of Fundamental Rights shortly after its entry into force. In Test-Achats it is evident from the Court’s reasoning that it sees the EU Charter as the natural starting point of the Court’s inquiry, even if the facts of the case were prior to the entry into force of the Charter.17 Since the Charter rights are at the ‘at the heart’ of EU’s legal structure, it makes perfect sense that the EU Charter is to ‘constitute the reference legislation each time the Court is called upon to rule on the compliance with an EU measure or a national provision with the fundamental rights protected by the Charter’.18 But may we go so far as to argue that the EU Charter should be brought into the proceedings ex officio by the Court of Justice? In a number of post-­Charter cases, the Court of Justice has clearly done so, particularly when it comes to Article 47 of the Charter (see further discussion below).19 However, the nature

17  Case C-236/09 Association Belge des Consommateurs Test-Achats ASBL and Others, EU:C:2011:100, paras 16 and 32. One should highlight here that the question put by the national court was formulated in light of Art 6(2) EU. As put by the Court, ‘Article 6(2) EU, to which the national court refers in its questions and which is mentioned in recital 1 to Directive 2004/113, provides that the European Union is to respect fundamental rights as guaranteed by the European Convention for the Protection of Human Rights and Fundamental Freedoms and as they result from the constitutional traditions common to the Member States, as general principles of Community law. Those fundamental rights are incorporated in the Charter, which, with effect from 1 December 2009, has the same legal status as the Treaties.’ 18  Case C-108/10 Ivana Skattolon, EU:C:2011:211, Advocate General Bot’s opinion, para 108. 19 Case C-373/09 Peñarroja Fa, EU:C:2009:637; Case C-279/09 DEB, EU:C:2010:811; and Case C-300/11 ZZ, EU:C:2013:363.

82 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

of proceedings differs, and the Court’s hands may be bound by the nature of the proceedings or for jurisdictional reasons.20 However, this shaping of the EU Charter as Leitnormen in the Court’s proceedings is still in its early stages, and it is unclear if the Court will be equally as willing to apply ex officio ‘substantive’ fundamental rights as compared to ‘structural’ remedies.21 At the same time, there is a line of fundamentally important cases in which the Court has decided not bring the EU Charter into the picture, where that seems to have been appropriate. Römer,22 Ruiz-Zambrano,23 Dominguez24 and Fag og Arbejde (FOA)25 serve as good ­illustrations of this situation. Furthermore, since the EU Charter is only applicable ‘within the scope of EU law’, there are, additionally, a number of recent cases where the scope of EU law seems exceptionally narrowly construed—in the presence of the EU Charter.26 Above it is in fact submitted that the EU Charter entered into EU’s primary law ‘with the unaltered presence of too many competitors’, eg the general principles of EU law and the ECHR—in other words, it entered a ‘crowded house’.27 But what has happened post the entry into force of the EU Charter? What type of ‘evolution’ is taking place? Survival of the fittest, the strongest or the one closest to ‘the heart’? Yet another line of recent cases seems to indicate that the norms of the EU Charter will gradually extinguish all the competition. In Chalkor the Court highlighted that ‘Article 47 of the Charter implements in European Union law the protection afforded by Article 6(1) of the ECHR. It is necessary, therefore, to refer only to Article 47.’28 This view was subsequently repeated, and slightly sharpened, in the preliminary ruling in Otis, where the Court stated that ‘Article 47 of the Charter secures in EU law the protection afforded by ­Article 6(1) of the ECHR. It is necessary, therefore, to refer only to Article 47.’29 In Inuit II a similar message was given, but here not concerning Article 47 of the EU Charter, but rather Articles 7, 10, 11 and 17 of the EU Charter, which ‘secure in

20  See, for example, Case C‑398/13 P Inuit Tapiriit Kanatami and Others v Commission (Inuit II), EU:C:2015:535, para 57. 21  The Court has held that it is to raise at its own motion issues of public policy, but not ‘pleas in law’ when undertaking legality review in inter partes proceedings, see Case C-389/10 P KME v Commission, EU:C:2011:816, para 131 and Case C‑386/10 P Chalkor v Commission, EU:C:2011:815, para 64. 22  Case C-147/08 Römer, EU:C:2011:286. 23  Case C-34/09 Ruiz Zambrano, EU:C:2011:124. 24  Case C‑282/10 Dominguez, EU:C:2012:33. 25  Case C-354/13 Fag og Arbejde (FOA), EU:C:2014:2463. 26 See eg Cases C-483/12 Pelkmans, EU:C:2014:304; C-333/13 Dano, EU:C:2014:2358; and C-67/14 Alimanovic, EU:C:2015:597 See more on this ‘hiding effect’ of the EU Charter in Groussot and ­Petursson (n 9). 27  P Cruz-Villalón, ‘Rights in Europe: The Crowded House’, King’s College Working Paper 01/2012, 14. See also and more generally EA Posner, The Twillight of Human Rights Law (Oxford, Oxford University Press, 2014). 28  Case C‑386/10 P Chalkor v Commission, EU:C:2011:815, para 51, emphasis added. 29  Case C-199/11 Otis and Others, EU:C:2012:684, para 47, emphasis added.

General Principles and the Many Faces of Coherence

 83

EU law the protection conferred by the provisions of the ECHR relied on by the appellants and that it is appropriate, in this instance, to base the examination of the validity of the basic regulation solely on the fundamental rights guaranteed by the Charter’.30 In the very recent JN case, of 15 February 2016, concerning the validity of an EU secondary norm, the claim of the Charter’s superiority seems unqualified, since ‘an examination of the validity of point (e) of the first subparagraph of Article 8(3) of Directive 2013/33 must be undertaken solely in the light of the fundamental rights guaranteed by the Charter’.31 Additionally, in JN there is no reference to the case law of the European Court of Human Rights (ECtHR). That seems also to be a growing trend, perhaps to be seen in the light of the now frequent reference by the Court of Justice to the fact that the ECHR ‘does not constitute, as long as the European Union has not acceded to it, a legal instrument which has been formally incorporated into EU law’, and even to the outcome in Opinion 2/13.32 Irrespective of this, neither the ECHR nor the general principles seem to have fully left the ‘crowded house’ yet. In the Deutsche Bahn case the Court breathed life into both the ECHR and the general principles of EU law when stating that: ‘It must be remembered in that regard that that fundamental right constitutes a general principle of EU law, as currently expressed in Article 47 of the Charter, which is the equivalent under EU law of Article 6(1) of the ECHR.’33 Furthermore, the general principles of EU law are still present in the case law of the Court of Justice. In Kokopelli the Court could have mentioned the relevant provisions of the Charter in relation to the principle of non-discrimination and the freedom to pursue an economic activity but preferred to rely instead on the case law on general principles of EU law.34 Similarly in Global Trans Lodzhistik the Court of Justice made a reference to the rights of the defence, as a general principle of EU law, in the context of the right to be heard, but made no reference to the EU Charter of Fundamental Rights.35

30  Case C‑398/13 P Inuit Tapiriit Kanatami and Others v Commission, EU:C:2015:535, para 46, emphasis added 31  Case C-601/15 PPU JN v Staatssecretaris voor Veiligheid en Justitie, EU:C:2016:84, 32  See Case C‑617/10 Åkerberg Fransson, EU:C:2013:105, para 44; Case C‑501/11 P Schindler ­Holding and Others v Commission, EU:C:2013:522, para 32; and Case C‑295/12 P Telefónica and ­Telefónica de España v Commission, EU:C:2014:2062, paras 41, 45; JN v Staatssecretaris voor Veiligheid en J­ustitie, para 45, n # and Inuit II, para 45 of CN. Compare also Case C-293/12 and 594/12 Digital Rights ­Ireland, EU:C:2014:238, with a substantial amount of reference to ECHR case law, with Case C-362/14 ­Maximillian v Facebook, EU:2015:650, with a reference to Digital Rights Ireland, but none to ECHR case law. In those type of situations the’ dilution’ of ECHR case law happens’ naturally’. 33  Case C-583/13 P Deutsche Bahn AG et al, EU:C:2015:404, para 47, emphasis added. 34  Case C-59/11 Association Kokopelli, EU:C:2012:447. 35  Joined Cases C-29/13 and 30/13 Global Trans Lodzhistik OOD, EU:C:2014:140, para 51. See also Case C-206/13 Siragusa, EU:C:2014:126, para 34, where the Court is making a reference to the principle of proportionality, one of the general principles of EU law ‘which must be observed by any national legislation which falls within the scope of EU law or which implements that law’.

84 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

D.  Transnational Coherence Transnational coherence is another face of coherence. Substantive general ­principles—by creating a bottom-up effect—are perfect tools for building a ­transnational legal culture or a common constitutional space. Indeed, the true function of the general principles must be assessed in the light of a mutual ­cross-fertilization which creates a continuous flux of ideas and exchange of opinions between the Court of Justice and its national counterparts. This gives rise to a ‘common constitutional space’ defined by a dynamic dialogue.36

The general principles lie at the core of the inter-legality of EU law (which is also premised on this inter-legality). This interpretation is also confirmed by the recent decision of the CJEU in Åkerberg:37 29  That said, where a court of a Member State is called upon to review whether fundamental rights are complied with by a national provision or measure which, in a situation where action of the Member States is not entirely determined by European Union law, implements the latter for the purposes of Article 51(1) of the Charter, national authorities and courts remain free to apply national standards of protection of fundamental rights, provided that the level of protection provided for by the Charter, as interpreted by the Court, and the primacy, unity and effectiveness of European Union law are not thereby compromised (see, in relation to the latter aspect, Case C-399/11 Melloni [2013] ECR, paragraph 60).

Another aspect of transnational coherence and perhaps the most important is the spillover of substantive general principles into national law, eg legitimate expectations in France or proportionality in Sweden.38 A recent example of the building of this transnational law is given by the impact of the principle of good administration in Irish law. It should be emphasised that regardless of whether the EU general principles of an administrative law nature flow directly from Article 41 of the EU Charter or from the case law of the CJEU, these principles are an important complement to the substantive EU rules. These principles of administrative law add structural or procedural guarantees to the application of different substantive EU law principles, such as equal treatment, proportionality or mutual recognition,39 depending on the circumstances of 36  See K Lenaerts and J Gutiérrez-Fons, ‘The Constitutional Allocation of Powers and General Principles of EU Law’ (2010) 47 CMLRev 1629, 1630. 37  Case C-617/10 Åkerberg, EU:C:2013:105. See D Ritleng, ‘De l'articulation des systèmes de protection des droits fondamentaux dans l'Union: les enseignements des arrêts Ákerberg Fransson et Melloni’ RTDE 2013 (April–June), 267ff. 38  The consistency argument (‘two-speed law argument’) provides another element supporting the spillover of the principles. It considers that the existence of the two standards of law (EU law and internal law) goes against the coherence of the legal system and that, consequently, only one standard (the higher one) should be applied. This voluntary choice to adapt the judicial system should be made in order to ensure the coherence of the legal system. 39  cf Regulation (EC) No 764/2008 of the European Parliament and of the Council of 9 July 2008 laying down procedures relating to the application of certain national technical rules to products lawfully marketed in another Member State and repealing Decision No 3052/95/EC, [2008] OJ L218/21.

General Principles and the Many Faces of Coherence

 85

each case.40 This legal development, which has long led to a strengthening of the four freedoms and the rules on citizenship, is now proving to be of primary importance for the development of the protection for asylum seekers. In this context, it is interesting to note that Ireland—as well as the UK—has chosen to opt in to the first phase of EU asylum measures adopted between 1999 and 2004 but did not participate in the second phase, resulting in the Recast Procedural Directive 2013/32/EC.41 Still, the key rules of the secondary phase legislation apply to Ireland as a result of the CJEU’s interpretation of the first-phase rules.42 This is evidenced by the reforms of the procedural rules applicable to the refugee and subsidiary claims that followed the ruling in N.43 On 24 March 2015 the Irish government approved the general scheme of the International Protection Bill, which has as its main purpose the establishment of a single procedure for the examination of applications for refugee and subsidiary protection status in Ireland.44 The Irish Minister for Justice and Equality offered the summary that: ‘This Bill will achieve the desired balance in treating asylum seekers with humanity and respect while ensuring more efficient immigration procedures and safeguards.’45 The immediate actions of the Irish government and subsequent relevant legislative changes in Irish procedural law are clear indications that N and other cases in this area are taking a strong stand in the creation of European procedural jus commune. It should be emphasised that EU law, including its fundamental rights, are to be applied coherently at the national level through national courts. There it may not be so easy to grant the EU Charter, as interpreted by the Court of Justice, the highest normative status—at least not if it ‘extinguishes’ other norms. In that light, the message from the Court of Justice regarding the applicative process needs 40  J Hettne, ‘Administrative Law as a Key to Market Integration?’ in P Cardonnel, A Rosas and N Wahl (eds) Constitutionalising the EU Judicial System—Essays in Honor of Pernilla Lindh (Oxford, Hart Publishing, 2012), 165. 41  Recital 58 of the Recast Procedural Directive 2013/32/EC states that: ‘In accordance with Articles 1, 2 and Article 4a(1) of Protocol No 21 on the position of the United Kingdom and Ireland in respect of the Area of Freedom, Security and Justice, annexed to the TEU and the TFEU, and without prejudice to Article 4 of that Protocol, the United Kingdom and Ireland are not taking part in the adoption of this Directive and are not bound by it or subject to it application.’ 42  Following the ruling in the N, Ireland is bound to decide the claims for subsidiary protection within a reasonable time. 43  Case C-604/12 HN v Minister for Justice, Equality and Law Reform and Others, EU:C:2014:302. Indeed, this judgment was positively received inter alia by the Irish Refugee Council and perceived as a ‘clear mandate’ to introduce a single procedure for determining refugee and subsidiary protection. The Irish Office of the Refugee Application Commissioner issued a notice regarding the making of applications for subsidiary protection by applicants for refugee status already only some months following the verdict in N. The notice put forward immediate arrangements to comply with the judgment, including allowing applications for subsidiary protection to be submitted at the same time as applications for refugee status for new applicants and asylum seekers whose applications for refugee status were still pending. The notice was followed by legislative changes. 44  Department of Justice and Equality, General Scheme of the International Protection Bill and Regulatory Impact Analysis’ at http://justice.ie/en/JELR/Pages/PB15000080 (last visited 10 June 2015). 45  See the statement at www.inis.gov.ie/en/INIS/Pages/Wednesday%2025%20March%202015 (last visted 15 June 2015).

86 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

to be clear and coherent—but not entirely without flexibility. The substantive general principles are vehicles for building the transnational legal culture, where ideas and norms flow freely between the national and international level. Where national authorities and courts are to ‘remain free to apply national standards of protection of fundamental rights’,46 trust must also be granted in reality. The Irish government’s recent reaction to N and other cases shows that in that case trust was warranted for the acceptance of the right to good administration.

III.  Good Administration and Coherence A.  Good Administration and the Charter General principles of law have for a long time provided guidance for the ­development of EU administrative law. These principles are typically relevant when the European Commission is dealing with matters in, for example, the ­competition and state aid sector and also when power has been attributed to a ­particular EU authority, eg the European Union Intellectual Property Office.47 One reason why general principles of law have been particularly important in this context is that the EU lacks a general administrative law of the type which normally exists in the Member States. The most important principles are now enshrined in Article 41 in the EU Charter, headed ‘Right to Good Administration’, which is worded as follows: 1. Every person has the right to have his or her affairs handled impartially, fairly and within a reasonable time by the institutions, bodies, offices and agencies of the Union. This right includes: (a) the right of every person to be heard, before any individual measure which would affect him or her adversely is taken; (b) the right of every person to have access to his or her file, while respecting the ­legitimate interests of confidentiality and of professional and business secrecy; (c) the obligation of the administration to give reasons for its decisions.

In addition, Article 47 in the Charter provides: Everyone whose rights and freedoms guaranteed by the Union law are violated has the right to an effective remedy before a tribunal in compliance with the conditions laid down in this Article. Everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal previously established by law. Everyone shall have the possibility of being advised, defended and represented. 46  47 

Case C-399/11 Melloni, ECLI:EU:C:2013:107. Former Office for Harmonisation of the Internal Market (OHIM).

General Principles and the Many Faces of Coherence

 87

Legal aid shall be made available to those who lack sufficient resources in so far as such aid is necessary to ensure effective access to justice.

Finally, Article 48 in the Charter states: 1. Everyone who has been charged shall be presumed innocent until proved guilty according to law. 2. Respect for the rights of the defence of anyone who has been charged shall be guaranteed.

These Articles safeguard the protection of individuals and constitute a tool for judicial control. Articles 47 and 48 regarding the right to an effective remedy and to a fair trial correspond to similar procedural guarantees in the ECHR and have been the basis for important developments in EU administrative law.48 Importantly, however, the right to good administration does not in itself fall under the ECHR. Instead, the Council of Europe has enacted non-binding resolutions within the area and has adopted a recommendation on a model code of good administration.49 Similarly, a model code has been adopted by the EU Ombudsman, namely the European Code of Good Administrative Behaviour.50 The purpose of this code is to ensure that the EU institutions, bodies, offices and agencies respect the principle of good administration. Good administration as an enforceable right has, however, been rather ill-defined, and marked by a terse and complex jurisprudence. Thus, Hanns Peter Nehl has concluded that ‘the principle of good administration’ exclusively exists as a phrase but not as a procedural standard with a particular content and meaning.51 The principle has generally been considered as an umbrella principle covering a cluster of rights. An attempt at consolidation was made by the General Court in Case T-54/99 Max Mobil: Since the present action is directed against a measure rejecting a complaint, it must be emphasised at the outset that the diligent and impartial treatment of a complaint is associated with the right to sound administration which is one of the general principles that are observed in a State governed by the rule of law and are common to the constitutional traditions of the Member States.52

It should, however, be observed that there are no detailed requirements imposed on the Member States (in the absence of specific secondary EU law)53 to apply 48  For instance, the right to be heard is a general principle of EU law pertaining, on the one hand, to the right to good administration, laid down in Art 41 of the Charter and, on the other, to observance of the rights of defence and the right to a fair trial enshrined in Arts 47 and 48 of the Charter. See the judgments of the Court of Justice in Case C‑349/07 Sopropé, EU:C:2008:746, paras 36–38 and Case C‑27/09 P France v People’s Mojahedin Organization of Iran, EU:C:2011:853, para 66. 49  See J Reichel, ‘Between Supremacy and Autonomy—Applying the Principle of Good Administration in the Member States’ in U Bernitz, J Nergelius and C Cardner (eds), General Principles of EC Law in a Process of Development (Dordrecht, Kluwer, 2008), 255. 50  Available at: www.ombudsman.europa.eu/en/resources/code.faces#/page/1. 51  HP Nehl, Principles of Administrative Procedure in EC Law (Oxford, Hart Publishing, 1999) 37. 52  Case T-54/99 Max Mobil, EU:T:2002:20, para 48. 53  See, for instance, Art 10 in Directive 2006/123/EC of the European Parliament and of the Council of 12 December 2006 on services in the internal market, [2006] OJ L376/36.

88 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

the principle of good administration in the same way as at the European level. This differs from the requirements following from Articles 47 and 48 in the Charter which directly concern national litigation when the situation is governed by EU law and also correspond to the requirements following from the ECHR. It is submitted that this is due to the fact that the Member States generally have their own laws on good administration which apply in the situations governed by EU law, and these national laws and legal traditions deserve to be respected. However, as mentioned above, the general principles of administrative law can nonetheless be of prime importance, not only in connection with the EU administration, but also in national processes. The CJEU has many times referred to general principles of administrative law in order to ensure a correct and effective application of substantive EU law in the national legal context.54

B. Ambiguity in the Jurisprudence Relating to Article 41 of the Charter An important aspect of Article 41 of the Charter is thus its institutional scope. At least a narrow reading of this Article suggests that it does not apply to the Member States’ institutions or administrations. Consequently, in Cicala55 the CJEU observed that Article 41 of the Charter is addressed only to the bodies and institutions of the EU. However, this reading of Article 41 departs from the general scope described in Article 51 of the Charter by explicitly limiting the principle of good administration to the measures taken by ‘the institutions, bodies, offices and agencies of the Union’. Article 51(1) states that the provisions of the Charter are addressed to the Member States when they are implementing EU law, which has been interpreted by the CJEU to mean when they act within the scope of EU law.56 In N,57 which concerned minimum standards on procedures in Member States for granting and withdrawing refugee status, Advocate General Bot claimed that the Member States are required to ensure observance of fundamental rights and the general principles of Union law when they adopt decisions falling within the ambit of EU law. There was therefore no doubt, according to the Advocate General, that the Irish authorities must ensure respect for the right to good administration enjoyed by the persons concerned, not only because that right constitutes a general principle of EU law, but also because it is a fundamental right affirmed in Article 41 of the Charter. In the view of Mr Bot, although the wording of Article 41(1) of the Charter refers to relations between individuals and the ‘institutions, bodies and agencies of the Union’, the right to good administration should be

54  See, for instance, Case C-157/99 Smits and Peerbooms, EU:C:2001:404, para 90; Case C-385/99 Müller-Fauré and van Riet, EU:C:2003:270, para 85; Case C-372/04 Watts, EU:C:2006:325, para 116. 55  Case C-482/10 Teresa Cicala v Regione Siciliana, EU:C:2011:868, para 28. 56  See Case C-617/10 Åkerberg, EU:C:2013:105, paras 17–23. 57  Case C-604/12 HN v Minister for Justice, Equality and Law Reform and Others, EU:C:2014:302.

General Principles and the Many Faces of Coherence

 89

incumbent in the same way on the Member States when they are implementing EU law.58 The CJEU (Fourth Chamber) responded that Article 41 of the Charter ‘reflects a general principle of EU law’ (paragraph 49). Therefore, where a Member State implements EU law, ‘the requirements pertaining to the right to good administration’, including the right of any person to have his or her affairs handled impartially and within a reasonable period of time, are applicable in a procedure for granting subsidiary protection. The Court held that Irish procedural law was compatible with these requirements. Accordingly, in N the Court seemed to suggest that Article 41 of the Charter applies to the bodies and institutions of the Member States when they act within the scope of EU Law. This approach is non-hierarchical, meaning that it makes no distinction between Charter rights and the general principles of EU law.59 In a similar vein, this understanding of the scope of Article 41 is in harmony with the interpretation of Article 51 of the Charter by the Court in Åkerberg.60 The approach in N seems to be confirmed in the later case, M,61 where the court (First Chamber) applied Article 41 to the actions of the Member States, in addition to arguing that the right to good administration is of general application. However, these rulings beg the question if the wording of Article 41 can be totally ignored. Soon after delivering the judgments in N and M, the CJEU interestingly seemed to change its approach with regard to the relationship between Charter rights and general principles. In the ruling YS and others62 the Court made a leap over its verdicts in N and M and went back to the Cicala approach. More precisely, in YS and others both the court (Third Chamber) and Advocate General Sharpston found that the right to good administration, as enshrined in Article 41, is addressed only to the institutions and bodies of the EU. This view was confirmed more recently in Khaled Boudjlida. In that case, ­Advocate General Wathelet stated that it is true that Article 41 of the Charter establishes the right to be heard only in respect of the institutions and bodies of the Union. According to the Advocate General, it seemed, however, neither consistent nor in accordance with the previous case N for the wording of Article 41 of the Charter to allow the introduction of an exception to the rule laid down in Article 51 thereof enabling the Member States not to apply an Article of the Charter, even when they are implementing Union law. He therefore argued in favour of the applicability of Article 41 of the Charter to the Member States when they are implementing Union law, but added that, in any event, the right to be heard, according to settled case law, is a general principle of EU law. Accordingly, the AG concluded that ‘the right to be heard (if not Article 41 in 58 

EU:C:2013:714, para 36. HCH Hofmann and BC Mihaescu, ‘Relation between Charter’s Fundamental Right and Unwritten General Principles’ (2013) 9 European Constitutional Law Review, 73. 60  Case C-617/10 Åkerberg, EU:C:2013:105. 61  Case C-277/11, EU:C:2012:744, para 84. 62  Joined Cases C‑141/12 YS v Minister voor Immigratie, Integratie en Asiel and C‑372/12 Minister voor Immigratie, Integratie en Asiel v M, S, EU:C:2014:2081, paras 66–69. 59 See

90 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

itself) must be observed by the national authorities when they adopt d ­ ecisions falling within the scope of EU law’.63 Thus, the AG suggested two alternative approaches. Either the principle of good administration is applicable in the Member States because this follows directly from Article 41 of the EU C ­ harter (cf Article 6.1 TEU) or certain principles covered by this umbrella principle apply because they constitute general principles of EU law (cf Article 6.3 TEU). The Court (First Chamber) opted in Khaled Boudjlida for the second approach, which is worth quoting: 32 As the Court stated in paragraph 67 of the judgment in YS and Others (C-141/12 and C-372/12, EU:C:2014:2081), it is clear from the wording of Article 41 of the Charter that it is addressed not to the Member States but solely to the institutions, bodies, offices and agencies of the European Union (see, to that effect, the judgment in Cicala, C-482/10, EU:C:2011:868, paragraph 28). 33 Consequently, an applicant for a resident permit cannot derive from Article 41(2)(a) of the Charter a right to be heard in all proceedings relating to his application (the judgment in Mukarubega, EU:C:2014:2336, paragraph 44). 34 Such a right is however inherent in respect for the rights of the defence, which is a general principle of EU law (the judgment in Mukarubega, EU:C:2014:2336, paragraph 45).64

In substance the Court’s solution does not differ from the first option suggested by the Advocate General, but the Court did clearly not follow the recommendation to disregard the wording of Article 41 of the EU Charter. This interpretation has more recently been confirmed in Bensada Benallal.65 These later cases also imply that general principles of EU administrative law have an independent meaning outside the scope of Article 41 of the Charter, at least when applied in the Member States. However, it now falls to be considered whether this inconsistency between Article 41 and Article 51 in the Charter can be explained and rationalised.

C. Jurisprudential Coherence, National Procedural Autonomy and Inter-legality The requirement of coherence is based on respect for the principle of legal ­certainty, but, as observed by Nic Shuibhne ‘certainty’ is not the most helpful term to apply in the context of case law.66 Coherence is not about striving for

63  Opinion in Case C-249/13 Khaled Boudjlida v Préfet des Pyrénées-Atlantiques, EU:C:2014:2032, paras 46 and 47. 64 EU:C:2014:2431. 65 Case C-161/15 Bensada Benallal, EU:C:2016:175 in comparison with the opinion from AG ­Mengozzi, EU:C:2016:3, para 32. 66  See Nic Shuibhne (n 1) 10.

General Principles and the Many Faces of Coherence

 91

an unrealistic degree of perfection or rigidity. It allows for the recognition and management of necessary differences. What coherence does demand is that ­differences must be explained and rationalised and that requirement is connected to the manifestation of fairness and integrity.67 As mentioned in the introduction of this chapter, it is important to keep in mind that the case law of the CJEU must be applied in all courts and tribunals across twenty-eight Member States. This means that the Court must articulate workable principles for widespread subsequent application. It is submitted, and will be explained below, that the apparent inconsistency between Article 41 and 51 in the Charter can actually be explained and rationalised. It has been pointed out above that the right to good administration is more strongly associated with national procedural autonomy and the respect for different legal cultures than other rights and principles in the EU Charter. Moreover, unlike Articles 47 and 48 of the Charter, Article 41 is not based on a common legal heritage arising from the ECHR. The Union also lacks legislative competence to implement, at a general level, the principle of good administration in the Member States. A reference should be made to Article 298 TFEU, according to which: 1. In carrying out their missions, the institutions, bodies, offices and agencies of the Union shall have the support of an open, efficient and independent European administration. 2. In compliance with the Staff Regulations and the Conditions of Employment adopted on the basis of Article 336, the European Parliament and the Council, acting by means of regulations in accordance with the ordinary legislative procedure, shall establish provisions to that end.’

This Article provides a new legal basis introduced by the Lisbon Treaty. It is unclear whether Article 298 TFEU can constitute the basis for a comprehensive framework on good administration in the Union, that is to say an Administrative Act which is binding for the institutions, bodies, offices and agencies of the Union. However, it is clear that this legal basis will not suffice to harmonise national law on good administration and cannot accordingly be the basis for binding administrative rules on national authorities in situations governed by EU law. National administrative law is part of the national procedural autonomy which forms, in the absence of specific secondary EU rules, the procedural framework also in situations governed by EU law in the Member States. This legal situation was clearly observed by the CJEU in N (paragraph 41), where the Court enunciated: In the absence of EU rules concerning the procedural requirements attaching to the examination of an application for subsidiary protection, the Member States remain competent, in accordance with the principle of procedural autonomy, to determine those requirements, while at the same time ensuring that fundamental rights are observed and that EU provisions on subsidiary protection are fully effective.

67 ibid.

92 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

However, as also indicated by the Court in N, EU law sets certain requirements. According to settled case law, the principle of equivalence requires that the national rule at issue be applied without distinction, whether the action is based on rights which individuals derive from EU law or whether it is based on an infringement of national law, where the purpose and cause of action are similar.68 Respect for that principle requires equal treatment of claims based on a breach of national law and similar claims based on a breach of EU law.69 Moreover, as underlined by the CJEU in N, the Member States’ authorities must also ensure that fundamental rights are observed. Against this background, it is understandable that Article 41 of the EU Charter is addressed to the institutions, bodies, offices and agencies of the Union. This Article can be seen in the same context as the aforementioned Article 298 TFEU and in the light of the principle of national procedural autonomy. Thus, Article 41 should not be binding on national authorities since they are already bound by similar and usually more detailed national rules on good administration. This can also serve to explain why the authors of the Charter introduced a distinction between the scope of the right to good administration and the general scope of the Charter (Article 51). If the Union has no competence to impose administrative requirements on national authorities, it is reasonable that the Charter contains a similar limitation, as the Charter should not change the scope of application of fundamental rights protection, and should respect the constitutional allocation of powers sought by the authors of the Treaties (cf Article 6.1 TEU).70 Nevertheless, in situations governed by EU law in the Member States, the same general principles of EU law must anyway be respected. However, in these situations there is more room for a balanced application of national administrative law and EU principles of administrative law. This situation can be described by the concept of inter-legality which was mentioned above. It means that ‘different legal spaces [are] superimposed, interpenetrated and mixed in our minds, as much as in our actions’.71 As underlined by Tuori, inter-legality is particularly important for the functioning of transnational EU law.72 In the mirror of inter-legality, legal orders 68 

Case C-93/12 Agrokonsulting-04, EU:C:2013:432, para 39. Case C-69/14 Târșia, EU:C:2015:662, para 34. 70  In the legal literature this question has been discussed long before the Charter became binding. Kanska and Lord Millet have both argued that Art 41 of the Charter cannot be invoked against national authorities implementing EU law: see K Kanska, ‘Towards Administrative Human Rights in the EU— Impact of the Charter of Fundamental Rights’ (2004) 10 European Law Journal 309 and Lord Millet, ‘The Right to Good Administration in European Law’ [2002] Public Law 317. On the other hand, N Fenger, Forvaltning & Faellesskabs. Om EU-rettens betydning for den almindelige forvaltningsret (Copenhagen, Jurist- og Ökonomförbundets Forlag 2004) 258 and 350, and T Fortsakis, ‘Principles Governing Good Administration’ [2005] European Public Law, 216, presuppose that Art 41 also binds the Member States. See for a later discussion J Reichel, ‘Between Supremacy and Autonomy—Applying the Principle of Good Administration in the Member States’ in U Bernitz, J Nergelius and C Cardner (eds), General Principles of EC Law in a Process of Development (Dordrecht, Kluwer, 2008) 255. 71  See B de Sousa Santos, Toward a New Legal Common Sense, Law, Globalization and Emancipation, 2nd edn (London, Butterworths, 2002) 347. 72  Tuori (n 5) 80. 69 

General Principles and the Many Faces of Coherence

 93

do not appear self-contained entities: they are seen as mutually overlapping and maintaining a dialogical relation with each other.73 Thus, in the EU, inter-­legality allows the policy-based objectives inherent in EU law to impact on national law without neglecting the significance of national legal cultures. Inter-legality is not a question of mechanical application of existing norms but, to some extent, the production of new law.74 The CJEU seeks in reality to achieve the interweaving by accepting legal cultural differences.75 The national procedural autonomy must be respected, but national law should meet the equivalence and effectiveness requirements of the EU law, which could lead to a trade-off between efficiency requirements and other important interests worthy of protection. National law may therefore have to be ‘adjusted’ if it prevents the effective application of EU law, but this is not the same thing as applying Article 41 in the Charter instead of national law. Against this background, one should also consider more carefully the actual wording used by the CJEU in N and M. In paragraph 49 in N, the Court stated that the right to good administration stipulated in Article 41 of the EU Charter ‘reflects’ a general principle of EU law. This is not the same thing as saying that Article 41 is directly applicable in the Member States. Furthermore, the Court’s statement in M (paragraph 84) that Article 41(2) of the Charter is of general application does not necessarily mean that it is, as such, applicable in the Member States. As pointed out by Advocate General Sharpston in YS and others,76 paragraphs 82–84 of the judgment in M, when read together, suggest rather that the Court was focusing on the substance of the right to be heard and who can invoke it; and in so doing the Court emphasised both the very broad scope of that right and the place that it has long held in the EU legal order. However, it is respectfully submitted that the Court, if this was the intended meaning, should have pointed out much more clearly that Article 41 is not, in itself, applicable in the Member States and should have explained why.77

IV.  Ideological Coherence It is argued in this contribution that Opinion 2/13 reflects ‘ideological coherence’. Bengoetxea refers to ideological coherence as being characterised by the values of 73 ibid.

74  See M Amstutz, ‘In-Between Worlds: Marleasing and the Emergence of Interlegality in Legal ­Reasoning’ (2005) 11 European Law Journal 769. 75  ibid, 775. 76  Joined Cases C-141/12 YS v Minister voor Immigratie, Integratie en Asiel and C-372/12 Minister voor Immigratie, Integratie en Asiel v M, S, EU:C:2013:838, para 90. 77  As stated by Nic Shuibhne, uneven articulations of clear rules because of sporadic opportunities for refinement through case law channels is one thing; persistently shifting or unsettled principles expressed in variable ways and departing from the case law mainframe without rationalised explanations, is something else. See Nic Shuibhne (n 1) 10.

94 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

the body politic as stated in and interpreted by the constitution.78 Defining the relationship between law and ideology is not an easy task; they are in fact intimately interlinked. In that sense, if law is a system of enforceable rules governing social relations and legislated by a political system, it might seem obvious that law is connected to ideology, which refers, in a general way, to a system of political ideas.79 The complexity is also thickened by the multitude of linguistic terminologies used to express ‘ideology’ in the legal field.80 In his recent book on the relationship between the academy and the judiciary, Richard Posner mentions the ‘prior’ which is defined as a belief or inclination, conscious or (frequently) unconscious, that one (a judge) brings to an issue.81 Is there any difference between a prior and an ideology?82 Opinion 2/13 constitutes a ‘constitutional speech act’ towards international law.83 In other words, the Court of Justice claims authority over the law of the ECHR. This claim of authority is visible in many ways but its most interesting and surprising aspect takes the form of mutual trust as a ‘justificationary principle’.84 The CJEU used in the past different types of principles to justify claims of authority over national law and international law. For instance, unity/effectiveness (Costa),85 loyalty (Pupino)86 or rule of law (Kadi)87 were relied on in order to justify, respectively, the primacy of EU law, the application of general principles within the third pillar in the national legal orders or the priority of EU human rights law over the law of the United Nations. The Court in Opinion 2/13 is particularly concerned by a (ideological) clash of jurisprudence between the two European courts.88 This clash is illustrated by

78  J Bengoetxea, ‘Rethinking EU Law in the Light of Pluralism and Practical Reason’ in M Maduro, K Tuori and S Sankari (eds), Transnational Law: Rethinking European Law and Legal Thinking (Cambridge, Cambridge University Press, 2014) 145, 147. 79 See Stanford Encyclopedia of Philosophy, ‘Law and Ideology’, first published 22 October 2001; substantive revision 24 October 2014, http://plato.stanford.edu/entries/law-ideology/. 80  JB Thompson, Studies in the Theory of Ideology (Berkeley, University of California Press, 1984) 36. The study of ideology is fundamentally concerned with language. Ideology is present in any political programme, system of thoughts and system of belief. 81  RA Posner, Divergent Paths: The Academy and the Judiciary (Cambridge, MA, Harvard University Press, 2016) 2 and 17. 82  DM Kahan, ‘Ideology in or Cultural Cognition of Judging: What Difference Does it Make?’ (2009) 92 Marquette Law Review 413. 83  See Tuori (n 5). In this book, Tuori relies often on the concept of ‘constitutional speech act’. See also J Habermas, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy (Cambridge, MA, MIT Press 1997). For Habermas, an ideal speech situation is connected to rationality. There is an internal connection between the validity claims raised by speech acts and the notion of reason or rationality. Validity claim is justified only if the statement would command the consent of anyone who could enter into the discussion with the speaker. 84  Tuori (n 5). 85  Case 6/64 Costa, ECLI:EU:C:1964:66. 86  Case C-105/03 Pupino, ECLI:EU:C:2005:386. 87  Case C-402/05 P Kadi, ECLI:EU:C:2008:461. 88  G Butler, ‘A Political Decision Disguised as Legal Argument? Opinion 2/13 and European Union Accession to the European Convention of Human Rights’ [2015] Utrecht Journal of International and European Law 104.

General Principles and the Many Faces of Coherence

 95

the tension between the protection of human rights and the effectiveness of EU law. The general tone in the Opinion is very defensive and based on a rhetoric of autonomy, exclusive jurisdiction and effectiveness reflecting the life-world of the CJEU. The Melloni case is at the heart of the reasoning of the Court, which also goes very deep into the specific nature of EU law.89 It is underlined that the EU constitutional framework is based on a particularly sophisticated institutional structure and a full set of legal rules to ensure its operation, which has consequences on the procedure for and conditions of accession to the ECHR.90 As stated by the Court: These essential characteristics of EU law have given rise to a structured network of principles, rules and mutually interdependent legal relations linking the EU and its Member States, and its Member States with each other. … This legal structure is based on the fundamental premiss that each Member State shares with all the other Member States, and recognises that they share with it. …That premiss implies and justifies the existence of mutual trust between the Member States that those values will be recognised and, therefore, that the law of the EU that implements them will be respected.91

At the hub of this legal structure are the fundamental rights recognised by the EU Charter of Fundamental Rights. But the CJEU quickly refers to the need to respect the EU objectives which are closely connected to ensuring the effectiveness of the internal market and the process of integration.92 The principle of mutual trust is a key element of Opinion 2/13.93 The Court sees a tension between this fundamental principle of EU law and the Draft Accession Agreement.94 As it puts it: [T]he approach adopted in the agreement envisaged, which is to treat the EU as a State and to give it a role identical in every respect to that of any other contracting party, specifically disregards the intrinsic nature of the EU and, in particular, fails to take into consideration the fact that the Member States have, by reason of their membership of the EU, accepted that relations between them as regards the matters covered by the transfer of powers from the Member States to the EU are governed by EU law.

This obligation of mutual trust is jeopardised by accession, which is liable to upset the balance of the EU and undermine the autonomy of EU law.95 Mutual trust is particularly present in the Area of Freedom Security and Justice, where the Member States are required, except in exceptional circumstances, to comply with EU

89  See Opinion 2/13, n, paras 153–71. Another example is given by the case law of the ECJ in ­ kerberg Fransson and Melloni, nn 37 and 46, respectively, where effectiveness is used as a policy yardÅ stick together with primacy and unity in order to assess the potential application of national standards of fundamental rights (see especially para 63 in Melloni). 90  Opinion 2/13, para 158. 91  ibid, paras 167–68. 92  ibid, paras 169–72. 93  ibid, para 167. 94  ibid, para 191. 95  ibid, paras 194.

96 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

law and particularly with the fundamental rights recognised by EU law.96 This obligation based on automaticity may also conflict with the case-by-case analysis used in ECtHR cases to determine a breach of human rights.97 Is mutual trust the raison d’être of EU law?98 Is it so after Opinion 2/13? What is the true nature of mutual trust? Opinion 2/13 is built on references to mutual trust. In principle, mutual trust requires each Member State to consider all the other Member States to be complying with EU law and particularly with the fundamental rights recognised by EU law. For the Court of Justice even though EU law imposes an obligation of mutual trust between those Member States, accession is liable to upset the underlying balance of the EU and undermine the autonomy of EU law. The logic of mutual trust (based on automaticity and acceptance of the—possibly lower—standard of protection) is here clearly used as an argument against the accession to the ECHR and is tied to the notion of effectiveness. The concept of mutual trust, as used in Opinion 2/13, appears to go beyond its usual sense, reaching a kind of constitutional status that also reflects its transformative nature.99 Mutual trust is here described, not as the raison d’être of EU law (as it has been in the past in NS),100 but as a fundamental principle of EU law. This change of language is certainly much better. How can mutual trust be the raison d’être of EU law? It is now recognised as a fundamental principle just like loyalty and effectiveness. These fundamental principles are also structural to the EU legal order and are closely interconnected. As seen before, these principles have been used by the CJEU to justify various claims of authority. What is new with Opinion 2/13 is that mutual trust is relied on for the very first time as a partisan motivation to justify a claim of authority of EU law towards international law, ie the law of the ECHR. The core of Opinion 2/13 is about mutual trust and reflects the ideological coherence of the CJEU—its true nature. But to understand the full meaning of mutual trust in the post-Lisbon era, it is important in our view to look back at its juridical origins: the Cassis ruling.101 A parallel, in that regard, must be drawn

96 See, to that effect, judgments in Joined Cases C‑411/10 and C‑493/10 NS and Others, EU:C:2011:865 and Melloni, EU:C:2013:107. See also V Mitsilegas, ‘The Limits of Mutual Trust in Europe’s Area of Freedom, Security and Justice: From Automatic Inter-State Cooperation to the Slow Emergence of the Individual’ (2012) 31 Yearbook of European Law 319. 97  See eg App No 14737/09 Sneersone and Kampanella [12 July 2011] and App no 29217/12 ­Tarakhel v Switzerland [4 November 2014]. 98  Joined Cases C‑411/10 and C‑493/10 NS and Others, n 97, para 83. 99 See K Lenaerts, ‘The Principle of Mutual Recognition in the Area of Freedom, Security and ­Justice’, The Fourth Annual Sir Jeremy Lever Lecture, All Souls College, University of Oxford, 30 January 2015. 100  Joined Cases C‑411/10 and C‑493/10 NS and Others, n 97, para 83. 101  SK Schmidt, ‘Mutual Recognition as a New Mode of Governance’ (2007) 14 Journal of European Public Policy, 667, 673. According to the author, ‘in the Community different legal positions match the different principles of integration. Under the Cassis de Dijon case law, the ECJ gives a broad meaning to the market freedoms (for goods, services, persons, and capital). As long as goods or services are legally marketed in a member state, in other words, if they conform to the regulations of their home state, they can also be marketed in all other member states. This is the obligation to mutually recognize

General Principles and the Many Faces of Coherence

 97

between Cassis and Melloni (the heart of Opinion 2/13). As a preliminary remark, it is worth stressing the interest of Cassis for political scientists.102 This case— undeniably related to the issue of ‘power’—has become a cult in the EU scholarship interested in the role of the CJEU in making law.103 Melloni is arguably of the same kind and is founded on the logic of mutual recognition or mutual trust to claim authority of EU law over national laws in the human rights context. Do we observe a resurgence of mutual trust in the CJEU case law long ago after Cassis? Indeed, it is now common knowledge that the principle of mutual recognition had become a ghost in the rulings of the Luxembourg court.104 Such invisibility is certainly not unconnected to the strong criticism related to the regulatory use of mutual trust by the Commission in the EU legislation105 and its uneasy spillover into the other economic freedoms (beyond the free movement of goods).106 But after Tampere, mutual trust made its way in the Freedom Security

goods and services from the other member states. There is, however, a caveat. The Treaty of Rome gives member states the right to hinder trade and to regulate their markets, if they can claim overriding concerns. While Article 34 assures the freedom of goods, Article 36 mentions different exceptions. The other freedoms face similar restrictions. Additionally, by broadening the reach of the market freedoms under the Cassis de Dijon case law, the ECJ simultaneously enhanced the possibilities of member states to regulate their domestic markets, by invoking mandatory requirements to which goods would have to adhere (“rule of reason”).’ 102  See for examples (below nn 103, 105 and 106) the writings of Alter, Héritier, Kelemen, Lavenex, Nicolaidis and Schmidt. 103  SK Schmidt and D Kelemen (eds), The Power of the European Court of Justice (Abingdon, Routledge, 2013), 131. According to them, free movement rights provide the Court with general power to require the deregulation and liberalization of national economic regimes. Cassis is often related to the expansion of the freedoms. For Tuori (n 5) 143, Cassis is probably the most famous substantive ruling in the chain of economic constitutionalisation and the development of the microeconomic constitution. See also C Joerges, ‘Unity in Diversity as Europe’s Vocation and Conflicts Law as Europe’s Constitutional Form’, LEQS Paper No 28/2010, 28. For Joerges, through the upgrading of economic freedoms to constitutional rights, the CJEU has indeed assumed en passant constitutional functions. The issue, then, is of whether the Court has gone a step too far when complementing the recognition of the constitutional status of economic freedoms by its authoritative definition of the kind of concerns which are deemed to be compatible with the establishment of a common European market. 104  See J Pelkmans, ‘Mutual recognition: Economic and Regulatory Logic in Goods and Services’, Bruges European Economic Research Paper, 24/12. 105  R Barents, ‘New Developments in Measures Having Equivalent Effect’ (1981) 18 CMLRev 271, 296. The Commission has used a dubious interpretation of the judgment as a weapon to revive its crusade against protectionism. See also K Alter and S Meunier, ‘Judicial Politics in the European Community: European Integration and the Pathbreaking Cassis de Dijon Decision’ in K Alter, The European Court’s Political Power (Oxford, Oxford University Press, 2009) 139, 148. For them, Cassis is a good example of the CJEU attempting to make policy by promulgating a legal principle. This ruling introduces the general principle of recognition and the rule of reason was offered as potential qualification of the principle. In other words, the rule of reason served a sweetener. 106  JHH Weiler, ‘The Constitution of the Common Market Place: Text and Context in the Evolution of the Free Movement of Goods’ in P Craig and G de Bùrca (eds), The Evolution of EU Law (Oxford, Oxford University Press, 1999) 349, 367–68. Because of the inherently uncertain nature of where equivalence starts and where it stops, Weiler, for instance, deems mutual recognition inappropriate for building the single market. See A Héritier, ‘Mutual Recognition: Comparing Policy Areas’ (2007) 14 Journal of European Public Policy 800, 805. She compares the case of free movement of goods with the case of free movement of services. For her, it becomes evident that, in the area of services, there has been

98 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

and Justice legislation107 and now impacts again the CJEU case law as it is visible in Melloni and Opinion 2/13.108 Will this ‘new’ mutual trust follow the same fate as its economic counterpart? Or can it become a fully fledged principle of EU constitutional law recognised as such by the CJEU? The fate of mutual trust as a juridical principle is intimately linked to the constitutional reactions it will create. As we know, mutual trust is a non-absolute principle in EU law.109 There is no ‘pure mutual trust’, only ‘managed mutual trust’, to borrow the words of Nicolaidis.110 ‘Managed mutual trust’ is also a highly political concept as it results from Opinion 2/13. Reactions have been many. Even before its delivery, the ECHR case law on issues such as asylum and child abduction was clashing with the effectiveness approach taken by the CJEU in NS or Zarragua.111 After its delivery, Opinion 2/13 has led to the ideological reaction of another crucial constitutional actor: the Federal Constitutional Court of Germany. On 26 January 2016, the Federal Constitutional Court delivered its verdict in R, which addressed quite thoroughly the limits of mutual trust in relation to the sacrosanct human right of dignity within the context of the European Arrest Warrant (EAW).112 The decision dealt with the extradition proceedings of Mr R founded on an EAW issued by Italy. Mr R, a US citizen, had been sentenced in 1992 to thirty years imprisonment for cocaine smuggling and mafia membership by a court in Florence. He was not present during the trial in Italy and

no comparable judicial activism (807). See also K Nicolaıdis and SK Schmidt, ‘Mutual R ­ ecognition “On Trial”: The Long. Road to Services’ (2007) 14 Journal of European Public Policy 717, 719. The authors rightly point out that services come with people. They are frequently based on the ‘uno-actu’ ­principle, ie the service requires the movement of the provider to the point of consumption. See also S Lavenex, ‘Mutual Recognition and the Monopoly of Force: Limits of the Single Market Analogy’ (2007) 14 ­Journal of European Public Policy 762. She analyses the problems of applying mutual recognition to justice and home affairs: while markets are liberalised through mutual recognition, in justice and home affairs it serves as a tool of governmentalisation. 107  See eg Recital 21 of the Brussels IIbis Regulation, which states that: ‘The recognition and enforcement of judgments given in a Member State should be based on the principle of mutual trust and the grounds for non-recognition should be kept to the minimum required.’ 108 See Melloni (n 46) and Opinion 2/13 (n 3). 109  Lenaerts (n 99). 110 K Nicolaidis, ‘Kir Forever? The Journey of a Political Scientist in the Landscape of Mutual ­Recognition’ in M Maduro and L Azoulai, The Past and Future of EU Law (Oxford, Hart Publishing, 2010) 447, 451. Nicolaidis draws the distinction between managed mutual recognition and pure mutual recognition. For her, managed mutual recognition belongs to the political sphere. 111 See Case C-491/10 PPU Aguirre Zarraga, ECLI:EU:C:2010:828; Case C‑211/10 PPU Povse, ECLI:EU:C:2010:400 and Case C‑403/09 PPU Detiček, ECLI:EU:C:2009:810. For instance, in Aguirre Zarraga, a case dealing with the non-return of a child from Germany to Spain, it was asked whether the certificate provided for by Art 42 of the Brussels IIbis Regulation ordering the return of a child could be disregarded by a court in the Member State of enforcement in circumstances, where its issue amounted to a serious violation of fundamental rights (Art 24 of the Charter), or where that certificate contained a statement that was manifestly incorrect. The national court asked in particular whether it could oppose the enforcement of a judgment ordering the return of a child where—contrary to what is provided for by Art 42(2)(a) of the Brussels IIbis Regulation—that child had not been given the opportunity to be heard. 112  Decision No 4/2016 of 26 January 2016, ECLI:DE:BVerfG:2015:rs20151215.2bvr273514.

General Principles and the Many Faces of Coherence

 99

claimed not to have known about the trial. The German court which heard the case approved the extradition.113 The Federal Constitutional Court considered the issue of mutual trust in many places of its core reasoning.114 Even though the Federal Constitutional Court underlined the importance of mutual trust in EAW cases, it also pointed out that this trust may be shaken.115 More precisely, the German court that decides on an extradition is under an obligation to investigate and establish the facts of the case, The relevant facts in particular include what kind of treatment the requested person will have to expect in the requesting state. It does not follow from this obligation that the German courts always have to review in detail the reasons for a request for extradition. This holds true in particular in the context of extradition proceedings within Europe, where the principle of mutual trust applies. Yet, this trust is shaken if there are factual indications that the requirements that are absolutely essential for the protection of human dignity will not be met if the requested person is extradited.116 In other words, the mutual trust as applied within the EU would signify that no full examination of the principle of individual fault would be needed.117 Nevertheless, where the person subject to the extradition proceedings would claim and provide evidence that the Member State issuing the EAW would not comply with the minimum constitutional requirements the German court would have to investigate this claim further.118 The Federal Constitutional Court considered that even if the EAW would normally take precedence over German law this would not mean that the national courts would not be required to comply with the requirements of human dignity. Mutual trust can be limited by human dignity as protected by the German constitution.119 And therefore the automaticity of EAW decisions based on mutual trust may be put into question by the national judicial authorities. These authorities are in fact under an obligation to review whether the requirements under the rule of law have been respected even if the EAW formally meets the requirements of the Framework Decision. This national ruling post Opinion 2/13 puts clear limits on the application of mutual trust and on the procedural effectiveness of the EAW system when the right to dignity is in jeopardy. It means that the principle of mutual trust can not only be internally limited (ie by the EU Charter) but also externally 113  The FCC found that the constitutional complaint by Mr R (based on human dignity under the German Constitution) in the form of the principle of individual fault was admissible and that the national court’s decision violated this fundamental right. 114  J Nowag, ‘A New Solange Judgment from Germany—or Nothing to Worry About?’, Völkerrechtsblog, 22 March 2016. 115  Decision No 4/2016 of 26 January 2016 (n 112) paras 67 and 73. 116  D Robertson, The Judge as Political Theorist (New Haven, CT, Princeton University Press, 2010) 376. According to the author, one problem with dignity as a prime value used to flesh out other rights is that it is so expansive a concept. 117  Decision No 4/2016 of 26 January 2016 (n 112) para 73. 118  ibid, paras 73–74. 119  ibid, para 83.

100 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

limited (ie by a national constitution). This interpretation strikes a blow at the Melloni reasoning,120 which as seen before also forms the core of Opinion 2/13. Mutual trust is obviously a non-absolute principle. But the wide scope of its non-­absoluteness brings the risk of endangering its procedural effectiveness, its raison d’être. Interpreted broadly, the Decision in R may even be seen as an overruling of Melloni. As a matter of fact, for the German Court even fully harmonised areas of EU law can be reviewed in the light of the national constitution. On a more positive note, R can be viewed as an exemplification of the constitutional spillover of mutual trust and gives fuel to the claim that mutual trust is a constitutional principle. Indeed, it is possible to contend that mutual trust has become a recognised constitutional principle of EU law. After all, and at a more abstract level, isn’t it the case that trust is a mechanism for the reduction of social complexity and simplifies life by taking a risk.121 Trust is also about the recognition of the other or the recognition of other standards of protection, to put it in EU law terms.122 In other words, trust brings effectiveness and is a positive concept with positive consequences. Janssens in her recent study on the principle of mutual recognition concludes avec brio that a cross-policy approach to this principle is indeed both feasible and desirable.123 The feasibility of this approach is undeniable. However, its desirability perhaps deserves another answer in the light of the present situation. Looking at the constitutional reactions post Opinion 2/13 and at the momentum of perpetual crisis in which we are stuck,124 it is more than doubtful that mutual trust will and should spill over into the case law of the CJEU. After Cassis, l’histoire se répète-elle? In fact, we may have unfortunately entered the era of mutual distrust—the European spleen.125

120 

See in particular para 63. N Luhmann, Trust (Daidalos, 2005) see particularly ch 10 (at 127ff, of the Swedish translation). See also N Luhmann, ‘Familiarity, Confidence, Trust: Problems and Alternatives’ in D Gambetta (ed), Trust: Making and Breaking Cooperative Relations (2000) electronic edition, Department of Sociology, University of Oxford, ch 6, 94, 97–98. For the author, ‘Trust, on the other hand, requires a previous engagement on your part. It presupposes a situation of risk. You may or may not buy a used car which turns out to be a “lemon”. You may or may not hire a babysitter for the evening and leave him or her unsupervised in your apartment; he or she may also be a “lemon”. You can avoid taking the risk, but only if you are willing to waive the associated advantages.’ 122  P Ricoeur, The Course of Recognition (Cambridge, MA, Harvard University Press, 2005). Ricoeur starts by discussing the asymmetry of the self and the other, by referring to Husserl and Levinas who give directly opposed versions of the dissymmetry. Ricoeur’s point is to warn against forgetting the real dissymmetries in the search for mutuality between the self and the other. See also K Nicolaidis, ‘Trusting the Poles, Mark 2: Towards a Regulatory Peace Theory in a World of Mutual Recognition’ in I Lianos and O Odudu (eds), Regulating Trade in Services in the EU and the WTO: Trust, Distrust and Economic Integration (Cambridge, Cambridge University Press, 2012) 282. 123  C Janssens, The Principle of Mutual Recognition in EU Law (Oxford, Oxford University Press, 2013) 319. 124  J Habermas, The Crisis of the European Union: A Response (Cambridge, Polity, 2012) and K Tuori and K Tuori, The Eurozone Crisis: A Constitutional Analysis (Cambridge, Cambridge University Press, 2014). 125  C Baudelaire, ‘Spleen’, in Les Fleurs du Mal (recueil de poèmes, 1857). 121 

General Principles and the Many Faces of Coherence

 101

Is mutual distrust the new black, then?126 Unfortunately the answer seems to be ‘yes’ if we look introspectively at Opinion 2/13 once again. The message sent by the Court in Opinion 2/13 is limpid: the CJEU distrusts the ECHR case law since it may impede on the effectiveness of EU law. The list of changes asked by the CJEU to make possible the accession to the ECHR is long and unrealistic without changing the very nature of the Strasbourg Court: a human rights court founded on the principle of equality of the contracting parties.127 It appears clear that the Luxembourg Court is not ready to reach a compromise. It will do everything in its power to preserve the effectiveness of EU law and, therefore, to keep a wide margin of discretion when it comes to the review of acts of EU law impacting at the national level as is presently the case under the Bosphorus doctrine. In a recent article concerning the lessons to be drawn from Opinion 2/13, Judge Malenovský considers an alternative approach to the accession to the ECHR.128 For him, a new and more nuanced reflection is needed. This reflection should be articulated around taking into consideration the fundamental interests of the Union.129 The system and goals of the EU are here contrasted with the nation-state approach adopted within the Council of Europe. According to Malenovský, there is an essential need to respect the specificities of the EU law and therefore to be able to derogate from the principle of equality of contracting parties which lies at the foundation of the accession agreement.130 Malenovský insists in his conclusion that the renegotiation of the draft accession agreement should be more faithful to the will of the European constituant and respect the autonomy of EU law.131 To this end, he considers that the maintenance of the Bosphorus doctrine deserves particular attention.132 Yet the trust between the two European courts is arguably broken after Opinion 2/13. Suffice it to look at the comments of Dean Spielmann (the former president of the ECtHR) in his foreword to the 2014 ECHR Report.133 According to him, the decision not accede to the ECHR is a great disappointment—bearing in mind that negotiations on EU accession have been under way for more than thirty years, that accession is an obligation under the Lisbon Treaty and that all the Member States along with the European institutions had already stated that they considered the draft agreement compatible with the TEU and TFEU.134 These are not light words

126  Luhmann notes importantly that distrust is not only the contrary to trust; distrust is also its functional equivalent (Trust, n 121) 127 (ch 10). 127  See NL Arold, X Groussot and GT Petursson, ‘The Paradox of Human Rights Protection in Europe: Two Courts, One Goal?’ in OM Arnardottir and A Buyse, Shifting Centers of Gravity in Human Rights Protection (Abingdon, Routledge, 2016). 128 J Malenovský, ‘Comment Tirer Parti de l’Avis 2/13 de la Cour de l’Union Européenne sur l’Adhésion à la Convention Européénne des Droits de l’Homme’ [2015-4] RGDIP 705. 129  ibid, 733. 130  ibid, 734, 131  ibid, 740. 132  ibid, 735 and 740. 133  D Spielmann, Foreword, Annual Report 2014, Registry of the ECHR (Strasbourg 2015) 6. 134 ibid.

102 

Xavier Groussot, Jörgen Hettne and Gunnar Thor Petursson

in the mouth of the president of one of the most powerful courts in Europe. He then concludes by stating the blatant truth: ‘Let us not forget, however, that the principal victims will be those citizens whom this opinion (no 2/13) deprives of the right to have acts of the European Union subjected to the same external scrutiny as regards respect for human rights as that which applies to each Member State.’135 As discussed previously in this chapter, the tension between autonomy and assimilation (through recourse to the principle of equivalence under Article 52(3) of the Charter) has marked the case law of the CJEU on EU human rights since the entry into force of the Lisbon Treaty, which incorporates an obligation for the EU to accede to the ECHR. Will this affect the principle of comity (trust) r­ egulating the relationship between the CJEU and the ECtHR, which is epitomised by the principle of equivalence under Article 52(3) of the EU Charter of Fundamental Rights and the Bosphorus doctrine? Will this affect the coherence of human rights law in Europe (one of the rationales for acceding to the ECHR)? One thing appears to be rather clear after Opinion 2/13: the ideological coherence of the CJEU does not cohere with human rights protection in Europe.136

V. Conclusion The apocryphal saying ‘May you live in interesting times’ is perhaps fitting to describe the case law of the Court of Justice on general principles after the entry into force of the Lisbon Treaty and its binding Charter. Although the Charter’s purpose was undoubtedly to bring coherence into the fundamental rights jurisdiction of the EU, it has instead brought about numerous foreseen and unforeseen ‘interesting’ situations, where the level of coherence is tested. In this environment we have sought to ‘inject’ coherence, and demonstrated how coherence may be used to reflect the different functionalities of the general principles of EU law, including the fundamental rights. In that sense, the authority, applicability and inter-legality of fundamental rights was described via three interlinked concepts of coherence: creative coherence, normative coherence and transnational coherence. The creative background of the general principles in EU law is a part of its unique success as an instrument of (supra)international law that is applied throughout the national levels of the twenty-eight Member States of the EU. Allowing itself to be inspired by the national (constitutional) orders of the Member States, and their international obligations, when ‘observing’ the law, the fundamental rights were introduced by the Court of Justice in order to legitimise the reach of EU law. But here lies also the challenge, namely to reconcile the authoritative claims of

135 ibid. 136 

See Arold, Groussot and Petursson (n 127).

General Principles and the Many Faces of Coherence

 103

fundamental rights ‘recognised by the EU Charter’ vis-à-vis the general principles of EU law, both those found in the ECHR and those with a national reference. At the same time, it becomes clear that the normative coherence for EU fundamental rights, ie its applicative process and framework, is still very much ‘under construction’—even if no one expects full coherence to be reached. As a minimum it needs to be clear when and how the fundamental rights, which lie at the heart of EU’s legal structure, are applied in legal proceedings before the Court of Justice. From the recent legal development as regards the scope of the Article 41 of the EU Charter, it can be concluded that the CJEU, through divergent approaches, has created confusion when it comes to the application of this Article in the Member States. The latest development shows that Article 41 cannot be applied in itself in that context which means that the general scope of the Charter, defined in Article 51, is not respected. Certain Advocates General have tried to convince the Court that the exact wording of Article 41 should not be regarded as more important than the creation of a coherent general scope for the Charter. The CJEU, or rather its different Chambers, have however responded with expressions in variable ways without convincing explanations and reasoning. In this chapter we have tried to demonstrate that the difference between the scopes of Article 41 and 51 of the Charter actually can be explained and rationalised. National procedural autonomy must be respected, but national administrative law can meet the equivalence and effectiveness requirements of EU law, which could lead to a trade-off between efficiency requirements and other important interests worthy of protection. National law may therefore have to be’ adjusted’ if it prevents the effective application of EU law, but this is not the same thing as applying Article 41 in the Charter instead of national law. However, we think it is regrettable that the CJEU has not put more effort into providing a rational explanation based on coherence. We therefore argue that the Court must take its constitutional responsibility more seriously and strive for a coherent legal development of the Charter and the general principles, also when it comes to apparent inconsistencies. This is necessary if its case law is to be applied and respected in all national courts and tribunals in the Member States. On the other hand, coherence is difficult to achieve when the substantive general principles are rubbing shoulders with structural principles such as mutual trust and effectiveness. The Opinion 2/13, in that respect, constitutes an axiomatic illustration of the ‘ideological coherence’ of the CJEU.

104 

7 General Principles and Customary Law in the EU Legal Order SAMANTHA BESSON La formation d’une coutume communautaire est-elle concevable? Les traités sont encore trop récents pour en juger. Mais en soi, il ne faut pas exclure la possibilité, en présence surtout des nombreuses lacunes des traités, que des usages s’implantent, qu’ils soient acceptés par les institutions et par les Etats membres et que, pour finir, une conviction juridique vienne à s’installer à ce sujet. … Il n’est pas exclu que des conventions puissent s’établir sous forme coutumière. P Pescatore, L’ordre juridique des Communautés européennes—Etude des sources du droit communautaire (Liège, Presses universitaires de Liège, 1975) 174

I. Introduction Focusing on European Union customary law (EUCL) in a discussion pertaining to the general principles of EU law (EUGP) may sound surprising at first. Why broach one source of EU law, and a prima facie practically irrelevant one at that, to address another one? It is precisely because they are both (unwritten) sources of EU law that their comparison and relationship are arguably relevant. And this even more so as it is from the perspective of the doctrine of the sources of EU law that I propose to approach EUGP in this chapter. What is striking about EUGP is their disproportionate quantitative and qualitative importance in EU law1 by comparison to what is the case both in domestic

1  See eg S Besson, ‘General Principles in International Law—Whose Principles?’ in S Besson and P Pichonnaz (eds), Les principes en droit européen—Principles in European Law (Zurich, Schulthess, 2011) 19–64; J van Meerbeeck, ‘Les principes généraux du droit de l’Union européenne’ in Y Cartuyvels et al (eds), Les sources du droit revisitées—normes internationales et constitutionnelles, vol 1 (Bruxelles, Anthemis, 2012) 161–206; J Bengoetxea, ‘Principia and Teloi’ in S Besson and P Pichonnaz (eds), Les principes en droit européen—Principles in European Law (Zurich, Schulthess, 2011) 69–88; A Arnull, ‘What Is a General Principle of EU Law?’ in R de la Feria and S Vogenauer (eds), Prohibition of Abuse of Law: A New General Principle of EU Law (Oxford, Hart Publishing, 2011) 7–24; A von Bogdandy,

106 

Samantha Besson

legal orders2 and in the international legal order.3,4 This is surprising on at least two counts. In contrast to what applies in domestic law5 and in international law,6 first of all, there is hardly any mention in practice of customary law among the unwritten sources of EU law. This is surprising because EU law is usually conceived of either as a regional regime of international (organisation) law or as a special municipal legal order. Even if it is correctly conceived as neither, it is striking to encounter a legal order that does not, or only rarely, recognises customary law among its sources.7 This is even more peculiar as the role of EUCL has been limited in practice since the inception of the EU legal order. As a matter of fact, it is still a

‘Founding Principles of EU Law: A Theoretical and Doctrinal Sketch’ (2010) 16 European Law Journal 95–111; D Simon, ‘Les principes en droit communautaire’ in S Claudal (ed), Les Principes en Droit (Paris, Economica, 2008) 287–304; U Bernitz, J Nergelius and C Cardner (eds), General Principles of EC Law in a Process of Development (Austin, Wolters Kluwer, 2008); X Groussot, General Principles of Community Law (Groningen, Europa Law Publishing, 2006); T Tridimas, The General Principles of EU Law, 2nd edn (Oxford, Oxford University Press, 2006); C Flaesch-Mougin, ‘Typologie des principes de l’Union européenne’ in Centre d’excellence Jean Monnet (ed), Le droit de l’Union européenne en principes, Liber amicorum en l’honneur de Jean Raux (Rennes, Apogée, 2006) 99–152; U Bernitz and J Nergelius (eds), General Principles of European Community Law (The Hague, Kluwer Law International, 2000); M Herdegen and K Wolters, ‘General Principles of EU Law—The Methodological Challenge’ in U Bernitz, J Nergelius and C Cardner (eds), General Principles of EC Law in a Process of Development— Reports from a Conference in Stockholm, 23–24 March 2007, organised by the Swedish Network for ­European Legal Studies (Alphen aan den Rijn, Kluwer Law International, 2008) 343–55; O Wiklund and J Bengoetxea, ‘General Constitutional Principles of Community Law’ in U Bernitz and J Nergelius (eds), General principles of European Community Law (The Hague, Kluwer Law International, 2000) 119–35; JA Usher, General Principles of European Community Law (London, Longman, 1998); D Simon, ‘Y a-t-il des Principes Généraux du Droit Communautaire?’ (1991) 14 Droits 73–86. 2  See eg G Tusseau, ‘Métathéorie de la notion de principe dans la théorie du droit contemporaine. Sur quelques écoles de définition des principes’ in S Caudal (ed), Les principes en droit (Paris, Economica, 2008) 75–112; R Guastini, ‘Les principes de droit en tant que source de perplexité théorique’ in S Caudal (ed), Les principes en droit (Paris, Economica, 2008) 113–26. 3  See eg G Gaja, ‘General Principles of Law’ in Max Planck Encyclopedia of Public International Law (Oxford, Oxford University Press, 2007), online edition, available at www.mpepil.com/, accessed 1 September 2015; R Kolb, ‘Principles as Sources of International Law (with Special Reference to Good Faith)’ (2006) 53 Netherlands International Law Review 1–36; E Jouannet, ‘L’ambivalence des principes généraux’ in R Huesa Vinaixa and K Wellens (eds), L’influence des sources sur l’unité et la fragmentation du droit international (Bruxelles, Bruylant, 2006) 115–54; H Ascensio, ‘Principes généraux du droit’ in Répertoire international Dalloz 2004. 4  Importantly, my argument does not depend on the political nature of the EU, and in particular on whether it should be regarded as an international organisation or as a state. I am indeed assuming it has the (autonomous) legal order it claims to have and am only focusing on its law’s sources. The type of international subjecthood of the EU does not matter for its capacity to have a legal order in the first place (see J Klabbers, ‘The Concept of Legal Personality’ (2005) 11 Ius Gentium 35–66). Of course, the EU’s political nature can be relevant with respect to the identity of its law’s subjects and hence, partly at least, to the grounds for its legitimacy. However, assuming for now that both a state and an international organisation can be democratic, the political nature of the EU does not seem decisive for the relevance of EUCL. 5  See eg Art 1(2) of the 1907 Swiss Civil Code. 6  See eg Art 38(1)(b)(c) of the 1945 Statute of the International Court of Justice. 7  I am agnostic here about whether the usual sources of law (legislation, custom, principles) are a contingent feature of a legal order or a conceptual-normative one. The fact is that there are striking

General Principles and Customary Law in the EU Legal Order

 107

relatively young legal order and unwritten law should therefore be called to play an important gap-filling role,8 as confirmed ironically by the comparatively important gap-filling function of EUGP in EU law. Actually, even in long-established domestic legal orders where material customary law tends to diminish in relevance, customary law is still called to play a role in the institutional or constitutional realm.9 And the same could be said of the regional or functional customary law that often develops through established inter-state or institutional practice within international organisations (IOs).10 Secondly, in contrast to domestic and international law, there are no discussions in practice pertaining to the relationship between EUGP and EUCL. This is equally surprising because general principles are usually regarded, either interchangeably or at the same time, as both a type of norms and a source, at least in domestic and international law. This raises the difficulty of their relationship to other sources of law and especially to unwritten sources such as customary law. In fact, customary law is the source of many general principles of law qua norms, thus making its relationship to general principles qua source even more ambiguous. This may be exemplified by reference to the debate surrounding the legal positivist understanding of general principles,11 and especially the idea of ‘judicial custom’ qua social source of these principles in a legal order. Finally, general principles and customary law are not only both unwritten sources of law, but share common features in their process of identification and formation, thereby making the distinction between them often difficult. These three issues are well known in international law where the conflation between general principles and customary international law is common. The relationships between both sources are also frequently considered in domestic law. Of course, the lack of existence or relevance of EUCL in practice depends on which practice one considers. To start with, there is very little discussion of EUCL in the case law of the Court of Justice of the European Union (CJEU),12 for reasons

commonalities on sources from one legal order to the other in practice. Thus, even if having customary law among its sources is a contingent feature of the domestic and international legal orders, it is worth wondering why EU law differs in this respect from both domestic and international law. See also JL Martí, ‘Legitimacy and the Sources of International law’ in S Besson and J d’Aspremont (eds), Oxford Handbook on the Sources of International Law (Oxford, Oxford University Press, 2018, forthcoming). 8  See also P Pescatore, L’ordre juridique des Communautés européennes—Etude des sources du droit communautaire (Liège, Presses universitaires de Liège, 1975) 174. 9 See eg T Blanchet and J Keller-Noëllet, ‘Peut-on parler de “coutume” en droit de l’Union ­européenne?’ in Y Cartuyvels et al (eds), Les sources du droit revisitées—Les normes internationales et constitutionnelles, vol 1 (Brussels, Anthemis, 2012) 207–52; JP Jacqué, ‘La pratique des institutions communautaires et le développement de la structure institutionnelle communautaire’ in R Bieber and G Ress (eds), The Dynamics of EU Law (Baden-Baden, Nomos, 1987) 377–405. 10  See eg C Peters, ‘Subsequent Practice and Established Practice of International Organizations: Two Sides of the Same Coin?’ (2011) 3 Goettingen Journal of International Law 617–42. 11  See eg Besson (n 1); C Semmelmann, ‘General Principles in EU Law between a Compensatory Role and an Intrinsic Value’ (2013) 19 European Law Journal 457–87. 12  As a matter of fact, the only explicit reference to the term ‘custom’ can be found in Opinion of AG Maduro in Case C-133/06 Parliament v Council, [2008] ECR I-03189 EU:C:2007:551, paras 27–29.

108 

Samantha Besson

that I will identify and discuss later. True, this need not be representative of the rest of the practice of EU law. EUCL can arise from inter-institutional or inter-state practice and may, as a result, be identified and, arguably, even validated by other EU institutions than the CJEU. Besides, one should not underestimate the role of scholarship in the acknowledgement of the existence and relevance of a given source in the legal practice. This is because of the role of the doctrine of sources in the latter’s functioning. Just as curiously, however, there has been very little scholarly interest in EUCL to date,13 and even less so in its relationship to EUGP.14 Most of the EU law literature that discusses EUCL is quite ancient. Moreover, it stems from French-speaking or Dutch-speaking EU law scholarship, thereby confirming, of course, how little interest there has been overall for the topic of sources in EU law in English-speaking scholarship over time.15 All the same, the little EU law literature there is about EUCL is generally favourable to it and emphasises its interesting features.16 Curiously, even the literature on international law in the EU legal order, and how both legal orders influence each other, does not pick up on EUCL. International lawyers and EU external relations lawyers alike focus on customary international law (CIL) and how it has been received within EU law (eg its validity, rank and effects) or, conversely, how the EU’s practice has been contributing to the identification and/or formation of CIL17—whether on its own, in the EU’s respective spheres of competence, or in lieu of and for its Member States as

13  See eg B de Witte, ‘Sources and the Subjects of International Law: The EU and the Sources of International Law’ in S Besson and J d’Aspremont (eds), Oxford Handbook on the Sources of International Law (Oxford, Oxford University Press, 2018, forthcoming); Blanchet and Keller-Noëllet (n 9); D Kochenov, ‘EU Enlargement Law: History and Recent Developments: Treaty—Custom Concubinage?’ (2005) 9(6) European Integration online Papers available at http://eiop.or.at/eiop/texte/2005-006a.htm, accessed 1 September 2015; T Koopmans, ‘General Principles of Law in European and National Systems of Law’ in U Bernitz and J Nergelius (eds), General Principles of European Community Law (The Hague, Kluwer Law International, 2000) 25–34; H Schermers and DF Waelbroeck, Judicial Protection in the European Communities, 5th edn (London, Kluwer Law and Taxation Publishers, 1992) 98; Jacqué (n 9); A Bleckmann, ‘Zur Funktion des Gewohnheitsrechts im europäischen Gemeinschaftsrecht’ (1981) 16 Europarecht 101–23; Pescatore (n 8). 14  Compare the separate chapters by Blanchet and Keller-Noëllet (n 9) and Van Meerbeeck (n 1) and their lack of cross-references. 15  Compare eg JP Jacqué, Droit institutionnel de l’Union européenne, 8th edn (Paris, Dalloz, 2015) 511ff; C Blumann and L Dubouis, Droit institutionnel de l’Union européenne, 5th edn (Paris, LexisNexis, 2013) 543ff with eg TC Hartley, The Foundations of European Union Law, 8th edn (Oxford, Oxford University Press, 2014) 83ff and 203ff; P Craig and G de Búrca, EU Law, Text, Cases, and Materials, 6th edn (Oxford, Oxford University Press, 2015) 105ff. 16  Contra: Blumann and Dubouis (n 15) 544. 17  See eg J Malenovsky, ‘Le juge et la coutume internationale: perspectives de l’Union européenne et de la Cour de justice’ (2013) 12 Law and Practice of International Courts and Tribunals 217–41; T Konstadinides, ‘When in Europe: Customary International Law and EU Competence in the Sphere of External Action’ (2012) 13 German Law Journal 1177–201; A Gianelli, ‘Customary International law in the European Union’ in E Cannizzaro et al (eds), International Law as Law of the European Union (Leiden/Boston, Martinus Nijhoff, 2012) 93–110; F Hoffmeister, ‘The Contribution of EU Practice to International Law’ in M Cremona (ed), Developments in EU External Relations Law (Oxford, Oxford University Press, 2008) 37–128; J Vanhamme, ‘Formation and Enforcement of Customary International Law: The European Union’s Contribution’ (2008) 39 Netherlands Yearbook of International Law 127–54.

General Principles and Customary Law in the EU Legal Order

 109

practised by other IOs.18 This neglect is particularly intriguing as the existence of EUCL definitely affects the contribution of the EU to the development of CIL. Of course, the latter may occur without EUCL, but the existence of EUCL conditions the development of some norms of CIL, such as international human rights law. Generally speaking, the existence of customary law internal to an IO is approached by international law scholars exclusively from the perspective of its contribution to CIL (or, worse, as subsequent international treaty practice under Article 31(3) (c) of the Vienna Convention on the law of treaties (VCLT)), and rarely as a source of IO law in itself and hence as functional or regional customary law.19 The neglect of EUCL, and its relationship to EUGP, in both the practice and scholarship of EU law is even more surprising as the stakes in matters relative to the sources of law are high in any legal order. This should be true of the EU legal order as well. Among these stakes, one should, first of all, mention the concern for the legitimacy of EU law. Sources should indeed be organised in such a way that the EU law they produce can claim legitimate authority. Given the specificities of the EU legal order, the duality of its subjects (individuals and states), and in particular its democratic claims to inclusion and participation,20 the issue of the legitimacy of EUCL and EUGP arises differently than in either domestic or international law, but especially in the latter. A second stake is the allocation and balance of powers, both horizontally among EU institutions (Article 13(2) TEU) and vertically among the EU and its Member States (Articles 4(1)(3) and 5(1)(2) TEU). Sources of EU law should indeed be organised so as to respect that idiosyncratic balance. Again, these procedural constraints differ from what may apply in the domestic legal order, but especially also in the international legal order. A third concern is a constitutional one. Sources of EU law should be organised so as to respect the constraints set by the EU constitutional order,21 and especially within the EU Treaties. This hierarchical dimension of the sources of EU law distinguishes them from those of the international legal order in particular, and affects both EUCL and EUGP and their potential relationship. There are many reasons one could put forward for the comparative imbalance between EUCL and EUGP in the EU legal order. Some are inherent to the Union’s supranational nature and idiosyncratic democratic legitimacy, whereas others pertain to changes in the relationship between EU law and international law, in the allocation of powers horizontally between EU institutions or vertically between

18  See eg J Wouters and P De Man, ‘International Organizations as Law-Makers’ in J Klabbers and Å Wallendahl (eds), Research Handbook on the Law of International Organizations (Cheltenham, Edward Elgar, 2011) 190–224; Third report (A/CN.4/682) on identification of customary international law presented by Sir Michael Wood, Special Rapporteur, on the sixty-seventh session of the ILC, 27 March 2015, 46–54. 19  See eg JE Alvarez, International Organizations as Law-Makers (Oxford, Oxford University Press, 2005). See, however, Peters (n 10). 20  See Opinion of AG Maduro in Case C-133/06 Parliament v Council (n 12) para 29. 21  See ibid, para 29. See, more generally, eg Opinion 2/13 (Opinion pursuant to Article 218(11) TFEU) EU:C:2014:2454, para 158.

110 

Samantha Besson

the EU and the Member States, and in the role of the CJEU and its case law in the EU legal order. As I will argue, some of these reasons are under strain currently, thereby leading to potential developments in the practice of EUCL. Understanding the specificities of EU law with respect to EUCL, and its relationship to EUGP, and the reasons for these peculiarities of the EU legal order in comparison to both domestic and international law constitutes this chapter’s first aim. Its second objective is to explain why and how some EUGP could be better approached as EUCL. Accordingly, my argument will be three pronged. In section II, I will briefly account for three theoretical issues pertaining to general principles in EU law: their types, sources and functions in the EU legal order in comparison to domestic and international law. Section III will address EUCL and especially its types and functions, but also how it differs from other written and unwritten sources of EU law, how it relates to them and why it has had so little relevance in practice, still by comparison to domestic and international law. In section IV, I will argue for a stronger relationship between EUGP and EUCL in terms of sources of EU law. I will illustrate the importance of doing so with respect to EU fundamental rights which are the epitome of EUGP.

II.  General Principles in EU Law General principles are broadly used and referred to throughout the EU Treaties and within the CJEU’s case law. Interestingly, general principles have a specific nature (II.A), specific sources (II.B) and a specific role (II.C) in EU law in comparison to what applies in domestic and international law.

A.  Types of General Principles in EU Law In a nutshell, EUGP can be understood as legal principles, ie as (i) fundamental legal norms that are (ii) structurally indeterminate.22 As it is the case domestically and internationally, the addition of the qualifications ‘general’ and/or ‘fundamental’ does not bring much to the extent that all legal principles are usually both general and fundamental. Although principles may apply to public authorities or to individuals,23 this article focuses on general principles of public law, ie on principles that apply to public authorities whether EU Member States or EU institutions. These general principles can also be referred to as constitutional principles.

22  23 

See Guastini (n 2). See Arnull (n 1).

General Principles and Customary Law in the EU Legal Order

 111

There are four types of general principles in EU law. EU lawyers, like international lawyers, are familiar, first of all, with general principles ‘common to the laws of the Member States’ that stem from common domestic traditions such as EU fundamental rights; second, with ‘general principles of international law’ such as pacta sunt servanda; and, third, ‘general principles of EU law’ such as direct applicability that are specific to the functioning of the EU. This third category entails all general principles that pertain to the constitutional and institutional articulation and organisation of the EU.24 EU law also knows of a fourth category of general principles, however, that comes close to the general principles one finds in domestic law: ‘general principles of law’. These are the principles one encounters in any domestic legal system in Europe, such as legal security in particular.

B.  Sources of General Principles in EU Law In terms of sources, EUGP are both considered a type of EU legal norm and a source of EU (unwritten) primary law.25 One may derive the latter from Article 263(1) and Article 340(2) TFEU in particular. Of course, general principles of EU law qua source of EU law include as a priority general principles that stem from common domestic traditions. There is actually an important mutuality of general principles that arise from domestic law, become European general principles and influence domestic principles in return. This has been the case for the principle of proportionality, in particular. Interestingly, the CJEU is quite specific regarding the method and criteria for the identification of the EUGP it considers as a source of EU law, ie those stemming from common domestic traditions. Denys Simon refers to that method as a ‘méthode éclectique, sélective et filtrante’.26 The CJEU starts by a comparative law analysis, although it has done so less and less rigorously in recent times and leaves it to the Advocate General in most cases.27 The criteria are, first of all, a commonality test that does not usually require unanimity among states or even a majority thereof;28 and, secondly, a transposability test the success of which depends on how well the principles fit the overall structure and objectives of

24  See von Bogdandy (n 1); N Tsagourias, ‘The Constitutional Role of General Principles of Law in International and European Jurisprudence’ in N Tsagourias (ed), Transnational Constitutionalism— International and European Models (Cambridge, Cambridge University Press, 2007) 71–106. 25  See P Pescatore, ‘Les principes généraux du droit en tant que source du droit communautaire’ in P Pescatore (ed), Études de droit communautaire européen 1962–2007 (Bruxelles, Bruylant, 2008) 691–730. 26 See Simon, ‘Y a-t-il des Principes Généraux’ (n 1) and Simon, ‘Les principes en droit communautaire’ (n 1). 27  See Arnull (n 1). 28 See Opinion of AG Lagrange in Case C-14/61 Koninklijke Nederlandsche Hoogovens en Staalfabrieken v ECSC High Authority [1962] ECR 485 EU:C:1962:19.

112 

Samantha Besson

the EU.29 In view of the case law, the two tests tend to merge into one another as certain domestic traditions are held to be compatible with the general objectives of the EU and hence to be sufficiently common, while others do not for the very same reasons.30 EUGP also have other sources than domestic general principles, and in particular (written) primary EU law, but also the CJEU’s case law. Starting with the latter, first of all, whether one sees the CJEU as only interpreting and specifying the EUGP one finds in EU primary law, or actually as creating them or turning them into EU law through judicial custom, its case law is crucial to general principles. Judicial law amounts indeed to the most important source of legal principles as the latter need to be identified, but also interpreted even if they are guaranteed legally and pre-exist judicial interpretation. By nature, legal principles require interpretation and a normative assessment in every case. Arguably, not only do legal principles require the judiciary,31 but judges need legal principles to exercise their functions fully.32 As in domestic and international law, secondly, general principles are gradually specified and codified into primary or secondary EU law. As I explained before, this does not prevent their judicial interpretations from developing them further (see eg Article 6(3) TEU in the context of EU fundamental rights). The distinction between written and unwritten general principles is fluid, as a result. Implicit principles often get codified and become explicit, but they subsist as implicit principles. Implicit principles may actually also be extracted from explicit ones through interpretation. As a result, codified principles may impact on codified law through a retroaction process. The CJEU has not yet defined a strict method as to how to identify those general principles of EU law that do not stem from primary EU law—and do not derive from domestic traditions qua autonomous source of EU law either.33 This may be explained by the unease the idea of judicial law creates in general.34 By contrast to international law, however, the issue is not so much the lack of hierarchy or exclusivity of the judiciary and its lack of compulsory jurisdiction or of

29  See Case C-11/70 Internationale Handelsgesellschaft mbH v Einfuhr- und Vorratsstelle für Getreide und Futtermittel [1970] ECR 1125 EU:C:1970:114, para 4. 30 Compare Case C-159/90 Society for the Protection of Unborn Children Ireland v Grogan and others [1991] ECR I-4685 EU:C:1991:378 or Case C-341/05 Laval un Partneri [2007] ECR I-11767 EU:C:2007:809 with Case C-36/02 Omega Spielhallen- und Automatenaufstellungs-GmbH v Oberbürgermeisterin der Bundesstadt Bonn [2004] ECR I-09609 EU:C:2004:614, paras 34ff. 31  See also P Brunet, ‘Les principes généraux du droit et la hiérarchie des normes’ in D de Béchillon, P Brunet, V Champeil-Desplats and E Millard (eds), L’architecture du droit. Mélanges en l’honneur de Michel Troper (Paris, Economica, 2006) 207–21. 32  See eg Groussot (n 1) 105ff. 33  See eg the discussion pertaining to the existence of a general principle of equality of shareholders or of abuse of law in the Case C-101/08 Audiolux and Others [2009] ECR I-9823 EU:C:2009:626 and Case C-255/02 Halifax and Others [2006] ECR I-1609 EU:C:2006:121. 34  See Herdegen and Wolters (n 1).

General Principles and Customary Law in the EU Legal Order

 113

democratic legitimacy and accountability. One may indeed argue that the CJEU, and the European judicial dialogue that is so specific to the European legal order lato sensu, have actually developed in the way they did through the elaboration of general principles.35 Rather, the challenge to judicially created EUGP has to do with the vertical and horizontal division of powers within the EU, on the one hand, and between the EU and its Member States, on the other.36 Recent case law shows how justified these fears about a general principles’ ‘competence creep’ are.37 Interestingly, the CJEU has actually responded to those concerns recently by not over-interpreting general principles.38 Finally, in terms of rank, it follows from the nature and role of EUGP that they need to be interpreted and specified when applied in concrete circumstances and, in most cases, balanced against other legal principles. Their balancing is guided by the moral and political values they protect. It cannot indeed occur by reference to a formal hierarchy of legal norms: principles stem from different sources, on the one hand, and even when they stem from one single source such as judicial law, they are meant to refer the interpreter to moral values and hence relate to an axiological hierarchy and not a legal one, on the other. This becomes clear when one thinks of constitutional principles whose constitutional nature can derive à la fois from their constitutional entrenchment and from their use in constitutional judgements, without any difference being made either way as to their rank in case of balancing. In the context of EU law, one may actually consider that the constitutional function of (most) general principles, combined with their normative importance, would justify granting them constitutional rank in most cases.39 This has consequences in turn for the formal hierarchy of sources within EU law: if there is a hierarchy between EUGP and EU primary law, indeed, it is a material one at most (see also Article 52(4) EU Fundamental Rights Charter (EUFRC) a fortiori regarding EU fundamental rights). And the same may be said of their relations to one another.

35 

See Groussot (n 1). S Prechal, ‘Competence Creep and General Principles of Law’ (2010) 3 Review of European Administrative Law 5–22. Contra: K Lenaerts and J Gutierrez Fons, ‘The Constitutional Allocation of Powers in the EU and General Principles’ (2010) 47 Common Market Law Review 1629–69, 1630. 37  See eg in the EU fundamental rights context: Case C-34/09 Gerardo Ruiz Zambrano v Office national de l’emploi (ONEm) [2011] ECR I-1177 EU:C:2011:124. See eg in the EU private law context: Case C-144/04 Werner Mangold v Rüdiger Helm [2005] ECR I-9981 EU:C:2005:709; Case C-555/07 Kücückdeveci [2010] ECR I-365 EU:C:2010:21; Case C-176/12 Association de médiation sociale [published in the electronic Reports of Cases] EU:C:2014:2. See T Tridimas, ‘Horizontal Effect of General Principles: Bold Rulings and Fine Distinctions’ in U Bernitz, X Groussot and F Schulyok (eds), General Principles of EU Law and European Private Law (Alphen aan den Rijn, Kluwer Law International, 2013) 213–232. 38  See eg Case C-101/08 Audiolux (n 33); Case C-255/02 Halifax (n 33). See also, most recently, Case C-176/12 Association de médiation sociale (n 37). 39  See Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat International Foundation v Council and Commission [2008] ECR I-06351 EU:C:2008:461, paras 307–08 (on fundamental rights); Opinion 2/13 (n 21) para 191 (on mutual trust). 36  See

114 

Samantha Besson

C.  Functions of General Principles in EU Law The functions of general principles in EU law follow closely the different functions general principles have in domestic law, but also in international law.40 General principles are both an aid to and a constraint on the judicial interpretation of EU law.41 Their reference to values helps to justify other EU law norms, but also contributes to the development of the rule of law in EU law and hence to the legitimacy of EU law as a whole. The two functions of general principles one can distinguish in domestic and international law, ie gap-filling and coherence, can also be identified in the CJEU’s practice. They have evolved both quantitatively and qualitatively in the EU legal order, however. General principles helped to fill gaps, first of all, while EU law was developing. It seems from the recent case law, however, that there are too many potential principles available currently and that the CJEU often chooses to leave a gap open nowadays rather than fill it with yet another principle.42 One may also observe how general principles codified in EU primary law are constantly being reinterpreted by the CJEU, thus ensuring a dynamic understanding of EU law. In qualitative terms too, the gap-filling function has turned into an intervalidation function where general principles allow principles to be transferred from one legal order to the other, and in particular from domestic or international law into EU law, while respecting the autonomy of each of the respective legal orders. Secondly, general principles also ensure the coherence of EU law. Here again, coherence has evolved with EU law and the greater integration between legal orders in the EU. Although their invocation does not imply hierarchy or unity, general principles contribute to the integrity of the European legal order lato sensu and hence preserve the legal pluralism that prevails between the EU legal order and that of its Member States.43 General principles can even be regarded as a way of providing materially superior norms and a transitive normative hierarchy that can allow EU law and general international law to be articulated in richer terms than dualist ones within the European legal order lato sensu.44

III.  EU Customary Law Although it is the most important unwritten source of law in both domestic and international law, customary law has received very little recognition in the practice 40 

See Guastini (n 2). See also Bengoetxea (n 1). 42  See Case C-101/08 Audiolux (n 33) and Case C-255/02 Halifax (n 33). 43  See S Besson, ‘From European Integration to European Integrity. Should European Law Speak with Just One Voice?’ (2004) 10 European Law Journal 257–81. 44  See eg Opinion of AG Maduro in 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 EU:C:2008:11. 41 

General Principles and Customary Law in the EU Legal Order

 115

and scholarship of EU law. In what follows, and by reference to the structure of the discussion of EUGP adopted in the previous section, I will consider the different types of EUCL one encounters in practice (III.A), their relationship to other sources of EU law (III.B) and their various functions (III.C), always by comparison to what is the case in domestic and international law.

A.  Types of EU Customary Law EUCL can be understood, like customary law in general, as a normative practice, ie (i) a general, coherent and regular practice (ii) regarded as law and hence as legally binding. Confirmation of the applicability of the so-called two-element doctrine in EU law may be found in the CJEU’s case law.45 There is very little information in that very case law, however, about how these criteria actually work. Of course, as I explained before, the CJEU’s case law is extremely limited on EUCL in general and the reasoning pertaining to its ascertainment even more. Domestic and international courts are not usually very specific about the criteria they use for the determination of customary law in any case. Depending on the identity of the subjects and the object of customary law, one may distinguish between different types of EUCL: inter-individual, inter-­ institutional and/or inter-state EUCL; and material and/or constitutional EUCL. First of all, the subjects of EUCL could be any of the subjects of EU law: individuals, EU institutions or EU Member States. This is what one would expect from a comparison with domestic and international customary law. The subjects of any norm of customary law are indeed those practising what could thereby become a custom. In EU law, EUCL has developed mostly between EU institutions and/ or between EU Member States. This is a consequence of the object of EUCL, as we will see next, but also of the sheer scope of the general EU practice needed to give rise to inter-individual EUCL and the difficulty of meeting that condition in practice at EU level. As it is the case in international law, inter-state EUCL may also arise from a sufficiently general, regular and coherent intrastate practice. Secondly, EUCL may be either material or institutional depending on whether its object is any part of EU material law or only EU institutional law. This is a distinction one encounters within domestic customary law where customary law has progressively, and primarily due to its gradual codification, become mostly institutional or constitutional. Constitutional customary law is also referred to domestically as ‘constitutional custom’ or ‘constitutional convention’. In EU law, constitutional customary law is also the most common type of EUCL one can encounter.46 Its norms pertain to various constitutional aspects of the EU, and in

45  46 

See Case C-230/81 Luxembourg v Parliament [1983] ECR 255 EU:C:1983:32, para 44. See Blanchet and Keller-Noëllet (n 9) 217; Jacqué (n 9) 377.

116 

Samantha Besson

particular to inter-institutional relations or intra-institutional organisation. However, this does not preclude the development of material EUCL, in principle. In this chapter, I will focus on constitutional EUCL, whether it is inter-institutional or inter-state, as this provides a nice contrast to the constitutional EUGP discussed in the previous section. The examples of constitutional EUCL one finds in practice may be organised under two headings: inter-institutional and inter-state.47 There are more examples of the former than of the latter, for reasons that have to do with the jurisgenerative effects of the self-organisation powers of EU institutions and their duties of loyal co-operation.48 Of course, examples of EUCL are mostly historical to the extent that EUCL, once it is sufficiently established to be qualified as such, is often rapidly codified into EU primary law for reasons of legal security. Inter-institutional EUCL is also often turned into inter-institutional agreements. To that extent, EUCL may be compared to EUGP whose relationship to EU primary law is dynamic and characterised by the toing and froing of codification and interpretation processes. Among historical examples of norms of inter-institutional EUCL, on the one hand, one may mention the 1966 Compromis du Luxembourg49 or the Parliament’s role in the consultation procedure.50 Both practices arose as custom and were then codified into EU primary law. Historical examples of inter-state EUCL, on the other, may be found in the EU standards for enlargement51 or for the election of CJEU judges (Article 255 TFEU). As I explained in the introduction, it is difficult to find many examples of EUCL in the CJEU’s case law. The latter only rarely mentions inter-institutional practices, and, even more rarely, inter-state practices. When it does, moreover, it does not refer to them expressly as ‘customary’. The only exception is Advocate General Maduro’s 2007 Opinion which refers explicitly to ‘extra-statutory customs’ and to ‘precedents’ from the institutions or Member States.52 Unsurprisingly, therefore, the Court’s reasoning is also usually very brief. The CJEU’s case law on EUCL confirms the distinction between inter-institutional and inter-state customary law, albeit not nominally, of course. The Court has never accepted a case of inter-state customary law to date. The reasons given are either that such law could be equated to an amendment of the EU Treaties contrary to Article 48 TEU, as in the Defrenne I case,53 or that it would be overriding or

47  See Blanchet and Keller-Noëllet (n 9) 221–22, 225–38, 244–47; Blumann and Dubouis (n 15) 455–56. 48  See Jacqué (n 9) 381–93. 49  See eg Blumann and Dubouis (n 15) 260ff; Craig and G de Búrca (n 15) 134–35. 50  See Case C-138/79 Roquette frères [1980] ECR 3333 EU:C:1980:249; Case C-65/93 Parliament v Council [1995] ECR I-643 EU:C:1995:91. 51  See Case C-93/78 Mattheus v Doego [1978] ECR 2203 EU:C:1978:206. See also Kochenov (n 13). 52  See Opinion of AG Maduro in Case C-133/06 Parliament v Council (n 12) paras 27–29. 53  See Case C-43/75 Defrenne v Sabena (Defrenne I) [1976] ECR 455 EU:C:1976:56, paras 57–58; Case C-59/75 Manghera [1976] ECR 91 EU:C:1976:14, paras 19–21.

General Principles and Customary Law in the EU Legal Order

 117

derogating from them.54 The Court’s concern seems to be the protection of the settlements agreed at times of Treaty revisions. The Court has been equally hard on ‘negative practice’ or custom when it is invoked to derogate from a norm of EU primary law that is no longer practised.55 It has been more generous in recognising instances of inter-institutional customary law,56 however.57 Thus, one could mention the confirmation by the Court of the legality of a resolution of the Parliament to hold meetings of its committees and political groups in Brussels58 or the legality of the management committee procedure.59 At first sight, this reluctance of the Court is surprising because, whether in domestic or international law, courts are the ones vested with the pre-enforcement identification of customary law, and arguably with its validation. There are many reasons one could venture, however. The first one, of course, and the most charitable one, is the lack of relevance of EUCL in practice and hence the few customary norms to actually enforce. The second reason may be that the CJEU in reality resorts to other sources to do what a domestic or international court would usually do with customary law. As a matter of fact, the CJEU uses general principles to fill gaps or to specify indeterminate EU law norms, both functions customary law usually fulfils in domestic and international law.60 This is not surprising given how closely related judicial reasoning and general principles are to one another in general. Nor should it come as a surprise given how unconstrained judicial lawmaking generally is in EU law, by comparison to domestic law, but especially to international law. As I argued before, however, the lack of recognition of EUCL in judicial practice does not take anything away from the latter’s independently acquired validity through practice. Nor does it prevent other EU institutions, such as the Council or the Commission, from recognising it, and thereby arguably validating it, if need be.

54 See Case C-174/84 Bulk Oil [1986] ECR 559 EU:C:1986:60, para 65; Case C-68/86 United ­ ingdom v Council [1988] ECR 855 EU:C:1988:85, para 24; Case C-327/91 France v Commission K [1994] ECR I-3641 EU:C:1994:305, para 36; Opinion 1/94 [1994] ECR I-5267 EU:C:1994:384, para 52; Case C-426/93 Germany v Council [1995] ECR I-3723 EU:C:1995:367, para 21; Case C-271/94 Parliament v Council [1996] ECR I-1689 EU:C:1996:133, para 24; Case C-84/94 United Kingdom v Council [1996] ECR I-5755 EU:C:1996:431, para 19; Case C-133/06 Parliament v Council [2008] ECR I-3189 EU:C:2008:257, paras 55–57, 59–60. See also Opinion of AG Maduro in Case C-133/06 Parliament v Council (n 12) paras 26–35. See also Jacqué (n 9) 401; Blanchet and Keller-Noëllet (n 9) 240–41. 55  See Case C-7/71 Commission v France [1971] ECR 1003 EU:C:1971:121, paras 18–22, 23–26. 56  See Case C-25/70 Einfuhr- und Vorratsstelle für Getreide und Futtermittel v Köster and Berodt & Co [1970] ECR 1161 EU:C:1970:115, paras 6–9; Case C-230/81 Luxembourg v Parliament (n 45) paras 43–46, 48–49; Case C-149/85 Wybot [1986] ECR 2391 EU:C:1986:310, paras 16–17; Case C-137/10 Région de Bruxelles-Capitale [2011] ECR I-3515 EU:C:2011:280, paras 19–22; Opinion of AG Cruz-Villalon in Case C-137/10 Région de Bruxelles-Capitale [2011] ECR I-3515 EU:C:2011:16, paras 48–49. 57  See Blanchet and Keller-Noëllet (n 9) 242–47. 58  See Case C-230/81 Luxembourg v Parliament (n 45) paras 43–46, 48–49. 59  See Case C-25/70 Köster (n 56) paras 6–9. 60  See eg Kolb (n 3); Tridimas (n 1).

118 

Samantha Besson

B.  EU Customary Law and the Other Sources of EU Law One of the features of the EU legal order, especially by comparison to international law, is that its sources are ranked by reference to one another. One may see this as a sign of the constitutionalisation of EU law, but the hierarchy of sources of EU law may also be justified more generally on grounds of procedural legitimacy. While EU primary law is usually ranked first and EU secondary law last, a growing trend in recent years has been the distinction, within EU primary law, between a constitutional and superior tier of EU primary law, which includes EU fundamental rights,61 and a lower tier. This has consequences in turn for the rank of EUCL in the EU legal order. That rank matters, first, by reference to written sources such as EU primary and secondary law, but also, second, to non-written sources of EU law such as EUGP.

(i)  EU Customary Law and the Written Sources of EU Law Like EUGP, EUCL is dynamic in its relationship to written EU law. Once it is sufficiently established, it is often codified into written law. Of course, nothing prevents codified EUCL from being interpreted specifically or differently and hence from giving rise to a new subsequent practice and hence to a new custom. EUCL, especially when it is constitutional, can be codified (‘germinated’)62 into EU primary law, but also, when it is inter-institutional, into inter-institutional agreements.63 As a matter of fact, EU primary law has been progressively amended and made more flexible through different instruments of amendment/revision, and this has enhanced these forms of codification of EUCL even further. Furthermore, the recognition of inter-institutional agreements as a source of binding EU law by the Lisbon Treaty (Article 295 TFEU)64 or the enhanced co-operation mechanism (Articles 20 TEU and 328 TFEU) amount to another way in which EU Treaties accommodate within EU primary law inter-state or inter-institutional practices that may become or already amount to EUCL. The CJEU itself has often contributed to the establishment of these constitutional conventions in the making, by turning them into correct interpretations of EU primary law.65 As a result, the relationship between EUCL and written sources of EU law is fluid. Importantly, given the superior rank of EU primary law in the EU legal order, EU primary law is taken as setting constraints on the validity of EUCL. This is one

61 

See Joined Cases C-402/05 P and C-415/05 P Kadi (n 39) para 285. See Blanchet and Keller-Noëllet (n 9) 231; Jacqué (n 15) 34–35. 63  See eg Blanchet and Keller-Noëllet (n 9) 218ff; Blumann and Dubouis (n 15) 455–56; Jacqué (n 15) 28ff. 64  See Blanchet and Keller-Noëllet (n 9) 218. 65  See eg Case C-138/79 Roquette frères (n 50) paras 33–36; Case C-139/79 Maizena GmbH v Council of the European Communities [1980] ECR 3393 EU:C:1980:250, paras 27–28. 62 

General Principles and Customary Law in the EU Legal Order

 119

of the key differences between EUCL and customary international law. Constitutional constraints on customary international law are quite common in domestic law, however.66 In EU law, constitutional constraints on EUCL include the following EU primary law rules and principles in particular: rules on the attribution of competences (Articles 4(1) and 5(1) TEU) and the principles of institutional balance (Article 13(2) TEU), of institutional autonomy and of loyal co-operation (Article 4(3) TEU), all of them both inter-state and inter-institutional. Interestingly, most of these constraints are general principles identified by the CJEU itself, whether or not they have yet been codified into EU primary law. This raises the question of constitutional EUCL and of the applicability of the formal priority of rank of EU primary law to that kind of EUCL. This is a problem currently faced by constitutional EUGP and their relationship to contrary EU primary law. In these cases, as I explained before, formal hierarchies do not seem to help resolving normative conflicts between EUGP and EU primary law norms, thereby leaving their resolution to material normative hierarchies. The constitutional equivalence of certain norms of EUCL and EU primary law may therefore constitute an exception to the CJEU’s ruling that inter-state and inter-institutional EUCL may not contradict or derogate from EU primary law.67 The justification for this ‘rigid’ approach, to quote Advocate General Maduro, to the priority of EU Treaties is the formal procedure of revision of those treaties.68 Of course, this argument is understandable when the priority of EU Treaties over EUCL is discussed by contrast to the relationship of equivalence between international treaties and customary international law. It is also envisaged with respect to the relevance of the VCLT in matters of EU Treaties—Article 48 TEU is meant to work as a lex specialis. This is particularly striking with reference to the exclusion of subsequent treaty practice under Article 31(3)(c) VCLT, but also of established practice under Article 5 VCLT or, more generally, of the possibility of treatyderogating customs (indirectly tolerated by Article 39 VCLT).69 However, once the issue is approached from an internal EU constitutional perspective, it is unclear that the ‘rigid nature of the founding Treaties’ can be a sufficient argument in order to rank constitutional EUCL below EU primary law. Conflicts between them should rather be resolved through balancing and by reference to material normative hierarchies when they exist.

(ii)  EU Customary Law and the Other Unwritten Sources of EU Law Even though the relationship between unwritten sources of EU law is not foreseen by the formal hierarchy of sources of EU law, it is important to consider it 66 

See eg Federal Supreme Court of Switzerland, 13 June 2013, Judgement 6B_218/2013, para 3.3. Case C-68/86 United Kingdom v Council (n 54) para 24; Case C-7/71 Commission v France (n 55) paras 18–22, 23–26; Case C-133/06 Parliament v Council (n 54) paras 55–57, 59–60. See also Opinion of AG Maduro in Case C-133/06 Parliament v Council (n 12) paras 26–35. 68  See Opinion of AG Maduro in Case C-133/06 Parliament v Council (n 12) paras 27–29. 69  See de Witte (n 13). 67  See

120 

Samantha Besson

carefully, not the least because it may affect our discussion of the relationship between EUCL and EUGP later in this chapter. Qua unwritten EU law, EUCL should be compared to other unwritten sources of binding and non-binding EU law. The latter include gentlemen’s agreements70 and other non-binding conventions between EU Member States or institutions. I will focus on the former in this section: they include EUGP and EU judicial law. First of all, the relationship between EUCL and EUGP. As I explained before, customary law and general principles are usually very difficult to distinguish. This may be explained by reference to the ambivalence about the source of general principles that are sometimes regarded as a type of norms and sometimes as both a type of norms and a source of these norms. Further, the unwritten nature of general principles and the fact that customary law often entails general principles also contribute to their frequent conflation. In international law, the tendency to conflate them has even become strategic since general principles tend to be easier to establish procedurally than customary international law for they do not require a practice.71 Curiously, in EU law, this discussion has not yet taken place. There is no trace of it in the CJEU’s case law either. This is even more surprising as many EUGP have arisen from inter-state practice, such as EU fundamental rights or the proportionality principle. Moreover, the comparative law method used by the CJEU in ascertaining the existence of some EUGP is reminiscent of the method used by domestic or international courts to establish the inter-state practice constitutive of customary law. Finally, EUGP and EUCL have a certain dependence on judicial reasoning and recognition in common. Second, EUCL and judicial law-making by the CJEU. When the Court interprets EU law infra or praeter legem, it contributes to its development, thereby contributing to EU law-making writ large. It actually sometimes does so in circumstances where there is an existing inter-state and/or inter-institutional practice. The question that arises in this context is whether EU judicial law should comply with the conditions for the identification or formation of customary law, ie establishing a general, coherent and regular practice and opinio juris, or whether it can expand the scope of EU law without constraints. This issue arose in the context of the Watson and Belmann case in the 1970s. Advocate General Trabucchi recommended that the Court ascertained the conditions for an inter-institutional and inter-state ‘rule of custom and practice’ before expanding the personal scope of the principle of free movement of persons to

70 

See Blanchet and Keller-Noëllet (n 9) 247–48. eg ICJ, 15 June 1962, Case concerning the Temple of Preah Vihear (Cambodia v Thailand), Merits, Judgment, ICJ Reports 1962, 23; ICJ, 20 April 2010, Case concerning Pulp Mills on the River Uruguay (Argentina v Uruguay), ICJ Reports 2010, 101. See also P d’Argent, ‘Les principes généraux à la Cour internationale de Justice’ in S Besson and P Pichonnaz (eds), Les principes en droit européen— Principles in European Law (Zurich, Schulthess, 2011) 107–20. 71  See

General Principles and Customary Law in the EU Legal Order

 121

include tourists under EU Treaties.72 He was not followed by the Court, however. The CJEU interpreted EU primary law by extending its personal scope without first establishing the existence of an inter-institutional or inter-state custom to that extent and without ascertaining the other conditions for the existence of a custom in that context.73 The CJEU’s approach to the potential relationship between judicial interpretations of EU law and EUCL confirms the way in which it identifies and develops general principles in its case law, ie without much attention to their sources. EU judicial law or, in the case of EUCL, EU judicial custom seems to be self-sufficient.74

(iii)  Functions of EU Customary Law The functions of EUCL are the same as those of customary law in domestic and international law. As this was famously captured in Article 1(2) Swiss Civil Code, its primary role actually lies in complementing written law. Customary law fills the gaps left by written law, on the one hand, and helps making it more specific when it is indeterminate, on the other. This has been confirmed for EUCL by Advocate General Maduro.75 The CJEU has not been more explicit about the role played by EUCL in the EU legal order. What is clear from the Court’s practice, however, is that, when there is EU primary law in place, EUCL may only be used to develop it infra or praeter legem, but never contra legem. EUCL cannot be invoked either to amend76 or derogate77 from EU primary law. This is a consequence of the primacy of EU primary law mentioned before. At first sight, the lack of relevance of EUCL in practice does not come as a surprise given its complementary functions. After all, as has been the case in domestic law, the development of the EU legal order and the progressive codification of practices imply a reduced relevance of EUCL over time. The difficulty, however, is that it has always been like that in EU law.78 Moreover, this has been the case both for material and constitutional EUCL, whereas domestic constitutional customs tend to remain in place and flourish even in well-established legal orders.

72  Opinion of AG Trabucchi in Case C-118/75 Watson and Belmann [1976] ECR 1185 EU:C:1976:79, 1204–05. 73  Case C-118/75 Watson and Belmann [1976] ECR 1185 EU:C:1976:106. 74  See also Blanchet and Keller-Noëllet (n 9) 214–15; Blumann and Dubouis (n 15) 455–56. 75  See Opinion of AG Maduro in Case C-133/06 Parliament v Council (n 12) para 29. 76  See Case C-43/75 Defrenne I (n 53) paras 57–58; Case C-59/75 Manghera (n 53) paras 19–21. 77  See Case C-7/71 Commission v France (n 55) paras 18–22, 23–26; Case C-174/84 Bulk Oil (n 54) para 65; Case C-68/86 United Kingdom v Council (n 54) para 24; Case C-327/91 France v Commission (n 54) para 36; Opinion 1/94 (n 54) para 52; Case C-426/93 Germany v Council (n 54) para 21; Case C-271/94 Parliament v Council (n 54) para 24; Case C-84/94 United Kingdom v Council (n 54) para 19; Case C-133/06 Parliament v Council (n 54) paras 55–57, 59–60. See also Opinion of AG Maduro in Case C-133/06 Parliament v Council (n 12) paras 26–35. See also Jacqué (n 9) 401; Blanchet and Keller-Noëllet (n 9) 240–41. 78  See also Pescatore (n 8) 174.

122 

Samantha Besson

The grounds for the sidelining of EUCL are to be found elsewhere, as a result. There are at least four reasons one could put forward for the limited role of EUCL in the EU legal order: the supranationality of EU law, its democratic legitimacy, its complex vertical and horizontal division of powers, and its jurisprudential nature. First of all, the supranationality of EU law. One of the specificities of the EU qua international organisation is its supranational nature. This may be verified in numerous features of its legal order, but especially in the EU law-making process. The sources of EU law are such that EU Member States are no longer the sole lawmakers. This even applies to the EU Treaties whose adoption and revision procedures include EU institutions. Some of them represent the interests of Member States albeit on majoritarian grounds (Council), while others represent EU citizens (Parliament) or even the interest of the EU itself (Commission) (Article 10 TEU). No wonder, then, that a source such as EUCL whose subjects could be either states or institutions only does not fare very well in this supranational setting. Of course, nothing precludes certain normative practices and hence EUCL from being both inter-state and inter-institutional, thereby abiding by the supranational specificity of EU law. This was actually Advocate General Trabucchi’s argument in the context of the free movement practice in Watson and Belmann.79 Thus, whereas EUCL may at first seem to be at odds with supranational EU law-making, it is in fact compatible with it provided it is both inter-state and inter-institutional. Secondly, the democratic legitimacy of EU law. Another characteristic of the EU’s supranational legal order is that its legitimacy is democratic and hence ultimately grounded in the political equality of EU citizens (Article 9 TEU). More specifically, it is through the involvement of the European Parliament in most EU law-making procedures that EU law’s democratic justification is acquired. Even if inter-state EUCL may be said to be indirectly democratic to the extent that it is through the general practice of democratic Member States that it is adopted, the implication of the EU Parliament qua requirement of supranational democracy is not necessarily given. The same may be noted of some instances of interinstitutional EUCL in which the equal participation of the European Parliament is not necessarily granted. It is without surprise therefore that the lack of democratic legitimacy also underpins Advocate General Maduro’s rejection of derogatory EUCL: he refers to ‘EU citizens’ and to the ‘reinforced legitimacy’ of the EU Treaties.80 Of course, there is a paradox in considering EUCL as lacking indirect democratic credentials when EUGP are increasingly used in the CJEU’s case law, including in order to derogate from EU Treaties, and without any reference to either EU citizens’ or the EU Parliament’s approval. Thirdly, the complex horizontal and vertical division of powers in EU law. Another feature of the EU legal order is its complex division of powers. It is a combination of horizontal (between EU institutions) and vertical (between EU

79  80 

Opinion of AG Trabucchi in Case C-118/75 Watson and Belmann (n 72) 1204–05. See Opinion of AG Maduro in Case C-133/06 Parliament v Council (n 12) para 29.

General Principles and Customary Law in the EU Legal Order

 123

Member States and the EU) division, the former replicating some of the latter to a certain extent. The way in which EUCL is made may at first be considered as a threat to this sensitive balance of powers.81 After all, even when it arises from a combination of inter-state and inter-institutional practice, it is difficult to see how it could abide perfectly by the constitutional constraints and procedures set by the EU Treaties to respect that balance. In those conditions, the fear is that EUCL could contribute to the ‘competence creep’ already at work in the EU. Interestingly, however, EUGP themselves have also been criticised for contributing to that ‘competence creep’, as I explained before. In this respect, one may consider the combined inter-state and inter-institutional dimension of EUCL as vouching for a greater respect of the sensitive vertical and horizontal balance of powers within the EU. After all, EUCL derives from inter-state and inter-institutional practice and opinio juris, and hence from what both the Member States and institutions see as being the right division of powers in practice. Finally, the jurisprudential nature of EU law. The CJEU has played a crucial role in the development of the EU legal order qua autonomous legal order. Nowhere is that role more tangible than in the context of the sources of EU law: they have been gradually moulded by the Court, whose role has been to ascertain norms arising from these sources and sometimes even to validate them. The lack of case law, and of careful reasoning, about EUCL in the Court’s practice may be explained by the very role the CJEU plays within the EU legal order itself. This came out clearly from its reasoning in Watson and Belmann, and from the hiatus between the proposed reasoning in the Advocate General’s opinion and the Court’s. The Court does not see the need to justify its reasoning by grounding it in (other) sources of EU law. In turn, this explains why it is not interested in identifying new constraints on its reasoning based on inter-state or inter-institutional practice. As a matter of fact, this accounts for the comparative advantage of EUGP over EUCL for the CJEU: EUGP are easier to establish than EUCL, do not constrain judicial reasoning as much as EUCL and fulfil some of the gap-filling functions of EUCL. In other words, the CJEU prefers to claim the autonomy of EU law and to avoid reliance on EUCL exactly because it gives itself more flexibility. If it had to identify customary principles at Member State level, it could never have decided cases such as Omega or Mangold in the way it did. Interestingly, some of these grounds for the limited role of EUCL in practice are currently under strain. This is because the EU is changing internally, but also externally in its relationship to other states and international organisations. First of all, the return of inter-state law in the EU. There has been a clear intergovernmental trend at play within the EU since 2011. This has in particular led to the (re)introduction of international law methods and sources in the relations between EU Member States, including on issues which would have been dealt with through the méthode communautaire and through EU law at earlier times.

81 

See Blanchet and Keller-Noëllet (n 9) 213.

124 

Samantha Besson

It suffices to mention the various inter-state treaties concluded in the wake of the financial crisis since 2008, but also the ways in which these instruments have led to the twisting or softening of the EU constitutional framework.82 Of course, the EU Treaty crisis is not a new phenomenon: it has been with us since 2004 and the failure of the Constitutional Treaty. What is new, however, is that the increased flexibilisation introduced or revised in the EU Treaties at Lisbon (eg simplified revision procedures under Article 48 TEU, flexibility clause under Article 352 TFEU or enhanced co-operation under Article 20 TEU) have not managed to bring EU Member States back to EU primary law. On the contrary, Member States have retreated into the inter-governmental periphery, both in terms of institutions and sources of law-making. Under these circumstances, one should make a virtue of a necessity, and bring inter-state conventions and customs back into the scope of EU law by broadening the sources of EU law.83 This would ironically ensure that the constitutional and democratic constraints of EU law are respected. Maintaining the integrity of the EU legal order may now come at this price. Secondly, the questioning of judicial law in the EU. The role of the Court in EU law-making is being challenged, especially when contrasted with that of EU Member States. Of course, the opposition between legislation and judicial law is common in any democracy. What is specific in the EU, however, is that it is also a polity of democratic states.84 This implies that the democratic legitimacy of judicial law also needs to be accounted for from an inter-state perspective. In this context, one may hope the CJEU could refer more regularly and more rigorously to EUCL to support its reasoning and interpretations in the future. This is even more important after Opinion 2/13 and the direction in which the CJEU is leading the Member States in terms of international law.85

IV.  General Principles qua EU Customary Law Based on the previous sections, I would now like to argue that some general principles of EU law, ie those that derive from the Member States’ legal orders, are best approached as customary law principles (IV.A). I will illustrate this point in a specific context, that of EU fundamental rights (IV.B).

82  See M Dawson and F de Witte, ‘Constitutional Balance in the EU after the Euro-Crisis’ (2013) 76 Modern Law Review 817–44. See also Case C-370/12 Pringle [published in the electronic Reports of Cases] EU:C:2012:756; Case C-62/14 Gauweiler and Others [nyr] EU:C:2015:400. 83  See Blanchet and Keller-Noëllet (n 9) 250–51. 84 See S Besson, ‘European Human Rights, Supranational Judicial Review and Democracy— Thinking Outside the Judicial Box’ in P Popelier, C Van de Heyning and P Van Nuffel (eds), Human Rights Protection in the European Legal Order: The Interaction Between the European and the National Courts (Cambridge, Intersentia, 2011) 97–145. 85 See PJ Kuijper, ‘Reaction to Leonard Besselink’s ACELG Blog’, https://acelg.blogactiv. eu/2015/01/06/reaction-to-leonard-besselinks%E2%80%99s-acelg-blog/, accessed 1 September 2015.

General Principles and Customary Law in the EU Legal Order

 125

A. Sources of General Principles Common to the Laws of the Member States Many EUGP arise from inter-state practice and come very close to customary general principles. This is clearly the case of the EUGP that are ‘common to the laws of EU Member States’. While these principles are one out of four types of EUGP, they are also the most specific ones in EU law and the kind of general principles that have been most prevalent in the EU legal order. As a matter of fact, EUGP and EUCL are even more intimately related than general principles and custom in other legal orders. Indeed, they share similar identification and formation criteria. First of all, the comparative law method used by the CJEU in ascertaining the existence of some EUGP86 is reminiscent of the method used by domestic or international courts to establish the interstate practice constitutive of customary law. Secondly, the inter-state practice the CJEU is after is general only, and need not be unanimous for a general principle to be considered as sufficiently common to become an EU general principle. Thirdly, the reference to domestic case law in that context comes close to the establishment of EU Member States’ opinio juris: the practice of domestic courts is indeed evidence of the grounds for compliance behind the practice. All this confirms that EUCL could in fact be regarded as the source of some EUGP. If this is the case, it is important the CJEU acknowledges and complies with customary law reasoning in all cases pertaining to the general principles common to EU Member States. This would be particularly justified now that the increasing relevance of EUGP in the EU legal order and their development by the CJEU are facing challenges. Some of these challenges could indeed be met were the customary source of some EUGP clearly recognised and the Court’s reasoning adapted accordingly. This is the case primarily of the need to enhance the constraints on the CJEU’s law-making powers and judicial reasoning. EUCL has ­indirect democratic credentials that derive from (non-judicial) inter-state practice that EUGP do not have in this respect. A second challenge lies in the ‘competence creep’ through EUGP that some have identified in practice.87 The customary law-ascertainment process in the Court’s reasoning could ­contribute to make those principles more attuned to an inter-state and inter-institutional practice in a given area and hence to preserving the allocation of powers within the EU.

86  87 

See Case C-11/70 Handelsgesellschaft (n 29). See eg Prechal (n 36).

126 

Samantha Besson

B.  EU Fundamental Rights qua EU Customary Law Among the EUGP that stem from EU Member States’ common principles, the most historical and fundamental ones in the EU legal order are EU fundamental rights. They developed as general principles through the CJEU’s case law.88 As a matter of fact, EUGP are still recognised as a source of these rights now that EU fundamental rights have been codified into the EU Fundamental Rights Charter (Article 6(3) TEU; Article 52(4) EUFRC; Preamble to the EUFRC) and in other parts of EU primary and secondary law. True, from 2010 onwards, the Charter has been treated as a starting point in the CJEU’s reasoning on EU fundamental rights.89 However, this has not always been followed strictly. Importantly, moreover, this does not imply a priority in reasoning with either a Charter right or a general principle, nor a formal hierarchy in case of conflict between them. As I explained before, indeed, the normative relations between EUGP, on the one hand, and between EUGP and other sources of EU law, on the other, do not fit the formal hierarchy of sources very well in practice. Thus, no formal priority or hierarchy should be assumed between the EU fundamental rights of Article 6(1) and (3) TEU depending on their respective sources (see also Article 52(4) EUFRC a fortiori).90 Based on my argument in the previous section, EU fundamental rights could actually be approached as customary in source. Not only does this interpretation fit the sources of EU fundamental rights, but it also best justifies their specificities. In terms of sources, on the one hand, EU fundamental rights are said to arise bottom-up from EU Member States’ respective internal practices of fundamental rights. Once there is sufficient commonality in that practice, even if it is not unanimous as confirmed by the CJEU in the Omega case,91 these rights may become EU fundamental rights. Moreover, EU fundamental rights work as protection maxima and not as minimal thresholds of protection.92 Both features are best explained by reference to their customary source. 88  See eg Case C-29/69 Stauder [1969] ECR 419 EU:C:1969:57, para 7; Case C-4/73 Nold [1974] ECR 491 EU:C:1974:51, para 13; Case C-11/70 Handelsgesellschaft (n 29) paras 3–4. See, more recently, Case C-521/09 P Elf Aquitaine v Commission [2011] ECR I-8947 EU:C:2011:620, para 112; Case C-571/10 Kamberaj [published in the electronic Reports of Cases] EU:C:2012:233, paras 60–61. 89  See eg Joined Cases C-92/09 and 93/09 Volker und Markus Schecke and Eifert Schecke [2010] ECR I-11063 EU:C:2010:662; Case C-236/09 Association Belge des Consommateurs Test-Achats and Others [2011] ECR I-773 EU:C:2011:100. 90 See H Hoffmann and B Mihaescu, ‘Relation between Charter’s Fundamental Rights and Unwritten General Principles’ (2013) 9 European Constitutional Law Review 73–101, 77–82. See also Case C-604/12 HN [nyr] EU:C:2014:302. 91  See Case C-36/02 Omega (n 30) paras 34–35. 92  See LFM Besselink, ‘Entrapped by the Maximum Standard: On Fundamental Rights, Pluralism and Subsidiarity in the European Union’ (1998) 35 Common Market Law Review 629–80 and JHH Weiler, ‘Fundamental Rights and Fundamental Boundaries: On the Conflict of Standards and Values in the Protection of Human Rights in the European Legal Space’ in JHH Weiler, The Constitution of Europe (Cambridge, Cambridge University Press, 1999) 102–29. See also Case C-550/07 P Akzo Nobel Chemicals and Akcros Chemicals v Commission [2010] ECR I-8301 EU:C:2010:512.

General Principles and Customary Law in the EU Legal Order

 127

On the other hand, the customary approach to EU fundamental rights best accounts for their specificities in practice. First of all, the function of EU fundamental rights may be compared neither to domestic (federal) fundamental rights nor to international fundamental rights.93 This can be checked easily by reference to their personal (Article 51(1) EUFRC)94 and material scope (Article 51(1) EUFRC), but also to their maximal rather than minimal nature (Article 53 EUFRC).95 These two features are best justified if the sources of EU fundamental rights are considered to be transnational and to grow out of the intra- and inter-state practice of EU Member States, rather than imposed top-down and in a centralised fashion on them. Second, there is no general human rights competence of the EU. EU fundamental rights have indeed grown out of domestic fundamental rights in joint spheres of competence.96 The customary account of the sources of EU fundamental rights therefore succeeds in squaring the human rights duties of EU Member States nicely with the principle of conferral of powers. Third, EU fundamental rights are only rarely applied and interpreted in conjunction with international human rights (except for the ECHR on grounds of Article 53(3) EUFRC). This is best justified by reference to the customary and transnational nature of EU fundamental rights: it is because EU fundamental rights have grown from the inside and from each Member State’s practice as an integrated97 fundamental rights regime that they are not concerned with minimal external guarantees that apply directly to those Member States’ own fundamental rights regimes. Finally, EU fundamental rights work as unwritten constitutional constraints on EU primary law. This is best justified again once their sources are considered as customary and concurrent to EU primary law, rather than internal to the latter (see also Article 52(4) EUFRC a fortiori with respect to the Charter itself and Article 6(1) TEU). This is a common argument made about domestic constitutional customary law and its role as supra-constitutional constraint.

93  Of course, with the development of EU democracy and citizenship, this intermediary transnational position held by EU fundamental rights has become increasingly uncomfortable: see eg Case C-617/10 Åkerberg Fransson [published in the electronic Reports of Cases] EU:C:2013:105, para 29 and Case C-399/11 Melloni [published in the electronic Reports of Cases] EU:C:2013:107, para 60. See S Besson, ‘The Bearers of Human Rights Duties and Responsibilities for Human Rights–A Quiet (R)Evolution’ (2015) 32 Social Philosophy & Policy 244–268, 265–66. 94  See Case C-617/10 Fransson (n 93) para 29. 95  See Case C-399/11 Melloni (n 93) para 60. 96 See S Besson, ‘The Human Rights Competence in the EU—The State of the Question after Lisbon’ in G Kofler, M Poiares Maduro and P Pistone (eds), Taxation and Human Rights in Europe and in the World, 5th GREIT Conference (Amsterdam, IBFD, 2011) 37–63. 97  See P Eeckhout, ‘Human Rights and the Autonomy of EU Law: Pluralism or Integration?’ (2013) 66 Current Legal Problems 169–202. See also S Besson, ‘Human Rights and Constitutional Law: Mutual Validation and Legitimation’ in S Rowan Cruft, M Liao and M Renzo (eds), Philosophical Foundations of Human Rights (Oxford, Oxford University Press, 2015) 279–99; S Besson, ‘European Human Rights Pluralism—Notion and Justification’ in M Poiares Maduro, K Tuori and S Sankari (eds), Transnational Law—Rethinking European Law and Legal Thinking (Cambridge, Cambridge University Press, 2014) 170–205.

128

Samantha Besson

One should therefore hope for more consideration of the customary source of EU fundamental rights in future reasoning by the CJEU. Some authors, such as Grainne de Búrca,98 have actually already recommended, albeit on different grounds, that the Court resorts to a more rigorous comparative law method in its human rights reasoning.99 This would actually merely imply a return to the comparative fundamental rights reasoning used by the Court in its original decisions on fundamental rights.100

V. Conclusions The doctrine of sources constitutes an interesting angle to grasp the comparative nature and role of EUGP in the EU legal order. Unlike previous research on the topic, this chapter’s comparative focus has not been on their relationship to EU primary law, but to another source of unwritten EU law: EUCL. Contrary to what is the case in domestic and international law, customary law has been largely neglected in the EU legal order, and so has its relation to EUGP. This is particularly striking as the latter relationship has long obsessed domestic and international lawyers. There have been different reasons for these debates in domestic and international law: both general principles and customary law are unwritten sources of law, and they actually share some dimensions in their identification and formation processes; there is an ambivalence between general principles qua source and qua type of norms, and customary law is actually the source of many principles qua norms; finally, that ambivalence has been used strategically, especially in international law, to escape the conditions of identification and formation of one or both sources. This chapter started by fleshing out the comparative imbalance in EU law between the qualitative and quantitative importance of EUGP and the relative irrelevance of EUCL. Some reasons for the reluctance towards EUCL in EU law have been identified and then debunked in the chapter: the supranationality of EU law, its democratic legitimacy, its idiosyncratic vertical and horizontal division of powers, and its jurisprudential nature. The chapter also explained how some of these grounds are currently under strain and are turning into arguments

98 See G de Búrca, ‘After the Charter of EU Fundamental Rights: The Court of Justice as a Human Rights Adjudicator?’ (2013) 20 Maastricht Journal of European and Comparative Law 168–184, 171–4. 99 On comparative or transnational human rights law in general, see eg Besson (n 97); S Besson, ‘Human Rights as Transnational Constitutional Law’ in A Lang and A Wiener (eds), Global Constitutionalism Handbook (London, Edward Elgar, 2017, forthcoming); C McCrudden, ‘A Common Law of Human Rights: Transnational Judicial Conversations on Human Rights’ (2000) 20 Oxford Journal of Legal Studies 499–532, 530. 100 See eg Case C-29/69 Stauder (n 88); Case C-4/73 Nold (n 88); Case C-11/70 Handelsgesellschaft (n 29).

General Principles and Customary Law in the EU Legal Order

 129

for EUCL. The EU is changing internally, but also externally in its relationship to other states and international organisations. In these circumstances, EUCL may have a new role to play, especially in light of two important challenges: the role of international law in the EU and the growing critique of the role of the CJEU. After exploring EUGP’s and EUCL’s respective types, sources (respectively, source-based relations) and functions in EU law by comparison to domestic and international law, this chapter uncovered and explained some of their ties both in terms of functions and of sources. On that basis, it argued that the most important EUGP for the development of the EU legal order, ie the general principles common to the laws of EU Member States, are best understood as customary law principles. This approach accounts in particular for their convergent practice of identification that is comparative, general and judicial. The argument was then exemplified in the field of EU fundamental rights. The proposed customary reading of EU fundamental rights not only accounts for the practice of identification of these rights and their role in EU law, but justifies some of their specific traits, and in particular: their transnational nature by contrast to both domestic or international human rights law; their remote relation to international human rights law; their concurrent relationship to EU primary law; and the lack of a general fundamental rights competence of the EU. The upshot of the proposed argument is the enhanced legitimacy of EUGP qua EUCL. The customary reading helps to address some of the current challenges facing EUGP. It makes those principles arguably less dependent on judicial law and the CJEU, more comparative and less threatening to the vertical and horizontal allocation of powers within the EU. In the context of EU fundamental rights, this argument is also a signal to the CJEU whose future case law on human rights could gain in democratic legitimacy were it to be more comparative and attuned to the domestic practices of EU Member States.

130 

Part II

Proportionality

132 

8 Proportionality ALISON L YOUNG AND GRÁINNE DE BÚRCA

When asked to choose a general principle which has most influenced the ­development of public law across Europe, it seems difficult—one might almost say manifestly disproportionate—to choose any principle other than proportionality. It is hard to deny the influence of proportionality, both as a general principle of EU law which pervades the case law of the European Court of Justice and as the means through which the European Court of Human Rights determines whether restrictions placed on Convention rights are ‘necessary in a democratic society’. As such, proportionality is a principle which is frequently applied by domestic courts, even within those legal systems that have not adopted a specific or general test of proportionality. And once courts have had experience of applying proportionality, it seems almost inevitable that the principle will expand its influence into domestic law. It seems also hard to conclude, in view of the extensive literature on the topic, that proportionality is anything other than a modern invention, an indication of the willingness of the judiciary to subject discretionary decisions of the administration to a more searching scrutiny. Proportionality, therefore, appears to be part of the recent move towards greater accountability, with more emphasis being placed on the enforcement of the rule of law by the judiciary in order to control actions of both the legislature and the executive. Proportionality is also seen by many as closely linked to the protection of human rights, both in terms of its role in decisions of the European Court of Human Rights and its influential role in German law. Hence if there is a tale to tell of the influence of proportionality, it seems to be one of its growing influence across Europe, leading to greater judicial scrutiny, more accountability and a stronger protection of human rights. We hope that this collection of essays goes some way to telling a more nuanced, if not a different story. Although it is hard to deny the influence of the European Union and the European Convention on Human Rights (ECHR), it is not necessarily true that proportionality has its origins in German jurisprudence, or that the growing influence of the principle is due solely to the role of the ­European Union and the ECHR. Nor is the influence of either the Strasbourg or the Luxembourg courts the same across the Member States of the European Union and the signatory states of the ECHR. And while it is clear that there is a c­ onnection

134 

Alison L Young and Gráinne de Búrca

between proportionality and human rights, it is not the case that proportionality is inevitably linked with human rights. When analysing different jurisdictions and the application of the proportionality test in the European Union we also see a role for proportionality in enforcing an aspect of distributive justice, ensuring in particular that administrative policies do not impose manifestly disproportionate burdens on particular individuals or groups. And whilst there may be, broadly speaking, a consensus surrounding what the test of proportionality comprises in the context of the ECHR, there is not necessarily a consensus surrounding the nature of the test of proportionality beyond this, particularly as concerns the connection between proportionality and the judicial control of rationality. Nor is the test of proportionality without its critics. If we learn anything from the fresh set of narratives contained in the chapters which follow, it is that the principle of proportionality is multifaceted and that a better understanding of its nature and practice may only be gleaned from separating out its different elements, assessing the extent to which different understandings of proportionality are suited to different subject matters, and examining how they are articulated and applied within different legal systems.

I.  The Origins of Proportionality and the Influence of Europe Paul Craig’s chapter provides the most robust challenge to the traditional narrative that the idea of proportionality in law is a modern creation, originating in German jurisprudence and strongly influenced by decisions of the German Bundesverfassungsgericht, which in turn influenced its adoption as a general principle of EU law and its role in the European Court of Human Rights. Craig’s analysis of seventeenth-century case law in England suggests that it is untrue that legal tests of review based upon the idea of proportionality are an invention of the twentieth century, or that judicial control over the discretionary powers of the administration is a purely modern phenomenon. Craig’s work also clearly demonstrates that, for English law at least, proportionality has a long history and a long pedigree of acceptance by both the judiciary and the legislature. It is no ­European import, originating instead in English law. Craig’s research demonstrates the use by the judiciary and the legislature in the seventeenth century of the terms ‘proportionable’ and ‘proportionability’. Proportionability was referred to expressly in legislation, prompting oversight by courts to ensure that a burden imposed upon an individual was proportionable. It seems that, where this was not referred to expressly in legislative provisions, the courts would nevertheless regard it as a principle of statutory interpretation, ensuring that benefits and burdens were allocated in a proportionable manner. Proportionability also operated as a stand-alone principle of judicial review in order to ensure a proportionable

Proportionality

 135

distribution of benefits and burdens. It was also used as a condition of regulatory intervention, such that even if there was provision for the King to impose a levy, it would be subject to the proviso that any such levies were imposed in a proportionable manner. Nor is the earlier existence of ideas of proportionality in domestic law, prior to the ‘Europeanisation’ of public law, unique to English law—although on the evidence of these chapters alone, proportionability in English law would appear to have the oldest pedigree. As Yoan Sanchez’s account of French law demonstrates, the test of proportionality in French law is often regarded as deriving from the 1933 decision of the Conseil d’Etat in Benjamin. However, Sanchez argues that the approach in Benjamin, which requires the court to ensure that exercises of police powers to protect public order can only restrict human rights when strictly necessary, derives from an earlier 1909 decision of the Conseil d’Etat in Abbé Olivier. Although not as old a pedigree as the seventeenth-century decisions identified by Paul Craig in English law, this development in French law nevertheless pre-dates both the European Union and the European Convention of Human Rights. It also differs from English law due to its specific focus on human rights, as opposed to the focus of a notion of proportionability which seeks to facilitate an equal or fair distribution of benefits and burdens. Martinico and Simoncini’s chapter on Italian law presents a similar picture to that of France, with the principle of proportionality arising originally as a tool to exercise control over the legitimate exercise of police powers. It is, however, hard to deny the influence of European law on the development if not the origins of proportionality, be that through the influence of EU law or through the work of the ECHR and European Court on Human Rights. This is particularly true in English law, where the test of proportionality is mostly applied in the context of the application of EU law and ECHR rights incorporated into English law through the Human Rights Act 1998. The Supreme Court in Keyu1 may have recently rejected the opportunity to adopt proportionality as a general test of review in English law—such a broad constitutional change requiring nine as opposed to five Justices of the Supreme Court—but it may well now be at least arguable that proportionality is the test of review for fundamental common law rights post Youssef,2 as well as being used in the protection of substantive legitimate expectations. Despite the undoubted role of EU law in the development of proportionality in English law, it is clear that it is the ECHR which has had the most influence in this context. A similar story is to be found in Danish law, where Helle Krunke remarks that there has been a rise in the application of the principle of proportionality through the influence of both EU law and the ECHR,

1  R (Keyu) v Secretary of State for Foreign and Commonwealth Affairs [2015] UKSC 69, [2015] 3 WLR 1665. 2  R (Youssef) v Secretary of State for Foreign and Commonwealth Affairs [2016] UKSC 3, [2016] 2 WLR 509.

136 

Alison L Young and Gráinne de Búrca

noting in particular the way in which the Convention is used to supplement the ­constitutional protection of rights found in older, perhaps one even might say outdated constitutions, helping to strengthen the protection of human rights. In Italian law, it is the European Union and not the ECHR which has had greater influence, with the application of proportionality in EU law leading to the development of the principle in Italian administrative law, its influence having culminated in the adoption of legislation requiring Italian administrative bodies to comply with EU law, regardless of whether their measures are based on EU law. There seems to be less of an influence, however, of either EU law or the ECHR in the Italian Constitutional Court, although the court is evidently more influenced by Strasbourg than by Luxembourg. In stark contrast, French law appears to be the least influenced by Europe, with Sanchez concluding that there is some influence of EU law when French courts apply proportionality in the field of EU law, as well as an influence of the ECHR in the domain of protection of rights in France. However, the French administrative courts have not been influenced by the structure of the proportionality test as applied by the Strasbourg court. It is hard to deny that both the European Union and the ECHR have played an influential role in the development of the principle of proportionality and in its adoption as a standard of review in domestic law across Europe. However, as this account of the origins of the test of proportionality suggests, care must be taken not to overstate or exaggerate the European influence; it is clear from the various chapters contained here that the test of proportionality—albeit in a different and less developed form—has national origins as well, some of which long predate its adoption and development by the European Union and the ECHR. It is true that the application of proportionability in English law was not the same as the sophisticated four-part test as currently applied in English law in the field of judicial review under the Human Rights Act 1998, or when English courts apply proportionality within the scope of EU law. In a similar manner, there is no mention by the Conseil d’Etat in Benjamin of a clearly structured three- or four-stage test. Nevertheless, it was clear that the English courts had a wealth of experience of applying a principle which is very similar to that of proportionality, ensuring that burdens were distributed in a proportionate manner, whilst being sensitive to the need to give weight to administrative determinations and the purposes of legislation governing the situation before the court. In a similar manner, the insistence of the French courts that restrictions on rights and liberties can only occur when necessary to ensure public order bears a remarkable similarity to more modern understandings of proportionality. However we classify these accounts of the origins of proportionality in England and France, they begin to cast doubt on the traditionally perceived narrative that the legal notion of proportionality is predominantly a creature of European law, be that from EU law or the ECHR, or even from German law.3 However, these accounts of the origins of proportionality 3  A Stone Sweet and J Matthews, ‘Proportionality Balancing and Global Constitutionalism’ (2008–09) 47 Columbia Journal of Transnational Law 73.

Proportionality

 137

also raise further questions as to the nature of proportionality and the scope of its application.

II.  What Is Proportionality? The test of proportionality would appear to have three key characteristics. First, it focuses in significant part on a control over outcomes (substance). Although this often includes an evaluation of the evidential basis on which the decision was made (process), proportionality is not limited to merely examining issues of process. Second, it is viewed as a more stringent form of control than a control of rationality. Third, it is a clearly structured three- or four-part legal test, depending on which elements are included. However, as the following chapters demonstrate, not all of these characteristics are uniformly present in all of the legal systems studied. Proportionality is designed primarily to control outcomes, ensuring that restrictions placed upon rights are proportionate, that benefits and burdens are allocated in a proportionate manner, or that, when achieving a particular ­outcome, the means chosen to achieve this aim place the least restriction on a particular right, or are the least restrictive means to achieve this particular outcome. Sanchez’s analysis of French law explains in particular that proportionality is meant to focus predominantly on outcomes, not on merely determining whether there is sufficient evidence on which to ground the administration’s assessment of proportionality. He uses this analysis to criticise some of the recent decisions of the Conseil d’Etat scrutinising the imposition of restrictions on civil liberties under the emergency powers provisions enacted following the terrorist attacks in Paris in November 2015. The Conseil Constitutionnel, when determining the constitutionality of the emergency powers provisions, required that any decision to place an individual under house arrest had to be verified by the administrative judges, focusing in particular on ensuring that these decisions were ‘adapted, necessary and proportionate’. Nevertheless, decisions of the Conseil d’Etat scrutinising house arrests did not refer to these provisions. However, the Conseil d’Etat was prepared to strike down a decision to place an individual under house arrest when the evidence on which this decision was based was not substantiated in fact. For Sanchez this fails to provide a proportionality control, given its focus on facts and evidence, as opposed to performing a substantive review as to whether placing an individual under house arrest is a proportionate response to the perceived threat to public security caused by this particular individual. Whether we see this as a failure of the principle of proportionality or not, it is clear that there is evidence of a more procedural turn in other jurisdictions. ­Advocate General Juliane Kokott and Christoph Sobotta’s analysis of EU law also points out the connection between factual assessments and the application of ­proportionality. As is well known, proportionality is a general principle in EU law,

138 

Alison L Young and Gráinne de Búrca

applying to all areas of EU law. Kokott and Sobotta’s analysis recognises a ­connection between the extent to which the CJEU applies a more substantive as opposed to procedural review and the area of law under review. As a general principle, proportionality in EU law applies to ensure that an act or measure does not exceed the limits of what is appropriate or necessary to achieve a particular objective, ensuring, where there is a range of means through which an objective can be pursued, that the EU adopts the least onerous means of achieving a particular objective. As a general principle, proportionality can apply to a range of situations. When applying proportionality, they argue, the CJEU performs a procedural as well as a substantive check. The Court ensures that the body under review provides an adequate statement of the reasons justifying its particular choice of means to achieve a certain policy objective. It also scrutinises decisions to ensure that the evidence on which the EU institution relies is factually accurate, reliable and consistent, and that it can substantiate the conclusions drawn by the body in question. Kokott and Sobotta suggest that the Court has been more prepared to carry out a full substantive review when faced with the application of proportionality to human rights, as in the recent Schecke4 and Digital Rights Ireland decisions.5 However, rather than seeing a procedural approach as a potential failure to apply the principle of proportionality, Kokott and Sobotta note that the CJEU applies a more procedural analysis when the subject matter to which the proportionality test applies is outside of the field of rights, to compensate for the less stringent application of proportionality over the outcome of the decision in these areas. This assessment leads nicely to our second assumption: is proportionality automatically a more stringent form of review than other forms of rationality control or reasonableness review? It clearly has been perceived as such across the legal systems studied in this section. Martinico and Simoncini note that, in Italian law, proportionality control arose out of reasonableness controls, but is generally regarded as going further than the standard of reasonableness. Whereas proportionality focuses on fairness and necessity, reasonableness focuses on suitability, which is generally a less demanding standard of review. This results in proportionality review being regarded as a more stringent form of control. Helle Krunke’s analysis of proportionality in Danish law notes not only that proportionality is generally regarded as a more stringent form of control, but remarks that the level of stringency found in the application of proportionality in EU law and the ECHR sits uneasily with the general attitude of judicial restraint in Denmark. This creates an element of ‘double standards’ and gives rise to reverse discrimination, where individuals whose circumstances fall within the scope of EU law may find that decisions governing them are subject to a more detailed scrutiny, including analysis of whether the administration has utilised the least restrictive means to achieve a p ­ articular o ­ bjective, 4  Case C-92/09 Volker und Marcus Schecke GbR v Land Hessen [2010] ECR I-11063, [2012] All ER (EC) 127. 5  Case C-293/12 Digital Rights Ireland v Minister for Communications, Marine and Natural Resources [2015] QB 127.

Proportionality

 139

or whether a measure is proportionate to the pursued purpose of administrative action, than the application of the principle of proportionality in Danish law. However, it is clear from a more detailed analysis of the chapters in this section that proportionality can be applied more or less stringently. It is not a monolithic or single-standard test, but is best understood as a variable standard of review. The theme of a variable standard of review runs through Craig’s analysis of English law and Philip Sales LJ’s account of the application of proportionality under the ECHR, with its well-known margin of appreciation. Sales argues that each of the aspects of the test of proportionality can be applied with varying degrees of stringency. This is recognised through the application of the margin of appreciation, which Sales regards as fulfilling three distinct purposes. First, it provides a means of recognising when weight should be given to democratic resolutions of rights, particularly in those areas that are politically sensitive and where there appears to be no general consensus across the signatory states to the Convention. Second, it serves the aim we associate more traditionally with the margin of appreciation, to allow each signatory state to develop its own culturally sensitive balancing of rights. Sales also advocates a new, further role for the margin of appreciation. He argues that the European Court of Human Rights has increasingly taken on the role of a superior constitutional court, scrutinising the decisions of national courts and national legislation when applying the proportionality test. His proposal for an application of the margin of appreciation in this application reflects concerns of what are usually referred to as comity, namely respect and consideration for the conclusions reached by other similarly legitimate and independent actors. All three roles or circumstances for an application of the margin of appreciation allow for a more or less stringent application of the proportionality test. Kokott and Sobotta’s analysis of EU law also recognises how the principle of proportionality may be applied more or less stringently, their focus being on how proportionality is applied less stringently in EU law in certain circumstances. In particular, the CJEU exercises restraint when it comes to the oversight of policy choices, or the evaluation of complex technological issues. When dealing with issues which might be described in this way, the Court has modified the standard of proportionality, striking down measures only when they are manifestly disproportionate. As discussed above, this weaker substantive control is, however, often coupled with a stronger procedural control, entailing a detailed scrutiny of reasons and evidence. The CJEU, however, applies proportionality more stringently when faced with the judicial review of measures affecting human rights. This modification of the stringency with which proportionality is applied is also remarked on in Giuseppe Martinico and Marta Simoncini’s account of the test of proportionality in EU law, where they note the oscillations of the proportionality test, and the different meanings it is given in different contexts of EU law. They also remark on the fact that, although the European Union has a three-part test of proportionality, the Court of Justice does not always or even often reach the third stage—the stage of proportionality stricto sensu, where the Court balances the aim of the policy against the burdens that the policy may produce.

140 

Alison L Young and Gráinne de Búrca

Third, proportionality is often regarded as a structured test, where the courts focus on whether the policy proposed by a public body has a legitimate aim, whether the measure taken is suited to achieving a particular aim, whether the measure is necessary to achieve this aim, and whether there is a proportionate balance between the aims pursued by the policy and the consequences of pursing this policy. In the field of human rights, the court is required to ensure that there is a fair balance between the impact on the right and the policy aim, or the impact on the rights of others. There is clear evidence of a structured test in the decisions of the European Court of Human Rights and in EU law, even given the evidence of Martinico and Simoncini that this three-stage test is not always applied. However, it is not the case that a structured approach to proportionality is always taken in domestic law outside the influence of the EU and the ECHR. Proportionality is still not a general standard of judicial review in English law and, as Craig’s work makes clear, although proportionability has a long history in English law, the test may have required courts to balance benefits and burdens but it did not require the imposition of a structured three- or four-stage test. Martinico and Simoncini note that, although the Italian Constitutional Court does apply the principle of proportionality, it absorbs the test into its control for reasonableness and does not use the three- or four-part test of proportionality. Krunke’s chapter illustrates how, although proportionality is used as a principle of interpretation in Denmark, it is not yet a constitutional principle given that it is not used in the pre-legislative scrutiny of legislation. Moreover, when applied in the administrative law and criminal law context, the Danish courts examine whether the least restrictive means is used, or whether the measure adopted is proportionate to achieve its particular purpose. However, Krunke’s chapter makes no reference to the use by the Danish Supreme Court of a clearly structured test. In French law, Sanchez argues that the courts may pay lip service to the test of proportionality—noting in particular the reference to the requirements that a measure be necessary, adapted to its purpose and proportionate in the Conseil d’Etat’s decision in Dieudonné. However, it is hard to find a clear focus on the use of these stages of the test in the decisions of the French courts. The comparative analysis provided in these chapters, therefore, casts doubt on some of the traditional narratives of the proportionality test. However, several clear and important themes run through the chapters. The principle of proportionality does appear to require a more searching review and the third part of the proportionality test, in particular, entails a more searching review than just an analysis of how an administrative decision was made, or the evidence on which it was based. However, the principle of proportionality is not always applied in a way that is more stringent than the review of reasonableness, especially when judicial review of reasonableness is modified to apply more stringently. The principle of proportionality may be applied by courts in particular jurisdictions more stringently in certain circumstances and less in others, but we can nonetheless perceive a general trend towards a more stringent application in the context of decisions affecting human rights and its less stringent application to the evaluation of policy

Proportionality

 141

choices, or of complex economic or technical decisions. The experience of France may be explained in part by the difficulty of applying the principle of proportionality in the national security context, with the focus on a detailed factual analysis appearing to be a more legitimate form of review over the decisions of a public body than a review which may require the court itself to evaluate substantive issues of national security policy. In addition, the more the test of proportionality is applied to issues of human rights; the more likely it is that the test will be applied in a more structured and rigorous form, ensuring that it also includes the final and perhaps most intrusive balancing stage.

III.  Is Proportionality Only or Best Suited to Rights Review? It is hard to deny the connection between proportionality and the protection of rights. This is particularly true as regards the influence of the ECHR and the extent to which proportionality in certain jurisdictions is either applied only, or is applied more stringently, in the field of human rights. This close connection between proportionality and rights review is clearly evident in English law, as documented well in Sales’s chapter which explains and evaluates the role of the ECHR in English law, in addition to the development of the protection of human rights through the common law. A similar connection can be found in Danish law, according to Krunke who notes the use by the courts of both the EU Charter of Fundamental Rights and Freedoms and the ECHR to provide a more stringent review of administrative powers through the use of proportionality. Both Sanchez’s account of proportionality in French law and Martinico and Simoncini’s account of Italian law also point to the way in which the use of proportionality review reflected a wish to exercise more stringent judicial control over police powers, with a view to ensuring that measures adopted to ‘protect public order’ should use the least restrictive means when limiting or impinging upon human rights, and particularly the right to freedom of expression. Moreover, as Kokott and Sabotta argue, although proportionality is a general principle of EU law, it is applied more stringently by the European Court of Justice in the field of human rights. However, it is also clear that there are other areas of the law that tend to attract the application of the principle of proportionality by courts. This is particularly true, for example, as concerns the issue of judicial scrutiny of sanctions. One of the key elements of the historic principle of proportionability discussed in Craig’s chapter was the need to ensure that sanctions are proportionate—it is unfair, for example, to punish all sailors equally for the depletion of the cargo of wine during its transit to England, such that no sailor was paid, when it may not be that all of the sailors drank the supplies, nor did those who partook of the wine necessarily drink the equivalent of their wages for the voyage. And although her chapter does

142 

Alison L Young and Gráinne de Búrca

not focus on sanctions per se, Krunke explains how proportionality in Danish law is a principle of criminal law as well as of administrative law, suggesting again that proportionality plays a role in ensuring fair punishment. Sanchez also remarks on how the principle of proportionality has been applied to sanctions cases in French law, pointing out in particular its application to prisoners and the military. Martinico and Simoncini remark on the connection between the principle of proportionality and of equality in Italian administrative law. This connection to equality suggests not only a connection between proportionality and sanctions, but also a possible role for proportionality in ensuring a fair distribution of benefits. This was a clear dimension of the application of the principle of proportionablity in English law, according to Craig’s analysis. In addition, concerns about a fair distribution of benefits and burdens is a common theme reflected in the understanding that the principle of proportionality requires that administrative bodies must select the least burdensome means of achieving a particular policy goal. This is evident from the analysis of Italian administrative law, in certain elements of French administrative law applying le bilan approach to proportionality, and in its more general application in EU law. These commonalities suggest that the principle of proportionality should not merely be seen as a legal test particularly suited to judicial review of rights given its tendency to be applied more stringently in cases involving impact on individual rights, not least because, as we have seen in Sanchez’s discussion of the recent cases in France following the state of emergency, proportionality can be applied by courts more or less stringently in cases involving adverse impact on individual rights just as it can when it is applied in other contexts. There would also appear to be some connection between the application of a proportionality principle in certain contexts and distributive justice requirements. Whilst it may not be the legitimate task of courts when applying a test of proportionality to ensure the equal distribution of all benefits and burdens resulting from all policy decisions, nevertheless the requirements of proportionality would appear to have clear implications when courts seek to ensure that policies do not excessively breach principles of distributive justice by imposing clearly disproportionate burdens or benefits on particular individuals or particular groups.

IV. Conclusion It is clear that proportionality is a multifaceted principle of administrative and constitutional law, having a role to play in the protection of rights and in ensuring distributive justice by challenging excessive inequalities, particularly in relation to burdens imposed but potentially also in relation to benefits conferred. Its flexible and multifaceted nature as a tool of judicial review is illustrated by the different origins and conceptions of proportionality across the legal systems discussed, in addition to the varied influence of the EU and ECHR legal systems. Whilst this

Proportionality

 143

may be seen as an advantage, it does not come without its disadvantages. As Sales sharply notes, the lack of certainty and clarity in the application of proportionality can have consequences for the rule of law. This danger is also remarked upon by Martinico and Simoncini. Moreover, the principle of proportionality, in challenging assumptions about the appropriateness of judicial restraint, may be perceived as having the potential to undermine democratic legitimacy should the judiciary stray too far into the evaluation of policy decisions. This has clearly been a concern in Danish law, where there is a tradition of judicial restraint, and also in English law as reflected in the development of greater deference and a less stringent application of the principle of proportionality, and in the development over time of the margin of appreciation by the European Court of Human Rights. Nevertheless, whilst pointing to the potential problems created by the application and spread of the principle of proportionality as a tool of judicial review, the chapters in this section also suggest some responses to these difficulties in the form of greater dialogue and exchange between judiciaries from different legal systems, in addition to greater academic exchange of ideas and critiques. The chapters in this section are hopefully one step in this process. It is clear that across Europe, and no doubt across other jurisdictions beyond Europe, we still have much to learn from one another about the scope, application and value of the principle of proportionality as a principle of judicial control.

144 

9 Proportionality and Judicial Review: A UK Historical Perspective PAUL CRAIG

I. Introduction Academic legal study is perforce temporally grounded. We most naturally consider the here and now. We may take a fleeting backward glimpse, most commonly to confirm our desired impression of the status quo, and to add substance to our prognostications, hesitant or not as they may be, about the future. Legal history is commonly left to legal historians in this age of increased specialisation. Their expertise is invaluable, but the historical terrain is vast, the numbers of specialist legal historians are limited and the subject matter coverage uneven. There are large gaps, as exemplified by the historical landscape of administrative law. Relatively little is known about legal development in this area, and the regulatory environment from which it grew, as attested to by the nostrum that the UK had no regime of administrative law until the mid-twentieth century, which is as erroneous as it is oft-repeated. We are of course not formally bound by anything that historical legal inquiry reveals. The academic has tools that bear analogy to the judicial. We can distinguish, praying in aid changed social circumstance, altered normative values and shifting policy preferences. Yet this academic ‘power’, like all other forms thereof, comes with responsibility. The argument must withstand scrutiny in the academic marketplace, and must also be grounded in some knowledge of what preceded the here and now. Insofar as this knowledge is exiguous it thereby diminishes the value of the academic judgement. This chapter is designed to cast historical light on the role played by the concept of proportionality in UK law. This very sentence may provoke scepticism, given the dominant view that proportionality was an import from continental Europe, the integration of which into UK law has been driven by membership of the European Convention on Human Rights (ECHR) and the European Union. There is no doubt that the classic three- or four-part proportionality inquiry has continental foundations, particularly Germanic, although it is noteworthy that the German eighteenth-century formulation of the principle was embryonic, with the

146 

Paul Craig

fully fledged test developing later.1 Integration of proportionality into UK law has been furthered by ECHR and EU jurisprudence. There is, however, also no doubt that the UK had a concept akin to that of proportionality, from the late sixteenth century onwards. The precise appellation varied, with terms such as proportionability, proportionable and disproportionate found in the legislation and case law. It was not a formal three-part test of the modern kind, but the older UK concept shared a common theme with its more modern offspring, which is the proscription of excessive regulatory burdens and the need to ensure that the burden was objectively justified. The story begins with four general misconceptions about UK administrative law, which sets the backdrop for the ensuing discussion. This is followed by four historical manifestations of proportionability. It will be seen that the concept was very commonly used as a criterion in regulatory legislation from the mid-­sixteenth century onwards, with the expectation that the courts would interpret and apply it to the particular regulatory regime, which is exactly what occurred. It was also used in the absence of express statutory mention, as a principle of legal interpretation, such that the courts would interpret regulatory legislation whenever possible so as not to impose excessive burdens on individuals. Proportionability was in addition deployed as a free-standing principle of judicial review, and also on some occasions as a condition for the legality of regulatory intervention. The magic number four frames the discourse in the final section of the chapter, where the focus shifts to the normative. There is discussion of the relationship between proportionality and proportionability, followed by reflections on the connection between the judicial and legislative use of proportionability. The final parts of this section consider the relevance of this material for modern doctrine, the determination of which is predicated on understanding that proportionability was judicially applied with sensitivity to regulatory purpose and appreciation of the limits to judicial oversight of the administration.

II.  UK Administrative Law: Four Misconceptions The image that we have of particular legal subjects is commonly constructed from a plethora of sources, real or imagined. This is especially true of UK administrative law, the image of which is based on four misconceptions. First, it is common for academics to believe that the UK had no system of administrative law until the mid-twentieth century, a sentiment fuelled by Dicey’s conclusion that the subject was not known in the UK.2 The reality is quite the contrary.

1  A Stone Sweet and J Mathews, ‘Proportionality Balancing and Global Constitutionalism’ (2008) 47 Columbia Journal of Transnational Law 73, 98–111. 2  AV Dicey, The Law of the Constitution, ed JW Allison (Oxford, Oxford University Press, 2013).

Proportionality and Judicial Review: A UK Historical Perspective

 147

The UK had one of the oldest bodies of administrative law in the post-Classical world, the doctrinal foundations of which were laid two centuries before anything comparable began to happen in continental Europe. It had a body of legal rules concerned directly with the legal constraints that should be placed on the administration broadly conceived for at least 400 years, and many of the core concepts that we use today would be recognised by our judicial forbears such as Coke, Holt, Hale, Abbott, Tenterden, Blackstone, Mansfield and Kenyon because they created them. In doctrinal terms, the courts developed many of the central concepts of judicial review with which we are familiar today.3 There was well-established case law on review of fact and law. There was doctrine dating back to the sixteenth century on the legal control of discretion, which was cast in terms of rationality review and also what was termed proportionability. There was jurisprudence on due process and damages liability. There was doctrine on principles of good administration, as exemplified by case law limiting the ability of a person who possessed a de facto or de jure monopoly to charge whatever prices he liked, the courts reasoning that such property was imbued with a public interest that limited the normal capacity to charge what the market would bear. The doctrine was given force through judicial creativity in relation to remedies. The number of such cases should, moreover, be borne firmly in mind. Public law cases concerned with direct and collateral challenge were at least 30 per cent of the total case law, and in some periods the ­figure was almost certainly higher. There were many thousands of such cases from the late sixteenth to the nineteenth centuries. And the incidence of such judicial review per year, when you take population into account (4.8 million at the beginning of the seventeenth century), and also the limited number of judges (King’s Bench had three judges), was no less than it was in the 1980s. A second assumption is that the courts applied the emerging principles of ­ judicial review with the sole objective of controlling the administration,­ betokening in this respect a narrow red-light conception of such review. This is mistaken. J­udicial review has always possessed a Janus-like quality. It is the mechanism through which the preceding doctrines were used when an individual contested the legality of a decision or regulatory norm made by a public or quasipublic body. This is the face that we perceive. Judicial review was, however, also the legal mechanism through which the courts commonly effectuated the regulatory schema challenged before them. The claimant challenged the legality of a decision and lost, because the court did not agree that there was such an illegality judged by the terms and purposes of the legislation. In reaching this conclusion the courts interpreted the statute to attain the specified objectives, and often filled gaps to render the legislation more efficacious. They were fully cognisant of the values served by the regulatory legislation. This duality is apparent when one reads the case law in any area where judicial review was applicable.4 3 P Craig, UK, EU and Global Administrative Law: Foundations and Challenges (Cambridge, ­Cambridge University Press, 2015) 29–44. 4  ibid, 62–95.

148 

Paul Craig

A third misunderstanding is that we did not need administrative law before the late nineteenth/twentieth century because there was not much governmental regulation, and thus no need for administrative law to render this accountable. This too is badly wrong as a matter of historical fact. England was highly centralised compared to its continental neighbours, more especially from the Tudors onwards. Social5 and economic legislation occupied a great deal of time in Elizabethan parliaments6 and ‘was considered, after the granting of taxation, to be the primary function of the House of Commons’.7 Adam Smith’s free market ideas were two centuries away, and there was statutory regulation of diverse matters, including trades such as leather, alcohol, iron and cloth; wages; bankruptcy; poverty, unemployment and vagrancy; land use; morality; police powers broadly conceived; tax and flood defences. The later advent of free market principles led to some diminution in trade regulation, but there was also increased regulation in areas such as factories, health and the like, which is the backdrop to continuing historical debates as to whether the nineteenth century really ever was an era of laissez-faire.8 It should moreover be emphasised that these areas were subject to detailed regulatory schemes laid down in enabling legislation, which was amended and fine-tuned repeatedly over the years. The fourth error, which is directly relevant to the present chapter, is the assumption that insofar as we exercised judicial control over discretion, that was done through reasonableness review and that proportionality was unknown in the UK prior to its introduction via EU and ECHR law. It has indeed been argued that it would be constitutionally improper for the courts to render this concept generally applicable as a tool of judicial review in the UK, this argument being predicated in part at least on the assumption that it is an alien import to be largely confined to the areas from which it is said to owe its origins, namely ECHR and EU law. This reasoning ignores history. The reality is that proportionality-type review existed in the UK from the seventeenth century onwards, and it was most commonly applied in non-rights-based cases. We did not have the classic three-part proportionality inquiry, and indeed if you search the legal database for the word proportionality you will get no hits. This is, however, because a range of different words were used in the context of judicial review actions, direct and indirect, including proportionable, proportionability, disproportion and proportionate. The semantic difference should not, however, conceal the substantive similarity: the courts were concerned to ensure that the regulatory burden placed on an individual was not excessive and that it was fair given the nature of the regulatory schema. The legislative and judicial manifestation of this will be explored in the next section. 5 

See eg P Slack, Poverty and Policy in Tudor and Stuart England (London, Longman, 1988). See eg GR Elton, The Parliament of England 1559–1581 (Cambridge, Cambridge University Press, 1986); DL Smith, The Stuart Parliaments 1603–1689 (London, Arnold, 1999). 7 R Sgroi, ‘Elizabethan Social and Economic Legislation’, www.historyofparliamentonline.org/ periods/tudors/elizabethan-social-and-economic-legislation. 8 See eg A Taylor, Laissez-faire and State Intervention in Nineteenth-Century Britain (London, ­Macmillan, 1972). 6 

Proportionality and Judicial Review: A UK Historical Perspective

 149

III.  Proportionability: Four Manifestations A.  Judicial Review and Statutory Provision Public lawyers, indeed lawyers in general, are considerably fonder of case law than legislation. Give them the choice and they will pick a complex case over a difficult statute pretty much every time. This is not the place to consider the rationales for this preference. Suffice it to say for the present that while the preference may be ingrained, we can learn much from the statutes that gave regulatory power to the administration and the terms on which this was done.9 The present inquiry is concerned with the extent to which such legislation conditioned the grant of regulatory power through requirements that burdens should be proportionate, or some word equivalent thereto. There were numerous statutes that explicitly contained ideas of due proportion in a public law context, which were then applied by the courts in judicial review actions, or in cases of indirect collateral challenge. Some sense of the range of legislation that conditioned intervention through such concepts can be conveyed from the numbers. From the late sixteenth century the term proportionably was used on 763 occasions in sections of statutes, while the term proportionable can be found in 1,230 statutory provisions.10 These citations used the relevant term at least once, and not infrequently more often. The concept was used in a plethora of regulatory contexts: economic, social, criminal and defence of the realm. Space precludes detailed elaboration, but the ensuing analysis will exemplify use of the concept in these areas, followed by three more particular examples of the way in which it was applied in relation to the poor law, bankruptcy and improvements. The economic usage of the concept is evident in the many statutes concerning tolls charged for roads, bridges, canals and rivers; the legislation commonly empowered collection of a specific sum for every ton of goods conveyed a certain distance, and this sum would then vary proportionably where the weight or distance was greater or lesser.11 Much the same was true in relation to many excise statutes enacted over four centuries; the legislation would prescribe a basic charge in relation to, for example, a barrel of beer, with smaller or larger amounts leading to proportionably less or more liability.12 Social purpose was to the fore in the provision for shipwrecked mariners; the legislation provided that ‘it is just and reasonable that all British Merchants

9  P Craig, ‘The Legitimacy of US Administrative Law and the Foundations of English Administrative Law: Setting the Historical Record Straight’, forthcoming. 10  There were also 1,209 legislative provisions that use the term proportioned, but many of these are less relevant to the present analysis. 11  See, e.g., Bedford Level Act 1756, 29 Geo 2, c. 9; Turnpike Roads Act 1822, 3 Geo. 4, c 126; Strick v Swansea Canal Company (1864) 16 CB (NS) 245. 12  See eg Excise Act 1670, 22 & 23 Cha 2, c 5; Taxation Act 1701, 13 & 14 Will 3, c 5.

150 

Paul Craig

t­ rading to Portugal, should proportionably contribute to the Relief of shipwrecked ­Mariners, and other distressed Persons, your Majesty’s Subjects there, and to the other pious and charitable Purposes herein after mentioned’.13 A variant on this theme is apparent in statutory provision for what was in effect a social insurance scheme for merchant seamen who became ill or died, with financial contributions made by ship-owners and the seamen themselves; the amount of the contribution was determined, inter alia, proportionably judged by length of time, and there was also provision for proportionate increase in the pension in certain circumstances.14 A different kind of social purpose was evident in legislation designed to prevent bribery and the like during elections; the regulatory legislation made provision for an election auditor to be recompensed by the candidates, including reasonable expenses incurred, which ‘shall be paid rateably and proportionably by the Candidates respectively’.15 The concept of proportional contribution was part of the changing world of criminal detection, as evidenced in the law relating to hue and cry; the malaise whereby the entire financial burden of recompensing a person who had been robbed was de facto imposed on a particular occupant of the locality where the robbery took place was ameliorated by legislation requiring that the cost be borne proportionably by each inhabitant, judged in terms of their ability to pay.16 This dimension is evident once again in statutes to deal with Moss Troopers, which was the vernacular for ‘lewd, disorderly and lawless Persons, being Thieves and ­Robbers’, who resided in Scotland or counties adjacent thereto, the name d ­ eriving from their ability to disappear into tracts of moss and heather to avoid ­capture; the cost of protection was borne by the counties, and unsurprisingly those ­further from the border were less inclined to contribute, being less affected by the ­incursions, and thus the legislation was designed to ensure that they paid their proportionable share towards defence.17 Proportionability was also part of the regulatory criteria concerning defence of the realm; thus where there was a shortfall in the number of private militia his Majesty’s Lieutenant was instructed to ‘discharge by Lot proportionably out of each respective Hundred, Rape, Lathe, Wapentake, or other Division, so many private Militia Men as shall exceed the Number so fixed and settled as aforesaid’.18 Mutiny and desertion endangered defence of the realm, and were the subject of successive statutes; the legislation commonly made provision for the billeting of soldiers, such that: Petty Constables, Headboroughs and Tithingmen, shall … billet and quarter every such Officer and Soldier in such Houses so subjected thereto by this Act equally and

13 

Shipwrecked Mariners Act 1721, 7 Geo 1, c 17. Merchant Seaman’s Widows Act 1834, 4 & 5 Will 4, c 52, ss 5–6. 15  Prevention of Corrupt Practices Act 1854, 17 & 18 Vict, c 102, s 34. 16  Hue and Cry Act 1584, 27 Eliz 1, c 13, s 5; Hue and Cry Act 1734, 8 Geo 2, c 16; 17  Moss Troopers Act 1662, 14 Cha 2, c 22. 18  Militia Act 1757, 31 Geo 2, c 26, s 21. 14 

Proportionality and Judicial Review: A UK Historical Perspective

 151

­ roportionably according to the Number of such Officers and Soldiers so to be billetted p and quartered, and of the Houses so subjected to receive them.19

Economic incentives for defence of the realm were evident in the law relating to prize, whereby enemy fortresses, ships and the like could be so treated; where, however, the venture was jointly undertaken by the navy and army, they shall ‘have such proportionable Interest and Property as His Majesty’ shall think fit.20 It is therefore readily apparent that proportionability was a common feature of much economic and social regulatory legislation, and that this was true also in relation to statutes where the objectives were police powers broadly conceived. The role played by the concept is evident in the following examples drawn from the poor law, bankruptcy and improvements. Provision of poor relief began in earnest with the Elizabethan poor law. The Poor Relief Act 160121 placed the primary obligation of support on the parish. This public statute was later complemented by many local Acts, whereby p ­ arishes joined together in order to discharge their responsibilities, since this was a more effective and efficient method of doing so.22 The legislation provided that the overseers of the poor should distribute the financial burden of the workhouse in due proportion as between the parishes. This was the approach in numerous statutes enacted after 1601,23 and was embodied in the Poor Law Amendment Act 1834, section 32.24 It gave the Poor Law Commissioners broad power to create or dissolve Poor Law Unions made up of separate parishes, and to make ­appropriate rules for their management, as they saw fit.25 Provided always, that in every such Case the said Commissioners shall and they are hereby required to ascertain the proportionate Value to every Parish of such Union of the Workhouses or other Property held or enjoyed by such Union for the Use of the Poor or Benefit of the Rate-payers therein, and also the proportionate Amount chargeable on every Parish in respect of all the Liabilities of such Union existing at the Time of such Dissolution or Alteration of the same, and the said Commissioners shall thereupon fix the Amount to be received, or paid or secured to be paid, by every Parish affected by such Alteration.

It was for the courts to decide what the due proportion was, and enforce it accordingly. This is exemplified by the Westmoreland Justices case.26 The legislation provided for legal redress where the rate assessed for particular parishes was not proportionate to the ratable value of the property contained therein. The ­claimants

19 

Mutiny Act 1813, 53 Geo 3, c 17, s 52. Manning of the Navy Act 1805, 45 Geo 3, c 72. Poor Relief Act 1601, 43 Eliz 1, c 2. 22  S Webb and B Webb, English Local Government: Statutory Authorities for Special Purposes ­(London, Longmans, 1922) 107–51. 23  See eg Relief of the Poor Act 1782, 22 Geo 3, c 83; Poor England Act 1815, 55 Geo 3, c 47. 24  Poor Law Amendment Act 1834, 4 & 5 Will 4, c 76. 25  ibid, s 32. 26  R v The Justices of Westmoreland (1829) 10 B & C 226. 20  21 

152 

Paul Craig

argued that there was a disproportion in the rates levied on certain individuals from the township of Shap, as compared to individuals from the parish of Bampton. The court acknowledged that such an action could in principle lie, but held that it failed on the facts because it was improperly pleaded. The statute provided that it was the parish or township that must be aggrieved by the disproportionate rate, and that it did not suffice in this regard for particular individuals from Shap to show that they paid a higher rate than those from Bampton. The same combination of statutory condition and judicial application is ­evident in relation to legislation concerning improvement of localities. It was common practice to vest regulatory power in local improvement commissioners, who would be given statutory authority to pave roads and improve local amenities.27 The costs thereof would commonly be borne by the local inhabitants, or the more particular section that benefited from the work. This practice was carried into effect through statutory provisions that each occupier should pay a proportionable share of the costs of the work, with the added incentive that if they did not do so the commissioners could levy distress to recover the relevant sums. It was for the courts to determine the meaning and application of such provisions. Thus in Cole28 Tindal CJ held that on the proper construction of the statute the landowners could not be called on to pay in anticipation of the work being undertaken, but that when it had been done, ‘the commissioners were to be in a condition to call upon the land owners for payment of their proportionable shares of the expenses incurred’, with the court adjudicating on disputes in this respect. A third area that exemplifies statutory conditions cast in terms of proportionability, which were then applied by the courts, is bankruptcy. There was, as we have seen, much economic and social regulation in the sixteenth and seventeenth centuries. Bankruptcy was a prominent concern, as attested to by the number of statutes dealing with the matter. Thus there were major pieces of legislation dealing with bankrupts in 1542, 1571, 1603, 1625, 1705, 1706, 1731, 1745, 1763, 1772, 1783, 1821 and 1822, with the law being consolidated in 1824.29 It was conceived very much as a public law problem, which could destabilise the economy, and it was regulated by Commissioners of Bankruptcy. The initial legislation was enacted in the reign of Henry VIII, and the preamble attests to the social ill to be combatted, due to the fact that:30 [D]ivers and sundry Persons craftily obtaining into their Hands great Substance of other Mens Goods, do suddenly flee to Parts unknown, or keep their Houses, not minding to pay or restore to any their Creditors, their Debts and Duties, but at their own Wills and Pleasures consume the Substance obtained by Credit of other Men, for their own ­Pleasure and delicate Living, against all Reason, Equity and good Conscience.31

27 

Webb and Webb (n 22) Cole v Green (1843) 6 Man & G 872. Bankruptcy (England) Act 1824, 5 Geo 4, c 98. 30  Statute of Bankrupts Act 1542, 34 & 35 Henry VIII, c 4. 31  Statute of Bankrupts Act 1542, 34 & 35 Henry VIII, c 4. 28  29 

Proportionality and Judicial Review: A UK Historical Perspective

 153

Elizabeth I instituted the regime of Commissioners to administer the bankruptcy legislation,32 and they were given extensive powers.33 Proportion and proportionability were aspects of the regulatory regime that were then policed by the courts if the issue should be contested. Thus, to take but one example, the Bankrupts Act 1720 sought to redress an ill in the pre-existing bankruptcy schema by providing that persons who took bills or bonds payable at some future date for goods that had been delivered, and the buyer then went bankrupt, could nonetheless be admitted to prove their bill or bond and ‘be intitled to a proportionable Part of the Bankrupt’s Estate, discounting such Securities after the Rate of 5 l. per Cent. for what they receive’.34

B.  Judicial Review and Statutory Interpretation Principles of statutory interpretation have a firm place in all legal areas, including administrative law, as exemplified by the principle of legality formulated in Simms, whereby Lord Hoffmann made clear that a statute that infringed fundamental rights would be interpreted narrowly and that Parliament would have to use express words, or something very close thereto, if it wished the limitation to have effect, thereby paying the political cost of doing so.35 Such interpretive principles are not a modern creation,36 although the particular ways in which they are deployed may change over time. It is therefore unsurprising that courts lent against the interpretation of a statute where it would place a disproportionate burden on a particular party. This is exemplified by the decision in the York Justices case.37 The applicants sought mandamus to compel the justices of the peace to reconsider a rate charged on their parish as its contribution to payment of the poor law for the wider area. They argued that the charge was excessive. The justices refused to reconsider the rate on the ground that they were not authorised by existing law to vary the fixed proportions of the county rates from the form in which they had existed for many years. Abbott CJ accepted the argument of the applicants’ counsel, who pointed to a statute that gave broad power to remedy county rates.38 Counsel argued that the necessity for a new rate could only be brought before the justices by appealing against the disproportion of the existing rate, and that the statutory wording was broad enough to accommodate challenge not only to new rates, but also to those

32 

Statute of Bankrupts Act 1571, 13 Eliz I, c 7. Craig (n 9) s 2.D.1. Bankrupts Act 1720, 7 Geo 1, c 31; Goddard v Vanderheyden (1771) 3 Wils KB 262. 35  R v Secretary of State for the Home Department, ex p Simms & O’Brien [2000] 2 AC 115; R (Morgan Grenfell & Co Ltd) v Special Commissioner of Income Tax [2003] 1 AC 563 R (Anufrijeva) v Secretary of State for the Home Department [2004] 1 AC 604; A v HM Treasury [2010] 2 AC 534. 36  J King, ‘The Pervasiveness of Polycentricity’ [2008] PL 401. 37  R v The Justices of the Peace for the City and County of the City of York (1824) 2 B & C 771. 38  County Rates Act 1815, 55 Geo 3, c 51, s 14. 33  34 

154 

Paul Craig

that existed previously, as where they had been made by fixed proportions. Abbott CJ accepted this argument, concluding that the legislation could be construed so as to accord a right of appeal, and that this interpretation cohered with the policy of the legislation. The disproportion in the parish rate was thus a reason to construe the statute so as to enable the justices to hear an appeal on this issue. The same principle is evident in Thompson.39 The plaintiff claimed for money withheld from his wages as a seaman. The sum had been deducted because some crew had broken into a cargo hold in which some fortified Madeira wine was held, and consumed 162 gallons on the voyage back to the UK. There was no proof that the plaintiff had taken part in this bibulous episode. Counsel for the ship owner sought to rely on a statute, which provided that each seaman who undertook the voyage was entitled to his wages, ‘provided always that there be no Plunderage, Embezzlement, or other unlawful Acts committed on the said Vessel’s Cargo or Stores’.40 He contended that if there was any such plunder, the right to the wage was ipso facto forfeited, and that ‘the terms of the proviso being clear and precise, cannot be done away with by construction’.41 Mansfield J was unpersuaded, pointing to the disproportionate impact of this interpretation. He noted the statutory provision set out above which counsel regarded as unequivocal, responding that ‘from these general loose words, the Defendant would contend that, if any good to the value of 5s be plundered or embezzled by A., or any unlawful act committed by A, B and every other sailor on board shall lose his whole wages’.42 This was unacceptable and therefore the statutory words should be construed in relation to each sailor respectively, with the consequence that there was no foundation for forfeiture of the plaintiff ’s wages, and Mansfield J doubted, moreover, whether any proportionable deduction in common with all other sailors was warranted.

C. Principle of Judicial Review: Excessive Burden and Denial of Benefit Judicial recognition of proportionability as a free-standing principle of judicial review dates back to the late sixteenth century, and the seminal decision in Rooke, given in the twilight of the Elizabethan age five years before James I took the throne.43 Commissioners of Sewers levied charges on one person for the repair of river banks, notwithstanding that numerous landowners benefited from the work. The Commissioners were given broad discretion as to who should be charged with the payment, although the Statute of Sewers 1531 indicated that all those in the relevant area could be liable to contribute.44 39 

Thompson v Collins (1805) 1 Bos & Pul (NR) 347. Desertion of Seamen Act 1797, 37 Geo 3, c 73, Sched A. 41  Thompson (n 39) 348. 42  ibid, 349. 43  (1598) 5 Co Rep 99b. 44  Statute of Sewers 1531, 23 Henry 8, c 5. 40 

Proportionality and Judicial Review: A UK Historical Perspective

 155

The court found the Commissioners’ action to be unlawful, since the ‘­commissioners ought to tax all who are in danger of being damaged by the not repairing equally, and not him who has the land next adjoining to the river only’.45 The reasoning strikes a remarkably modern chord. If the charge could be levied solely on the owner with land nearest the river, this might defeat the purpose of the statute ‘for perhaps the rage and force of the water might be so great, that the value of the land adjoining will not serve to make the banks’,46 and it thus followed that he who derived the benefit should share the burden.47 The statute required equality which ‘well agrees with the rule of equity’.48 The court had this to say about the discretion accorded to the Commissioners: [N]otwithstanding the words of the commission give authority to the commissioners to do according to their discretions, yet their proceedings ought to be limited and bound with the rule of reason and law. … For … discretion is a science or understanding to discern between falsity and truth, between wrong and right, between shadows and substance, between equity and colourable glosses and pretences, and not to do according to their wills and private affections.49

The link with proportionability and the difference between this concept and equality became evident in Hetley.50 The Commissioners of Sewers had satisfied a levy on a town by ordering the seizure and sale of the plaintiff ’s cattle, and then his imprisonment when he had the temerity to complain of this. Chief Justice Coke had no hesitation in ordering certiorari against the Commissioners, since their action was in direct conflict with the ruling in Rooke. He reiterated that ‘the commissioners of sewers cannot tax a whole township, but it ought to be done severally and proportionably to every inhabitant to himself ’,51 and that discretion was framed through law to attain justice. It was the court that framed the principles against which it was decided whether the regulatory burden on a particular individual was proportionable. The criterion used in the seventeenth century was to judge liability to pay in accord with the degree of danger to the particular property if an inundation should occur, and the degree of fault or not in the person undertaking the repair. The courts when applying such principles were mindful of the dangers of substituting judgment for that of the administration, and equally mindful to ensure that the overall regulatory purpose was not undermined.

45 

(1598) 5 Co Rep 99b. ibid, 100a. Qui sentit commodum sentire debet & onus. 48  (1598) 5 Co Rep 99b, 100a. 49  ibid, 100a. 50  Hetley v Boyer (1614) Cro Jac 336; The Case of the Isle of Ely (1610–11) 10 Co Rep 141a; R v Peters (1758) 1 Burr 568, 570; Kerrison v Sparrow (1815) 2 Ves Jun Supp 587; R v Commissioners of Sewers for Tower Hamlets Tower (1829) 9 B & C 517, 521; Soady v Wilson (1829) 3 Ad & E 248. 51  (1614) Cro Jac 336. 46  47 

156 

Paul Craig

This is apparent from Bow52 where the defendant challenged the need for river improvements in the form of a new sluice, and the amount charged to him by the Commissioners of Sewers. The defendant insisted that an old sluice sufficed and that the levy was unlawful because the amount had to be levied on each person proportionable to the damage that he might sustain. The Lord Chancellor was reluctant to reassess the need for the new sluice since if he did, ‘then the orders of commissioners of sewers, and of the sessions, would be made in this Court’.53 He nonetheless acknowledged that the mode of assessment was wrong, and that the particular lands should be assessed according to the danger to them. It was not, however, necessary to name the owners of such lands, since the Commissioners might not know them. Moreover, ‘if not naming the owners should make the assessment void, there would be an end of all assessments by commissioners of sewers’.54 The preceding principles are also apparent in Keighley,55 where the courts further refined their controls over discretion, and subtly gradated the obligations incumbent on those charged with river defences. Thus if a person was bound by prescription to repair a wall against the sea, which was broken through no fault of his by sudden and unusual flow of water, the Commissioners of Sewers should tax all according to the quantity of their land. If, however, the person charged with repair was at fault and the danger was not inevitable, the Commissioners could charge him to repair it. If the reality was that the danger became inevitable through his fault, those who suffered loss could have an action on the case against him. The principle of proportionability was not confined to cases concerning the Commissioners of Sewers. In Brownlow56 it was deployed in the context of regulatory burdens flowing from the Statute of Charitable Uses.57 A decree was made pursuant to the statute for the town of Market Raisen, and the monetary burden was placed solely on the defendant who owned land in the area. The defendant ‘insists to pay but his Proportion of the Money, there being several other Persons that have Lands in their Occupations chargeable with the said Charitable Use, yet the Plaintiff lays the whole Decree upon the Defendant’s Lands’.58 The court held that the whole charge must be initially borne by the defendant, but that thereafter there should be analysis of all lands liable to bear the charges and the ‘Commissioners to apportion each Party’s Payment with such proportionable Part of the Charges the Defendant hath been put to’.59 The principle was employed to determine the legality of fees for prisoners in Johnson.60 The treasurer of Hereford charged the city of Hereford in respect of

52 

Bow v Smith (1795) 9 Mod 94. ibid, 95. ibid, 95. 55  (1610–11) 10 Co Rep 139a; R v Commissioners of Sewers in Essex (1823) 2 Dowling & Ryland 700. 56  Villa de Market Raisen v Brownlow (1635–36) 1 Chan Rep 91. 57  Charitable Gifts Act 1601, 43 Eliz 1, c 4. 58  Brownlow (n 56) 92. 59  ibid, 92. 60  The Queen v Johnson (1839) 10 Ad & E 740. 53  54 

Proportionality and Judicial Review: A UK Historical Perspective

 157

prisoners that the city had sent to its jail. A dispute arose as to the amount that was due per prisoner, which was larger insofar as the city bore responsibility for the general expenses of the jail, and not merely the personal expenses of each prisoner. The treasurer levied the charges pursuant to the Municipal Corporations Act 1835.61 Lord Denman CJ held that the larger amount of the weekly charge per prisoner, which was made ‘upon a calculation of the proportion which the expence of each prisoner, bears to the total expences of the gaol, is reasonably made, and ought to be paid, seeing that the borough has a proportionable share of the benefits arising from the whole establishment’.62 The concept was applied in relation to bankruptcy, irrespective of whether there was specific provision to that effect in a bankruptcy statute. Thus in Vanacre63 a debtor was indebted to B and C, the former in a very large sum, the latter in a smaller amount. B did not realise that the debtor had become bankrupt, but C did and sued out a Commission of Bankruptcy. C sought to recover the value of some of his goods from B. The court held that the creditors should come to an account, and have proportionable satisfaction from the estate. The whole debt and estate should be accounted for, and there should be a proportionable division of the money recovered. In the similarly named but distinct case of Vanaker64 the plaintiff succeeded in an action brought against Commissioners of Bankruptcy, whereby he sought to be admitted as a creditor in order to obtain his proportionable benefit from the estate. It is readily apparent from the preceding discussion that proportionability could be pleaded to avoid excessive burden and to obtain benefit that was properly due. This duality is evident in Walton,65 where the dispute arose out of enclosure. The facts are complex, but the case in essence concerned enclosure and the extent to which allotments granted pursuant to the enclosure legislation constituted compensation for the loss of access to common land. Lord Ellenborough CJ concluded that the plaintiff had two distinct rights of common, that the allotment granted pursuant to the enclosure legislation only related to one right of common, with the consequence that he was entitled to proportionable compensation in relation to the loss of the second right of common.

D. Principle of Judicial Review: Condition of Regulatory Intervention Proportionability was also used in a different way, to determine the legitimacy of regulatory intervention. This is exemplified by the case of Customs, Subsidies and

61 

Municipal Corporations Act 1835, 5 & 6 Will 4, c 76, s 114. Johnson (n 60) 753. Vanacre’s Case (1677) 1 Chan Cas 303. 64  Vanaker v Nash (1673) Rep Temp Finch 60. 65  Hollinshead v Walton (1806) 7 East 485. 62  63 

158 

Paul Craig

Impositions.66 The court held that the King could not at his pleasure place any imposition on merchandise imported or exported, unless it was for advancement of trade. While it was open to the King to prohibit a person with some commodities to leave the realm, this was only where the end sought was public rather than private. The King could also impose tolls for the repair of highways and bridges for the benefit of the subjects, but ‘the sum imposed ought to be proportionable to the benefit’. Thus in this instance proportionability was used to limit the circumstances where the King could impose certain charges, ie tolls could be imposed provided that the sum charged was proportionable to the benefit to the King’s subjects. We see this same principle at work in Hill,67 where the defence against an action for trover and conversion was that the defendant was a bellman, charged with cleaning the streets and the like, for which as a matter of custom he was entitled to take a certain portion of corn from every amount brought into the town. In resolving the case the court drew on Coke CJ’s reasoning, to the effect that while the King could grant privileges of murage and pontage for the building and repair of town walls, the tolls thus charged were conditional on the privileges being pro bono publico, and proportionable. This was apposite in the instant case, because this principle was used to help justify the custom whereby the bellman took a share of corn from traders, the reasoning being that because the bellman had performed a valuable public service, therefore his share was a legitimate, proportionable quid pro quo. The same theme is apparent in Stamford,68 where Lord Alexander CB referred in the course of his judgment to the principle that the King could not grant a right to levy a toll for holding of a market or fair that had hitherto been free, unless there was some quid pro quo, some proportionable benefit to the public.

IV.  Normative Dimension: Four Issues The inquiry thus far has considered proportionability from a positive law perspective. A number of related, albeit distinct, normative issues will now be considered.

A.  Proportionability and Proportionality: The Relationship It is important to be clear about the relationship between proportionability and more modern conceptions of proportionality. I make no claim that the concept of

66 

(1607) 12 Co Rep 33. Hill v Hanks (1792) 2 Bulst 201. 68  Corporation of Stamford v Pawlett (1830) 1 C & J 57, 73. 67 

Proportionality and Judicial Review: A UK Historical Perspective

 159

proportionable burden or that of proportionability embodied the three-part test associated with the modern conception of proportionality. There is nonetheless a linkage between the older and more modern conceptions, insofar as both have in common the idea that regulatory burdens should not be excessive, and that they should be objectively justified. Parliament could determine that contributions should be predicated on relative capacity to pay, combined with assessment of the benefit received from the regulatory activity. The statutory provisions provided guidance as to the distribution of benefit and burden in the respective areas, but there was nonetheless often room for contestation as to what ‘proportionable’ meant in the circumstances. This was decided by the courts, as exemplified by difficult judicial determinations as to what the concept should mean for liability to contribute to the rates levied for poor relief, when a river or railway ran through several parishes. We shall return to the case law interpreting these provisions below. Suffice it to say for the present that the courts did not substitute judgment, but showed respect for the administrative determination,69 and interpreted the legislation with a keen eye on its overall purpose.70 The judicial contribution was a fortiori more significant where there was no direct guidance from the legislation concerning the distribution of benefit and burden, although they did not substitute judgment concerning the administrative task in this type of case either. It can be acknowledged that much of the historical material was concerned with regulatory burdens that were in some sense quantifiable in monetary terms, which stands in contrast to some modern case law on proportionality. The degree of difference in this respect should nonetheless be kept in perspective. I do not subscribe to the belief that every clash of interests can be reduced to monetary terms. This is both reductionist and normatively problematic. We should, however, be cautious about the converse proposition: we should not think that merely because the appropriate balance could be expressed in monetary terms, therefore this was somehow easy, or straightforward. In some instances it was, but in many it was not, since the quantifiable conclusion was the result of a complex calculus into which a plethora of values were fed. This was so in relation to many of the instances set out above where judgment as to proportionability turned on an admixture of capacity to pay, stake broadly conceived in the relevant enterprise, and relative benefit or burden. The legislation might provide some guidance on this, and it might not. It was for the courts to apply the legislation, fleshing out the criteria contained therein, or developing them where the legislation was exiguous in this respect. The judicial role was perforce further enhanced where there was no mention of the term in the enabling legislation. There is therefore no ready equation between a conclusion being capable of being expressed in quantifiable terms, and the ease of the calculation that led to

69  70 

See below, section IV(D). For more detailed consideration of this, see Craig (n 3) 69–95.

160 

Paul Craig

the conclusion. There is also no basis for the assumption that cases in which the conclusion cannot be expressed in quantifiable terms will necessarily entail consideration of a more difficult range of values than cases where the result can be so expressed.

B.  Proportionability: Legislative and Judicial It is equally important to be mindful of the linkage between statutory embodiment of proportionability as a criterion in the regulatory schema, and the courts’ application of the concept as a free-standing principle of judicial review. The courts created the architecture of judicial review, fashioning the relevant principles from the rule of law. These included the concept of proportionability, which was introduced in the late sixteenth and early seventeenth centuries. Its application in the leading cases made good sense in normative terms, and helped to legitimate and render accountable the considerable power accorded to the administration. There is no evidence of parliamentary dissatisfaction with this case law. To the contrary, the very fact that Parliament enacted very many statutes that conditioned regulatory intervention on proportionability is indicative of the fact that the legislature was mindful of the concept’s utility, and it did so knowing full well that the courts would adjudicate on its meaning and application. This in turn would have encouraged courts to regard intervention aimed at denying excessive burden, or ensuring proportionable benefit, as warranted and normal.

C. Proportionability and Proportionality: Democracy and the Temporal Dimension It has been argued that whatsoever the history might be, it is of limited relevance in the modern day, given that the principles of judicial review must now be seen against the backdrop of a democratic legislature, with the legitimacy that is attendant thereon, which it never possessed hitherto.71 There are several points to bear in mind in this regard. First, it assumes that Parliament lacked legitimacy prior to the modern extension of the franchise when women were granted the vote in 1928.72 There is no doubt that modern parliaments elected on this extended franchise enjoy increased democratic legitimacy. To reason from this to the conclusion that parliaments hitherto lacked legitimacy is, however, a non sequitur. The determination of who should be entitled to vote has perforce changed very markedly over time. There is no doubt that to modern eyes the property qualifications that conditioned the

71 

72 

This argument was advanced by Lord Justice Sales when this paper was delivered at the conference. Representation of the People Act 1928.

Proportionality and Judicial Review: A UK Historical Perspective

 161

franchise in earlier centuries appear outdated and unwarranted. This does not mean that the parliaments elected in previous centuries were perceived as illegitimate. There was, to be sure, some pressure for extension of the franchise, most notably in the more eloquent writings of the Levellers, but there was also considerable support for the limits that then existed. The courts did not, moreover, regard Parliament as illegitimate because of its limited franchise. To judge the legitimacy of earlier institutions by modern standards is to commit the inter-temporal error of assuming that criteria currently regarded as the hallmark of legitimacy were always thus perceived. Second, the argument assumes a clear relationship between the principles of judicial review and the democratic mandate. This is not the place to engage in a detailed exegesis on this topic, which would take us far beyond the confines of this chapter. Suffice it to say for the present that the reality concerning this relationship is complex. The argument assumes that principles of review developed in earlier years, such as proportionability, may be unwarranted now that Parliament has assumed its modern democratic form. This contention conceals more than it reveals. The legal reality is that many principles of judicial review have been enlarged since the advent of the current franchise. This includes review for error of law, fact, equality, legitimate expectation and human rights. There is no reason why the current democratic mandate should be regarded as a reason for denying proportionality a role in judicial review, although it may have implications for the intensity with which it should be applied. This is, however, already taken into account, as attested to by the vibrant judicial and academic debates concerning proportionality, deference, respect and the like that abound in the modern case law and secondary literature.73 In my view such respect is warranted on institutional, epistemic and constitutional grounds.74 The debate concerning proportionality as a general principle of judicial review is, moreover, predicated on variable intensity proportionality review, with lower intensity review being applicable in cases that 73  See eg R Edwards, ‘Judicial Deference and the Human Rights Act’ (2002) 65 MLR 859; J Jowell, ‘Judicial Deference and Human Rights: A Question of Competence’ in P Craig and R Rawlings (eds), Law and Administration in Europe, Essays in Honour of Carol Harlow (Oxford, Oxford University Press, 2003) ch 4; J Jowell, ‘Judicial Deference: Servility, Civility or Institutional Capacity’ [2003] PL 592; M Hunt, ‘Sovereignty’s Blight: Why Contemporary Public Law Needs the Concept of “Due Deference”’ in N Bamforth and P Leyland (eds), Public Law in a Multi-Layered Constitution (Oxford, Oxford University Press, 2003) ch 13; Lord Steyn, ‘Deference: A Tangled Story’ [2005] PL 346; TRS Allan, ‘Human Rights and Judicial Review: A Critique of “Due Deference”’ [2006] CLJ 671; Lord Justice Dyson, ‘Some Thoughts on Judicial Deference’ [2006] Judicial Review 103; R Clayton, ‘Principles for Judicial Deference’ [2006] Judicial Review 109; J King, ‘Institutional Approaches to Judicial Restraint’ (2008) 28 OJLS 409; A Young, ‘In Defence of Due Deference’ (2009) 72 MLR 554; A Kavanagh, ‘Defending Deference in Public Law and Constitutional Theory’ (2010) 126 LQR 222; T Hickman, Public Law after the Human Rights Act (Oxford, Hart Publishing, 2010) ch 5; TRS Allan, ‘Judicial Deference and Judicial Review: Legal Doctrine and Legal Theory’ (2011) 127 LQR 96; A Brady, Proportionality and Deference under the Human Rights Act 1998: An Institutionally Sensitive Approach (Cambridge, Cambridge University Press, 2012); P Daly, A Theory of Deference in Administrative Law: Basis, Application and Scope ­(Cambridge, Cambridge University Press, 2012); A Young, ‘Will You, Won’t You, Will You Join the Deference Dance? (2014) 34 OJLS 1. 74  Craig (n 3) ch 2.

162 

Paul Craig

do not involve rights. This coheres with the historical jurisprudence in this area, in which, as made clear above, the courts did not substitute judgment for that of the administrator. Thirdly, it is important to avoid two related errors concerning the earlier jurisprudence. Courts did not, as we shall see in the next section, simply substitute judgment in relation to proportionability, and they did not ignore the regulatory objectives when adjudicating on its application. Nor did courts regard what they were doing in proportionability cases as akin to Coke CJ’s strictures in Dr ­Bonham’s case concerning review of legislation against natural right and ­reason.75 Coke CJ’s ideas in this respect did not develop for a plethora of reasons, but proportionability review was very different, and the two issues were regarded as distinct. The principal target of proportionability was not the primary legislation, but the administration thereof by Commissioners, local justices and the like, and the substantive criteria was not natural right and reason, but excessive burden.

D.  Proportionability: Intensity and Legislative Purpose A willingness to engage with the older case law reveals much about judicial reasoning and serves to dispel beliefs as to how the courts applied these controls on discretionary power. There is often an implicit assumption that modern sensibilities concerning the limits of judicial power were lost on our forebears, and that they were insensible to attainment of legislative purpose. There is no warrant for such assumptions, whether viewed as general conjectures, or as hypotheses concerning the application of proportionability. We, the modern judicial and academic generation, did not suddenly have insights as to the relationship between courts and Parliament that were unknown to our predecessors. The very fact that Parliament and its administration were perceived as legitimate meant that while the courts regarded controls on administration as necessary to ensure its accountability, they were also mindful of the proper limits to judicial intervention. This is readily apparent from cases such as Bow76 and Keighley77 considered above. It is evident also in the following examples, drawn from different subject matter areas. The Cape of Good Hope case concerned a claim by the East India Company to a proportionable share of the benefits resulting from the capture of the Cape.78 The ships were used to carry troops to the Cape of Good Hope. Most of the operations necessary to subdue the colony had, however, been performed before the arrival of these ships, although one ship did perform military service, and it was accepted that it should be allowed to share in the prize. The case turned on whether the other ships could partake of a proportionable share, the legislation stipulating

75 

Dr Bonham’s Case (1609) 8 Co Rep 113b. Bow (n 52). 77  Keighley (n 55). 78  The Cape of Good Hope and its Dependencies (1799) 2 C Rob 274. 76 

Proportionality and Judicial Review: A UK Historical Perspective

 163

that in joint operations between navy and army the men would be given such ­proportionable share of the property as directed by His Majesty. Sir Walter Scott, giving judgment, proceeded on the assumption that the ships were originally no more than transport vessels, ‘liable to be called upon occasionally to act, with alacrity and vigour (for British vessels, of any character, are liable to be so called upon on extraordinary occasions of public necessity)’,79 albeit acknowledging the possibility that ‘a military character might be afterwards impressed upon them, by the nature and course of their subsequent employment’.80 He was, however, sceptical as to whether the requisite military character could be found in the claim that the very appearance of the ships intimidated the enemy, thus entitling the East India Company to a proportionable share of the property. This would, said the judge, lead to unacceptable consequences, since any intimidation was entirely passive in the instant case, given that the ships had no knowledge that they had been seen and no animus to intimidate. It was very different where the non-commissioned ship contributed materially to the act of capture. The judge was, moreover, mindful of the dangers of broadening the circumstances in which the private vessel could obtain some proportionable share of the prize. The interest of the prize is given to the fleet and army, and it would not be the mere voluntary interposition of a privateer that would entitle her to share. It would be a very inconvenient doctrine, that private ships of war, by watching an opportunity, and ­intruding themselves into an expedition, which the public authority had in no degree committed to them, should be at liberty to say, ‘we will co-operate’; and that they should be permitted to derive an interest from such a spontaneous act, to the disadvantage of those to whom the service was originally entrusted. Expeditions of this kind, designed by the immediate authority of the state, belong exclusively to its own instruments, whom it has selected for the purpose; and it might be attended with very grave obstruction to the public service of the country, if private individuals could intrude themselves into such undertakings, uninvited and under colour of their letter of marque.81

The same sound sense in the interpretation of proportionability is evident in Page.82 The case concerned the meaning to be accorded to the concept in the context of legislation imposing tolls, and in relation to liability to contribute to poor relief in a particular parish. A statute provided for payment of tolls on the river Kennet from Reading to Newbury, specifying that this should not be greater than four shillings per ton; that the charge should be proportionably larger or smaller, where the weight was greater or less; and that it should vary also depending on the distance for which the goods were carried. The case was, however, concerned not simply with the proportionable tolls for the river, but also with the proportion of the tolls thus raised that should be rateable for poor relief in a particular parish, which was relevant given that the river flowed through several parishes. 79 

ibid, 281–82. ibid, 282. 81  ibid, 287–88. Italics in the original. 82  R v Page (1792) 4 TR 543. 80 

164 

Paul Craig

Lord Kenyon CJ considered the principles on which a proportionable share of the poor rates should be distributed between the respective parishes. It was argued on grounds of ‘policy and justice’83 that the tolls should be considered to be due in each parish in respect of the quantity of land occupied by the navigation. Lord Kenyon rejected this criterion, stating that ‘hard would be the lot of the officers who are to make the rates in these several parishes’, since ‘they would have to measure not only the length but the breadth of the navigation in each respective parish, and to ascertain with precision the exact quantity of land covered with water’, and these ‘difficulties would be insuperable’, such that it would ‘be in vain to think of rating at all, if such were the rule’.84 The correct principle was rather that ‘where a person has a valuable interest in any parish or township, he ought to contribute towards the relief of the poor in that parish in proportion to such valuable interest’.85 This in turn generated interesting legal inquiry as to where the property interest in the collection of the toll vested. The claimant was rated only in respect of tolls that became due at Newbury—the place where the navigation finished and where the goods were delivered. The court held that this was valid, notwithstanding the fact that the statute allowed the proprietor of the toll to appoint the place of collection, since if this were the criterion for liability for poor rates the toll owner ‘might appoint a place of collection not in any parish through which the navigation passes’,86 or ‘fix a place of collection in some parish where the poor rates are the lightest, which could not be within the meaning of the Act’.87 Judicial sensibility to policy objectives is also apparent in Wilson.88 The corporation of Carlisle brought an action in assumpsit for tolls due from the defendant for the passage of their coaches and carriages, which were loaded with goods as well as passengers. The defendant claimed to be exempt, arguing that their carriages were principally used to convey passengers, not goods, and that the customary right to the toll was applicable only to the latter not the former. The court disagreed. Lord Ellenborough CJ held that the custom was to pay a toll for goods conveyed in carriages in proportion to the number of horses, the rationale being, as Lawrence J noted, that the damage to the roads was proportionably greater depending on the number of horses used, thereby justifying the higher toll. The type of carriage was therefore irrelevant, provided that it was used for carrying goods for sale the custom attached to it. The defendant contended that the toll charged on carriages that carried people as well as goods was disproportionately high. Lord Ellenborough CJ rejected the argument: ‘[A]s to the d ­ isproportion stated to arise from the application of the toll to carriages of this description,

83 

ibid, 547. ibid, 547. 85  ibid, 547. 86  ibid, 547. 87  ibid, 550. 88  Mayor of Carlisle v Wilson (1804) 5 East 2. 84 

Proportionality and Judicial Review: A UK Historical Perspective

 165

where the number of horses is adapted more to the carriage of ­passengers than of goods, that is the party’s own act, of which he cannot complain.’89 It would not have been practicable for the corporation of Carlisle to determine how much weight was attributable to passengers, and how much to goods: if the ‘coach-owners will multiply the number of their horses because of the additional weight of passengers which they carry, together with goods for which the toll is payable in proportion to such number, it is their own act, and the corporation have no means of ascertaining the proportion of horses used for each’.90

V. Conclusion There will be no attempt to summarise the entirety of the preceding argument. Suffice it to say the following. Common lawyers rightly note the absence of the divide between public and private law of the kind that exists in civil law systems. Historically we did not have a separate regime of public and private law courts commonly found in civil law regimes, and that remains true in the UK notwithstanding the creation of a more specialist administrative law jurisdiction within the High Court. The absence of separate courts for public law and private law cases did not, however, mean that the substantive principles applicable to the respective areas were the same. The nature of the subject matter, combined with the distinctive nature of the legal issues that arose in public and private law cases, led to doctrinal rules that differed. There were distinct concepts, or different conceptions of the same concept. We should nonetheless not veer too far in this direction, and assume that there was no overlap in the concepts used, or that there was necessarily significant divergence in the particular conception deployed. The commonality between the concepts of responsibility imposed on trustees, fiduciaries and public officials is merely one powerful exemplification of this point. This is not the place for exegesis on the reasons that drove this commonality or overlap, although it was at base underpinned by a similar normative view as to the nature of responsibility that pertained in the respective areas. This is salient in the present context. This chapter focused on the application of proportionability and related terminology in regulatory contexts, where the claimant sought relief against a body that was public, charged with carrying out its regulatory tasks by legislation. It is noteworthy that the concept was also used in a plethora of other contexts, including, inter alia, employment law, debt, commercial law, succession and private law aspects of bankruptcy. Those minded to

89  90 

ibid, 7. ibid, 7.

166 

Paul Craig

try to dismiss the analysis in this chapter on whatever ground should bear this in mind. The reality was that the courts applied the concept in those areas where it was felt to be normatively warranted, without any fixed boundaries between public and private law. It was not alien to the common law, and it was not confined to some narrow area of legal doctrine. I make no claim that legal doctrine ipso facto warrants application now, merely because of its historical provenance. There is, however, a fortiori no basis for predicating conclusions about present application of a legal concept on ignorance of legal history, or of the way in which the concept was applied, that do not cohere with reality.

10 The Evolution of the Principle of Proportionality in EU Law—Towards an Anticipative Understanding? JULIANE KOKOTT AND CHRISTOPH SOBOTTA*

I. Introduction No doubt there is a general principle of proportionality in EU law. The question, however, is whether it is applied in a coherent fashion. To examine this, there is a temptation to present an endless list of cases, starting with poor Mr Stauder in 19691 and probably not ending with Mr Lanigan2 from July 2015. To avoid that, we will focus on a more specific issue, namely the level of scrutiny applied by the Court of Justice, in particular with regard to legislation. This area is of particular importance for the practical effect of the principle. Moreover, there have been some interesting developments over time and in particular during the last years. These could generate an understanding of the standard of scrutiny that we would describe as ‘anticipative’. In the following, we will therefore in section II present the basic structure of the principle. Section III describes how the Court traditionally exercises judicial control and self-restraint in this regard. In section IV we will discuss some requirements that the Court underlines more and more often to compensate for this restraint and reluctance to thoroughly scrutinise proportionality. And finally, in section V, we will highlight some very recent developments, namely that the Court is, perhaps, starting to develop a substantial test to determine the level of scrutiny that needs to be applied.

*  Respectively Advocate General and Legal Secretary (référendaire), Court of Justice of the European Union, Luxembourg. This paper is a revised speech given by Advocate General Kokott at the conference ‘General Principles of Law: European and Comparative Perspectives’, Institute of European and Comparative Law, University of Oxford, 25 September 2015. 1  Judgment in Case C‑29/69 Stauder v Stadt Ulm, ECLI:EU:C:1969:57. 2 Judgment in Case C‑237/15 PPU Minister for Justice and Equality v Francis Lanigan, ECLI:EU:C:2015:474.

168 

Juliane Kokott and Christoph Sobotta

II.  The Basic Structure of the Principle of Proportionality The principle of proportionality is one of the general principles of EU law. It means that an act or measure must not exceed the limits of what is appropriate and necessary to attain the objectives legitimately pursued. When there is a choice between several appropriate measures, recourse must be had to the least onerous, and the disadvantages caused must not be disproportionate to the aims pursued.3 Therefore, a four-step test should be applied. For the first step the objective of the particular measure must be identified and the question whether this objective is legitimate needs to be considered. With the second step it must be assessed whether the measure in question can contribute to this objective. The third step requires an examination whether the measure is the least onerous to achieve the aim pursued. Finally, with the fourth step a balance must be struck between the impact of the measure and its contribution to the objective. If a measure fails at any of these steps it is disproportionate. From a German perspective this understanding really misses an important element. German lawyers primarily employ proportionality not in isolation, but to evaluate whether restrictions of specific rights are justified. They are guided by experience in the area of fundamental rights where proportionality is one of the conditions for restrictions. This approach is reflected in Article 52 of the EU Charter of Fundamental Rights and in the European Convention on Human Rights. For an assessment of proportionality we therefore need to identify a specific right that is being restricted. Both the right and the restriction provide important elements for the subsequent steps to take. In particular the final balancing can be performed more precisely if it is focused on a specific impact on a right and not abstractly on ‘disadvantages’. It is likely that the Court’s practice came into being because it is not a specialised court for the protection of fundamental rights, but a general court with a traditional focus on economic regulation. Nevertheless, even according to the version adopted by the Court, proportionality could be a very steep hurdle to pass if each of the four steps were subject to strict scrutiny. However, is it really appropriate that a court always applies strict scrutiny to all the steps of the proportionality test?

III.  Judicial Control and Self-Restraint In reality, there is a long line of cases where the Court has refused to scrutinise the proportionality of EU measures strictly. This judicial self-restraint has resulted in a rather strong presumption of the legality of EU law. 3  See eg Case C‑558/07 SPCM et al v Secretary of State for the Environment, Food and Rural Affairs, ECLI:EU:C:2009:142, Opinion of AG Kokott, para 68, with further references.

The Evolution of the Principle of Proportionality in EU Law

 169

From early on the Court considered that the legislator enjoys wide discretion when the implementation of a policy necessitates the evaluation of a complex economic situation.4 Over time this consideration has been applied to all types of complex assessments and policy choices. Examples of this wide discretion are, of course, agriculture,5 but also the Working Time Directive,6 the rules on tobacco marketing,7 the REACH Regulation on chemicals8 and the OMT programme of the European Central Bank.9 This is a very reasonable approach. Policy choices are not to be made by a court. They are a task for political institutions that enjoy democratic legitimacy. Similar considerations apply to complex situations. As regards the assessment of technical and scientific elements the executive and the legislature are in a much better position than a court. They can rely on the technical expertise of their services and consult the best experts. As far as uncertainty remains it is again the responsibility of political institutions to make or at least confirm the necessary prognosis and to decide on the appropriate action. Therefore, EU courts cannot substitute their view on policy choices and complex assessments for that of the institutions on which alone the Treaties have conferred that task.10 As a consequence of this discretion of the legislator the Court significantly limits judicial review. As regards proportionality the only relevant test is whether the measure in question is based on a manifest error,11 but this test is not clearly defined. A very restricted understanding of manifest errors would limit this concept to errors that are immediately obvious to the judge. However, this would be a very subjective test, depending on the specific expertise of the judge. In our opinion a more objective standard is preferable, eg the absence of reasonable doubt. In this vein, one of the authors has argued that a prognostic decision in the area of state aid would only be manifestly erroneous if it were not justifiable from any conceivable

4 Judgments in Joined Cases C‑15/59 and C‑29/59 Société métallurgique de Knutange v High Authority, ECLI:EU:C:1960:4, 9; Case C‑55/75 Balkan-Import Export GmbH v Hauptzollamt BerlinPackhof, ECLI:EU:C:1976:8, para 8; Case C‑166/78 Italy v Council, ECLI:EU:C:1979:195, para 14; and Case C‑138/79 Roquette Frères v Council, ECLI:EU:C:1980:249, para 25. 5  Judgments in Case C‑265/87 Schräder HS Kraftfutter v Hauptzollamt Gronau, ECLI:EU:C:1989:303, para 22; Case C‑157/96 The Queen v Ministry of Agriculture, Fisheries and Food, Commissioners of Customs & Excise, ex p National Farmers’ Union et al, ECLI:EU:C:1998:191, para 60; and Case C‑101/12 Schaible v Land Baden-Württemberg, ECLI:EU:C:2013:661, paras 47 and 48. 6  Judgment in Case C‑84/94 United Kingdom v Council, ECLI:EU:C:1996:431, para 58. 7  Judgment in Case C‑491/01 The Queen v Secretary of State for Health, ex p British American Tobacco (Investments) and Imperial Tobacco, ECLI:EU:C:2002:741, para 123. 8  Judgment in Case C‑558/07 SPCM et al v Secretary of State for the Environment, Food and Rural Affairs, ECLI:EU:C:2009:430, para 42. 9  Judgment in Case C‑62/14 Gauweiler et al v Deutscher Bundestag, ECLI:EU:C:2015:400, para 68. 10  Judgments in Case C‑425/08 Enviro Tech (Europe) v Belgian State, ECLI:EU:C:2009:635, para 47; and Case C‑15/10 Etimine v Secretary of State for Work and Pensions, ECLI:EU:C:2011:504, para 60. 11  See eg the Judgments in Case C‑138/79 Roquette Frères v Council, ECLI:EU:C:1980:249, para 25; and Case C‑343/09 Afton Chemical v Secretary of State for Transport, ECLI:EU:C:2010:419, para 33.

170 

Juliane Kokott and Christoph Sobotta

point of view, ie if no person behaving rationally would have conducted itself in the same way.12 This is certainly a radical understanding of the manifest error test that will need to be tempered according to the nature of the specific case. But in the end a manifest error will only very rarely be found. A corollary of this approach is that the initial understanding of the judge, in particular in very technical cases, is not decisive as it can be improved by additional information submitted in the course of the proceedings. This information can either prove a manifest error by demonstrating—beyond reasonable doubt— that any apparent justifications are in fact unsustainable. Or it can provide reasonable explanations that the judge at first glance had not identified. In reality, most cases will be characterised by uncertainty, even after exhaustive explanation. Uncertainty about the right decision does not justify the finding of a manifest error. This is the main reason why there is wide discretion. The practice of the Court provides examples for all types of judgments. For example, it once had to review an obligation that was imposed on producers of animal feedstuffs. On request, they had to disclose the exact quantitative composition of their product to customers. These customers had already received information about this composition on the label, but with a 15 per cent error margin. Was this obligation manifestly unnecessary for the purpose of safeguarding animal and human health? The Court found that it was unnecessary and the reasoning is quite convincing.13 However, we would dispute that there was no conceivable point of view to justify the obligation. On the other hand, in Kokopelli the Court could not be convinced that the disadvantages of the very restrictive rules on the marketing of seed varieties manifestly outweighed their advantages. The EU regime aims to guarantee that seed reliably produces the expected product and therefore imposes a very burdensome system of control. As a consequence only very profitable seed varieties are sold legally on the EU market. Less profitable traditional or local varieties are in practice only sold illegally. Not only are the options of seed traders limited as a consequence, but the choices of farmers and consumers are also severely restricted. Moreover, biodiversity in Europe’s agriculture suffers. One of the authors considered it obvious that the objectives of the legislation could easily be met by labelling requirements.14 However, the Court practised more self-restraint,15 because there are points of view that could conceivably justify the restrictive rules.

12 

Case C‑525/04 Spain v Lenzing, ECLI:EU:C:2007:73, Opinion of AG Kokott, para 71. Judgment in Joined Cases C-453/03, C-11/04, C-12/04 and C-194/04 The Queen, on the application of ABNA et al v Secretary of State for Health and Food Standards Agency et al, Fratelli Martini and Cargill v Ministero delle Politiche Agricole e Forestali et al, Assalzoo v Ministero delle Politiche Agricole e Forestali et al and Nevedi v Productschap Diervoeder, ECLI:EU:C:2005:741, paras 80–85. 14 Case C‑59/11 Association Kokopelli v Graines Baumaux, ECLI:EU:C:2012:28, Opinion of AG Kokott, in particular paras 88–94. 15  Judgment in Case C‑59/11 Association Kokopelli v Graines Baumaux, ECLI:EU:C:2012:447, in particular para 59. 13 

The Evolution of the Principle of Proportionality in EU Law

 171

And perhaps the Opinion really entered too far into a heated political debate. Two years after Kokopelli a Commission proposal that included less restrictive rules for certain niche markets was almost unanimously rejected by the European Parliament.16 However, to a certain degree the position of the Opinion was vindicated because most of the MEPs complained that this proposal was still far too restrictive.17 Regrettably, there is no sign that the Commission took this defeat to heart and will propose a less restrictive regime in the near future. In any event, what we should draw from the discussion in this section is that the manifest error test will only allow for a very limited review of proportionality and that the exact scope of this test remains unclear.

IV.  Compensatory Scrutiny However, the Court developed standards to compensate for the limited substantive scrutiny in most proportionality cases. It considers that where an EU institution enjoys broad discretion, a review of compliance with certain procedural guarantees is of fundamental importance. Those guarantees include, first, the obligation to examine carefully and impartially all the relevant elements of the situation in question; and, second, the duty to give an adequate statement of the reasons.18

A.  On the Facts However, as regards the obligation to examine all the relevant elements on closer inspection the jurisprudence appears contradictory. On the one hand, the Court sometimes extends wide discretion also to the finding of the basic facts. In this context the Court has accepted that it is open to the legislator to rely on general findings.19 One would not expect very strict scrutiny under this standard. On the other hand, the Court has in recent years elaborated on the necessary investigation of facts. Three conditions need to be met: (i) the evidence relied on must be factually accurate, reliable and consistent;

16 

European Parliament legislative resolution of 11 March 2014, P7_TA(2014)0185. Minutes of the debate on 10 March 2014, CRE 10/03/2014—13. 18 Judgments in Case C‑269/90 Technische Universität München v Hauptzollamt MünchenMitte, ECLI:EU:C:1991:438, para 14; Case C‑326/05 P Industrias Químicas del Vallés v Commission, ECLI:EU:C:2007:443, para 77; and Case C‑62/14 Gauweiler et al v Deutscher Bundestag, ECLI:EU:C:2015:400, para 69. 19  Judgments in Case C‑166/78 Italy v Council, ECLI:EU:C:1979:195, para 14; Joined Cases C‑248/95 and C‑249/95 SAM Schiffahrt and Stapf v Germany, ECLI:EU:C:1997:377, para 25; Case C‑4/96 NIFPO and Northern Ireland Fishermen’s Federation v Department of Agriculture for Northern Ireland, ECLI:EU:C:1998:67, para 42; Joined Cases C‑27/00 and C‑122/00 The Queen v Secretary of State for the Environment, Transport and the Regions, ex p Omega Air, and Omega Air et al v Irish Aviation 17 

172 

Juliane Kokott and Christoph Sobotta

(ii) the evidence must contain all the information which must be taken into account in order to assess a complex situation; and (iii) that evidence must be capable of substantiating the conclusions drawn from it. This could result in a very strict review of the factual preparation of discretionary measures, in particular as regards proportionality. However, looking into the cases it becomes clear that this is not a general standard in the jurisprudence of the Court. These conditions are mostly applied to individual decisions, eg in the areas of competition20 and state aid,21 and occasionally with regard to very technical implementing legislation.22 Some Opinions suggested extending this standard of review to general legislation,23 but the Court failed to take up this suggestion explicitly. There may be some good reasons for the Court’s reluctance. Individual decisions, even if they are taken in the exercise of wide discretion, are typically subject to much more specific conditions than general legislation. In the light of these conditions it may be possible to examine, in particular, whether the collected information was capable of substantiating the conclusions drawn from it. Conversely, general legislation mostly depends on legal bases that do not provide for such specific conditions. Therefore, it is a necessary part of legislative discretion to decide which conclusions can be drawn from the available information. Regardless, there are signs that the standard of review for the facts underlying legislation has been strengthened more recently. The most obvious indication is that today the Court usually insists that all relevant elements are taken into account, even if it accepts broad discretion for the finding of the basic facts.24

Authority, ECLI:EU:C:2002:161, para 65; Case C‑343/07 Bavaria and Bavaria Italia v Bayerischer Brauerbund, ECLI:EU:C:2009:415, para 84; and Case C‑34/08 Azienda Agricola Disarò Antonio et al v Cooperativa Milka 2000, ECLI:EU:C:2009:304, para 37. 20 Judgments in Case C‑12/03 P Commission v Tetra Laval, ECLI:EU:C:2005:87, para 39; Case C‑413/06 P Bertelsmann and Sony Corporation of America v Impala, ECLI:EU:C:2008:392, para 145; Case C‑386/10 P Chalkor v Commission, ECLI:EU:C:2011:815, para 54; Case C‑199/11 Europese Gemeenschap v Otis et al, ECLI:EU:C:2012:684, para 59; Case C‑510/11 P Kone et al v Commission, ECLI:EU:C:2013:696, para 28; Case C‑295/12 P Telefónica and Telefónica de España v Commission, ECLI:EU:C:2014:2062, para 54; and Case C‑67/13 CB v Commission, ECLI:EU:C:2014:2204, para 46. 21  Judgments in Case C‑525/04 P Spain v Lenzing, ECLI:EU:C:2007:698, para 57; Case C‑73/11 Frucona Košice v Commission, ECLI:EU:C:2013:32, para 76; Case C‑405/11 P Commission v Buczek Automotive, ECLI:EU:C:2013:186, paras 50 and 53; Joined Cases C‑214/12 P, C‑215/12 P and C‑223/12 P Land Burgenland et al v Commission, ECLI:EU:C:2013:682, para 79; Case C‑271/13 Rousse Industry v Commission, ECLI:EU:C:2014:175, para 59, and Joined Cases C‑533/12 P and C‑536/12 P SNCM and France v Corsica Ferries France, ECLI:EU:C:2014:2142, para 15. 22  Judgments in Case C‑326/05 P Industrias Químicas del Vallés v Commission, ECLI:EU:C:2007:443, para 77; Case C‑405/07 P Netherlands v Commission, ECLI:EU:C:2008:613, para 55, and Judgment in Case C‑360/14 P Germany v Commission, ECLI:EU:C:2015:457, para 37. 23  See Case C‑558/07 SPCM et al v Secretary of State for the Environment, Food and Rural Affairs, ECLI:EU:C:2009:142, Opinion of AG Kokott, para 77; Case C‑343/09 Afton Chemical v Secretary of State for Transport, ECLI:EU:C:2010:258, Opinion of AG Kokott, para 29, as well as Case C‑101/12 Schaible v Land Baden-Württemberg, ECLI:EU:C:2013:334, Opinion of AG Wahl, para 38. 24  Judgments in Case 310/04 Spain v Council, ECLI:EU:C:2006:521, paras 121 and 122, as well as Case C‑343/09 Afton Chemical v Secretary of State for Transport, ECLI:EU:C:2010:419, paras 32 and 33.

The Evolution of the Principle of Proportionality in EU Law

 173

Moreover, even for general legislation there are areas where the necessary information to be taken into account has been further specified. For health, safety, environmental protection and consumer protection legislation the Treaty on the Functioning of the European Union requires that the available scientific and technical data are taken into account.25 This concept has been developed in greater detail with regard to legislation that expresses the precautionary principle. This principle justifies the adoption of restrictive measures in case of uncertainty over the existence of a risk that could result in real harm to public health or the environment should the risk materialise. In our opinion, the Court aims to prevent this principle from opening the way for arbitrary measures to be adopted. Therefore, the application of the precautionary principle presupposes, first, identification of the potentially negative consequences for health or the environment, and, second, a comprehensive assessment of the risk based on the most reliable scientific data available and the most recent results of international research.26 To sum up this section, we can see that the limited substantive scrutiny of the application of the principle of proportionality is being accompanied by an increasing focus on the preparation of the factual basis of the decision. However, there is still some uncertainty as to how strictly the evaluation of facts will be reviewed.

B.  On the Duty to Give Reasons Ideally, these requirements on the investigation of facts would be supplemented by the duty to give reasons laid down in Article 296 (2) TFEU. Consequently, it could be expected that the reasons document in particular the necessary fact-finding and reflect the exercise of discretion. However, the jurisprudence is less clear on this point. According to the Court, the statement of reasons must be appropriate to the nature of the measure concerned. It must disclose in a clear and unequivocal fashion the reasoning followed by the EU institution which adopted the measure in such a way as to make the persons concerned aware of the reasons for the measure. However, it is not necessary for all relevant factual and legal aspects to be mentioned in the reasoning. The context and all the legal rules governing the matter in question must also be taken into account. In particular, if the contested measure clearly discloses the essential objective pursued, it would be excessive

See also the orders in Case C‑287/13 P Bilbaína de Alquitranes et al v ECHA, ECLI:EU:C:2014:599, para 20, and Case C‑290/13 P Rütgers Germany et al v ECHA, ECLI:EU:C:2014:2174, para 25, approving corresponding findings of the General Court. 25 

See Arts 114(3) and 191(3) TFEU. in Case C‑333/08 Commission v France, ECLI:EU:C:2010:44, paras 91–93, and Case C‑343/09 Afton Chemical v Secretary of State for Transport, ECLI:EU:C:2010:419, paras 60–62. 26  Judgments

174 

Juliane Kokott and Christoph Sobotta

to require a separate and specific statement of reasons for the various technical choices made.27 Under these criteria it could be argued that even very contentious measures do not require specific reasons if they can be qualified as technical choices. In particular it would not appear necessary to lay down the information that was taken into account if only the essential objective was stated. Nevertheless, in this regard the context of a measure becomes important. If technical choices have been discussed publicly during the preparation of legislation, the corresponding documentation should show whether the relevant facts and studies were part of the process. Commission consultations of stakeholders28 and hearings by the Parliament29 are the most important fora in this regard. We see that the duty to provide reasons complements the obligation to examine all the relevant elements of the situation. But the level of scrutiny applied to both obligations in the context of an assessment of proportionality is not very clearly defined.

V.  Towards a Substantial Criterion? However, two recent cases on the protection of personal data appear to change this setting significantly. These are Schecke30 and Digital Rights Ireland31 where the Court declared EU legislation incompatible with the fundamental right to privacy under Article 7 of the Charter of Fundamental Rights and the fundamental right to the protection of personal data under Article 8. Schecke was about EU rules that required the publication of the names of the beneficiaries of agricultural aid and of the amounts they received. The Court was not convinced that these rules were necessary to attain the objective of transparency. More precisely, the Court criticised the fact that the institutions could not show that they took into consideration less invasive methods of publishing the relevant information.32 There is no mention of broad discretion or manifest errors.33 27  Summed up in the Judgment in Joined Cases C‑27/00 and C‑122/00 The Queen v Secretary of State for the Environment, Transport and the Regions, ex p Omega Air, and Omega Air et al v Irish Aviation Authority, ECLI:EU:C:2002:161, paras 46 and 47. 28  Judgment in Case C‑58/08 Vodafone et al v Secretary of State for Business, Enterprise and Regulatory Reform, ECLI:EU:C:2010:321, para 65. 29  cf Judgment in Case C‑343/09 Afton Chemical v Secretary of State for Transport, ECLI:EU:C:2010:419, para 36. 30  Judgment in Joined Cases C‑92/09 and C‑93/09 Volker and Markus Schecke and Eifert v Land Hessen, ECLI:EU:C:2010:662. 31  Judgment in Joined Cases C‑293/12 and C‑594/12 Digital Rights Ireland v Minister for Communications, Marine and Natural Resources et al and Kärntner Landesregierung et al, ECLI:EU:C:2014:238. 32  Judgment in Joined Cases C‑92/09 and C‑93/09 Volker and Markus Schecke and Eifert v Land Hessen, ECLI:EU:C:2010:662, paras 81 and 86. 33  cf in this regard also the examination of proportionality in the Judgment in Case C‑283/11 Sky Österreich v Österreichischer Rundfunk, ECLI:EU:C:2013:28, paras 50ff.

The Evolution of the Principle of Proportionality in EU Law

 175

Is this case an isolated outlier? There is some temptation to think so because the Court refused to apply a similar level of scrutiny in subsequent cases. One example is the Kokopelli case discussed above on the marketing of seed varieties where the unwillingness to consider less invasive restrictions, such as labelling, was quite evident. However, Digital Rights Ireland can help us to understand the subtext of Schecke. This case concerned the retention of communication metadata, required by an EU Directive. The Court considered this measure to be disproportionate and therefore invalid. What is of interest here is the discussion of the discretion afforded to the legislature in this case. One would expect the standard consideration that the measure in question clearly requires policy choices in a very complex and sensitive environment, namely the fight against serious crime and in particular terrorism. However, the Court argued that where interferences with fundamental rights are at issue, the extent of the EU legislature’s discretion may prove to be limited. In this regard a number of factors were mentioned, namely the area concerned, the nature of the right at issue, the nature and seriousness of the interference, and the object pursued by the interference.34 Remarkably, this finding was based on the Marper case of the European Court for Human Rights.35 Similar considerations can be found in the jurisprudence of the German Bundesverfassungsgericht36 and in the different levels of scrutiny applied by the American Supreme Court with regard to the equal protection clause. While the rational basis test37 would correspond to the traditional assessment of proportionality by the CJEU the standard applied in Digital Rights Ireland corresponds more closely to intermediate scrutiny.38 Indeed it cannot be excluded that at some point even the level of strict scrutiny39 will be reached. This substantive criterion could explain what the Court did in Schecke without providing us with an explicit explanation: it did not assume broad discretion because of the policy choices involved, but focused on the substantial interference with privacy and personal data. As a consequence it considered the legislature’s discretion to be limited and required more stringent reasoning with regard to the necessity of the interference.

34  Judgment in Joined Cases C‑293/12 and C‑594/12 Digital Rights Ireland v Minister for Communications, Marine and Natural Resources et al and Kärntner Landesregierung et al, ECLI:EU:C:2014:238, para 47, confirmed by the Judgment in Case C‑362/14 Schrems v Data Protection Commissioner, ECLI:EU:C:2015:650, para 78. 35  S and Marper v the United Kingdom App nos. 30562/04 and 30566/04 (2008) ECHR 1581, para 102. 36  eg BVerfGE 42, 143, 149, on freedom of opinion. 37 eg FCC v Beach Communications, Inc, 508 US 307 (1993). 38 eg Craig v Boren, 429 US 190 (1976). 39 eg Loving v Virginia, 388 US 1 (1967).

176 

Juliane Kokott and Christoph Sobotta

In contrast, Kokopelli primarily concerned the freedom to conduct a business40 and this only at the margins. There was no legal market for the sale of seeds that did not meet the very restrictive conditions imposed by EU legislation. The interference with the mere hope to be able to sell such seeds in the future could not be considered as serious as the interferences discussed in Digital Rights Ireland or Schecke. The same idea could help us to understand the ambiguous findings on the investigation of relevant facts and the duty to provide reasons. We would argue that in these areas, too, the Court is already implicitly balancing the objectives with the interference to determine how deeply it will look into the matter. And this brings us back to the beginning and to the complaint that in EU law proportionality is examined in isolation from the right that is restricted. The stringent review of legislative discretion under the criteria of Digital Rights Ireland is only possible if the affected right is taken into account. Moreover, Digital Rights Ireland suggests that later stages of the proportionality test are already anticipated when the discretion of the legislator with regard to all steps of the proportionality test is specified. Under Digital Rights Ireland the starting point is the affected right as well as the objective and the seriousness of the interference. But the latter issue only comes into focus in the final stage of the proportionality test when the disadvantages and advantages of the measure are balanced. This anticipation of later stages of the test is the anticipatory understanding of proportionality mentioned in the title of this presentation.

VI. Conclusion To conclude, this presentation has covered the evolution of the principle of proportionality with regard to judicial scrutiny of EU legislative measures from its cautious beginning until today. In this respect Digital Rights Ireland has provided an important new perspective to discuss the level of scrutiny. For the time being the traditional way of reasoning still dominates the jurisprudence. It only requires consideration of whether the decision entails policy choices and/or the assessment of complex situations. But it is to be expected that in future the more refined anticipatory understanding of proportionality that has been introduced by Digital Rights Ireland will be explored more deeply. Already there is remarkable development towards a well-structured proportionality test, possibly inspired by other legal systems.

40  cf Judgment in Case C‑283/11 Sky Österreich v Österreichischer Rundfunk, ECLI:EU:C:2013:28, para 46.

The Evolution of the Principle of Proportionality in EU Law

 177

Obviously, there are other areas that would also merit discussion. The most important ones are probably the application of the principle of proportionality to and by the Member States. We would argue that the EU courts should, in principle, apply similar standards in those areas. However, we are even further away from explicit pronunciations in this regard than in the area addressed here. But this should not be surprising because we have seen that even in the latter area it took a lot of time to arrive at the current level of understanding. And if we put the Member States into the equation things will become much more complex and sensitive.

178 

11 Proportionality and the Margin of Appreciation: Strasbourg and London PHILIP SALES

In this essay I examine how the doctrine of proportionality currently seems to be operating under the law of the European Convention on Human Rights (ECHR) in the jurisprudence of the European Court of Human Rights (ECtHR) and in the jurisprudence of the UK’s Supreme Court. I begin with a short discussion about what the doctrine has to offer in terms of rule of law values.

I.  Proportionality and the Rule of Law The idea of the ‘rule of law’ is a cluster of values, some of which are in tension with each other.1 As is well known, there is a spectrum between more substantive and more formal versions of the idea. There is also a tension between an emphasis on being able to predict how laws will be applied (the rule of law, rather than men; promotion of individual—or state—autonomy and the ability to plan) and on the injection of substantive values into decision-making when a judge applies the law. Proportionality is a rather loose concept, operating as a standard rather than a clear-cut rule. Each stage of analysis—(i) identification of a legitimate aim for the measure in question; (ii) assessment of the suitability or rational connection between the measure and the aim; (iii) its necessity, in the sense of requiring minimal impairment of the affected right; and (iv) its proportionality ‘in the strict sense’, of providing a social gain which outweighs the impairment of the

1  RH Fallon Jr, ‘“The Rule of Law” as a Concept in Constitutional Discourse’ (1997) 97 Columbia LR 1; J Waldron, ‘Is the Rule of Law an Essentially Contested Concept (in Florida)?’ (2002) 21 Law and Philosophy 137; R Bellamy, ‘The Rule of Law and the Rule of Persons’ (2001) 4 Critical Review of International Social and Political Philosophy 221. As Bellamy puts it, the question is ‘whether one can link individual autonomy and legal authority in ways that avoid the Scylla of lawless anarchy and the Charybdis of legalistic despotism, without falling victim to personal tyranny. Theories of the Rule of Law aim to steer a course between these three dangers’ (224).

180 

Philip Sales

right2—carries within it potentially controversial value judgements. Its looseness can be both a strength and a weakness;3 and it has been criticised as oscillating between two different conceptions—one that ‘praises, eg, its technical rigor, its neutrality, and its manageability’, ie as a determinate legal mode of analysis in relation to discretionary power, and another that emphasises the way in which it allows ‘for every possible consideration to be assessed through unconstrained moral reasoning’,4 but where this ‘leaves the judge unguided, unaided in dealing with this complexity, exposing the judge to the pressure exerted by those directly interested in the decision of the case or by public opinion generally, as well as to challenges to the legitimacy of his decision’.5 If the point of law is taken to be to exclude generalised moral argument or to allow it to operate in very circumscribed ways so as to make disputes soluble in socially acceptable ways,6 the latter vision of proportionality tends to undermine it by reintroducing moral argument. The looseness of the notion of proportionality also makes it a notably malleable concept, open to influence by background considerations as regards the strictness or flexibility of the content given to it. In the language of the ECtHR’s case law, this is where the width or narrowness of the margin of appreciation is critical.7 It can be difficult to pin down these background factors and the extent of their impact, both descriptively and normatively. A standard list of factors which tend to widen the margin of appreciation would include where a measure relates to matters of social and economic importance, an area of general policy in relation to which opinions may reasonably differ in a democracy, or sensitive moral or ethical issues, or where a state has had to balance competing interests and rights (in particular, Convention rights) or there is an absence of a clear common approach across members of the Council of Europe.8 Even when one articulates the range of considerations which may apply, there still remains a gap between that articulation and the actual outcome of a proportionality analysis in any given case. This reflects the problem of the incommensurability of the values which need to be brought into account in a proportionality assessment. There is not a standard, agreed commensurability metric to which appeal can be made.9 This poses a 2  See, eg, J Rivers, ‘Proportionality and the Variable Intensity of Review’ [2006] Cambridge Law Journal 174, 181; HM Treasury v Bank Mellat (No 2) [2013] UKSC 39, [74] (Lord Reed). 3  cf Timothy Endicott, ‘The Value of Vagueness’ in A Marmor and S Soames (eds), Philosophical Foundations of Language in the Law (Oxford, Oxford University Press, 2011) ch 2. 4  FJ Urbina, ‘A Critique of Proportionality’ (2012) 57 American Journal of Jurisprudence 49, 50; and see G Webber, ‘Rights and the Rule of Law in the Balance’ (2013) 129 LQR 399. 5  ibid, 80. 6  See, eg, F Schauer, ‘Is the Common Law Law?’ (1989) 77 California Law Review 455; F Schauer, The Force of Law (Cambridge, MA, Harvard University Press, 2015) 105, referring to the settlement function of law in relation to morally contentious issues, to allow ‘decisions to be made, actions to be coordinated, and life to go on’. 7  P Sales, ‘Law and Democracy in a Human Rights Framework’ in D Feldman (ed), Law in Politics, Politics in Law (Oxford, Hart Publishing, 2013) ch 15, 228–33. 8  ibid, 230–31. 9  RH Fallon Jr, ‘The Supreme Court 1996 Term: Foreword: Implementing the Constitution’ (1997) 111 Harv LR 54; R Chang (ed), Incommensurability, Incomparability and Practical Reason (Cam-

Proportionality and the Margin of Appreciation: Strasbourg and London 181 problem for proportionality analysis in terms of rule of law values, since the absence of an agreed metric creates a tension with the rule of law values of predictability and the wish to avoid capriciousness and domination through the rule of men rather than law. However, this is hardly a problem unique to the application of the ECHR. Problems of the weighing of incommensurable values pervade most areas of the law, and in particular the common law, both in relation to the production of determinate doctrinal rules and in the application of broad standards.10 Leaving aside specific legislative action, there are two basic ways in which the uncertainty inherent in drawing determinate rules or determinate outcomes from the background of incommensurable issues can be addressed and kept within reasonable bounds: by relying on a normative consensus among the professional elite with responsibility for working with and applying the norms, and by gradual articulation of authoritative guidance through which expectations can be focused and judicial behaviour managed. As Richard Fallon notes in relation to the first, ‘In contexts marked by normative consensus, there might be broad agreement about how standards should be applied, and standards would permit both citizens and officials to be ruled by law’; ‘In short, the greater the normative consensus, the less the realization of Rule of Law values depends on the law being a law of rules.’11 Moreover, through judicial application on the basis of open reasoning, vague standards may come to crystallise into more brightline rules over time.12 In this way, courts can assume a leadership role to guide expectations. However, it is an open question how far either solution can be regarded as fully satisfactory, particularly in the context of application of the ECHR. Reliance for rule of law purposes on standards implicit in a legal culture can often seem to

bridge, MA, Harvard University Press, 1997); T Endicott, ‘Proportionality and Incommensurability’ in G Huscroft, BW Miller and G Webber (eds), Proportionality and the Rule of Law: Rights, Justification, Reasoning (Cambridge, Cambridge University Press, 2014) ch 14; J Gray, Isaiah Berlin: An Interpretation of his Thought (Princeton, Princeton University Press, 1996) ch 2; Urbina (n 4); Webber (n 4). 10  KM Sullivan, ‘The Justices of Rules and Standards’ (1992) 106 Harv L Rev 22; D Kennedy, ‘Form and Substance in Private Law Adjudication’ (1976) 89 Harv LR 1685; A Scalia, ‘The Rule of Law as a Law of Rules’ (1989) 56 U Chi L Rev 1175. 11  Fallon (n 1) 49 and 52. This is an aspect of law emphasised by, eg, K Llewellyn in The Common Law Tradition: Deciding Appeals (Boston, Little, Brown, 1960); and by M Eisenberg in The Nature of the Common Law (Cambridge, MA, Harvard University Press, 1988). The roots of the Western legal tradition in the creation of a specialist body of jurists and judges, with specialist training to use specifically legal materials, is highlighted in H Berman, Law and Revolution (Cambridge, MA, Harvard University Press, 1983) 8; and see F Schauer, The Force of Law (n 6) ch 11, ‘The Differentiation of Law’. 12  Webber (n 4) 416–17; cf H Dagan, ‘Private Law Pluralism and the Rule of Law’ in L Austin and D Klimchuk (eds), Private Law and the Rule of Law (Oxford, Oxford University Press, 2014) ch 7. As Fallon observes about constitutional law in the USA, ‘[o]nce established, doctrine frequently serves as a focal point for reasonable agreement between the justices’, even where they have divergent views about the constitution: Fallon (n 90) 151. In this way, using an enlarged sense of the rule of law, law may be given a new and wider part in the regulation of society, as an aspect of establishing and maintaining political authority: see N Johnson, In Search of the Constitution (Oxford, Pergamon Press, 1977) 149–50. It may be possible to speak of a European common law system of human rights: see Sales (n 7) 231–33.

182 

Philip Sales

fall short of those purposes.13 Specifically as regards the ECHR, there is a marked gap between the broad, abstract formulations of the rights set out in it, which are by reason of their abstractness underdeterminative of the result in any particular case, and their application in concrete circumstances. This leaves the courts which have to apply the Convention rights with a high degree of choice and decisionmaking power. The doctrine of proportionality seeks to bridge the gap with some form of doctrinal framework, but it is one that still leaves a significant area within which important choices fall to be made by individual judges. Yet the legal culture within which they operate is comparatively weak. The judiciary who staff the ECtHR come from disparate backgrounds and traditions, and national courts which seek to apply Convention rights domestically will have their own cultures and perspectives. The international judges may seem remote from the consensus standards prevailing in a particular contracting state, so that their judgments sometimes strike a dissonant note;14 on the other hand, the domestic judges may seem too narrowly parochial in their approach, and overinfluenced by domestic public opinion. There are various courts which operate with the Convention rights or similar (the ECtHR, the CJEU, national courts), and they may seek to guide their application in different directions. Moreover, the Convention rights have such wide ambit and potential penetration into the detail of resolution of contentious issues in society that they have implications for the outcome of politically controversial questions, in relation to which no social consensus exists and there is no generally accepted rational method of resolution.15 Such issues are not perceived as amenable to judicial resolution, but rather are regarded within a polity’s own traditions, and those of Western Europe more generally, as calling for resolution through political (in particular, democratic) procedures.16 Further, where judges are asked to resolve questions of this character, it may be difficult to sustain their claims to legitimacy

13  Schauer, ‘Is the Common Law Law?’ (n 6) 468–69; F Schauer, ‘The Failure of the Common Law’ (2004) 36 Arizona State LJ 765; AWB Simpson, ‘The Common Law and Legal Theory’ in W Twining (ed), Legal Theory and Common Law (Oxford, Blackwell, 1986) ch 2; and for an historical account of how in the common law the idea of custom as derived from the practices and values of the judiciary and legal profession became blurred with the idea of custom derived from the values of the general community, see A Cromartie, ‘The Idea of Common Law as Custom’ in A Perreau-Saussine and J Bernard Murphy (eds), The Nature of Customary Law: Legal, Historical and Philosophical Perspectives (Cambridge, Cambridge University Press, 2007) ch 9. In the US context, Scalia is sceptical about how effective this constraint is within US national legal culture: see Scalia (n 10). 14  The British concern—see, eg, the report of the Commission on a Bill of Rights, A UK Bill of Rights? The Choice Before Us (2012) and Lord Hoffmann, ‘The Universality of Human Rights’ (2009) 125 LQR 416—that the ECtHR’s interpretation of Convention rights appears alien and intrusive in some cases is partly a reflection of the absence of a strong, shared normative consensus, which may be easier to construct at a national level and within a national legal tradition operating on the basis of a stronger platform of shared cultural presuppositions. 15  ie where what Jeremy Waldron calls the ‘the circumstances of politics’ prevail: J Waldron, Law and Disagreement (Oxford, Clarendon Press, 1999) 101–03. 16  See, eg, J Waldron, ‘Do Judges Reason Morally?’ in G Huscroft (ed), Expounding the Constitution: Essays in Constitutional Theory (Cambridge, Cambridge University Press, 2009) ch 2.

Proportionality and the Margin of Appreciation: Strasbourg and London 183 in doing so in the face of democratic ideology17 and populist pressures.18 Judicial decision-making comes into more acute conflict with political modes of dispute resolution and compromise. Where judges are not perceived to be subject to some form of constraint arising from objectively identifiable social or legal norms, the legitimacy of decision-making by them may be difficult to establish. Judicial decisions applying Convention rights arise at the point of tension between two strands of thinking in liberal democracies, where ideas about the rule of law and respect for human rights on the one hand and about democracy on the other have an uneasy coexistence.19 As John Dunn writes, ‘Democracy is not the rule of law. It does not ensure it, and it is still far from clear that it may not sooner or later to varying degrees preclude it … each of these two is a distinct partial good, with every potentiality to conflict with the other.’20 Both principles have legitimising power, and may resonate more or less strongly depending on how closely a particular matter may approach the zones of central concern under each principle respectively. The tension is resolved through practice, through the acceptability of the outcomes achieved and (one hopes) the gathering trust in the decision-making processes which secure such outcomes. In understanding the application of the proportionality doctrine, the constitutional and political context is important in each case. Benjamin Cardozo noted in 1921 that there was a growing tendency for judges to leave legal development to legislation, in recognition of legislatures being in frequent session and being capable of establishing laws themselves.21 With the growing acceptance of democratic ideology, judges’ perception of what it is right for them to do in laying down legal principles and in deciding particular cases is affected by consciousness of whether there is a legislature that is both effective in making law in response to social needs and has strong democratic legitimacy in doing so.22 Here, the position of the CJEU, which is juxtaposed with legislative institutions with relatively weak democratic credentials, and of the ECtHR, which operates in a context of 47 very disparate national legislatures, none of which has competence to legislate across

17  For the comparatively recent dominance of democratic ideology, see J Dunn, Setting the People Free: The Story of Democracy (London, Atlantic, 2005). 18  cf A De Tocqueville, Democracy in America, vol I [1835] (1994) 280: ‘Scarcely any political question arises in the United States that is not resolved, sooner or later, into a judicial one’, but this is on the basis of a powerful loyalty to the constitution, supported by a relative openness of access to the legal profession and judicial office: cf L Siedentop, Democracy in Europe (London, Penguin, 2001) 150; M Loughlin, The Idea of Public Law (Oxford, Oxford University Press, 2003) 128–30. 19  G Webber, The Negotiable Constitution: On the Limitation of Rights (Cambridge, Cambridge University Press, 2009); Sales (n 7). 20  J Dunn, Breaking Democracy’s Spell (New Haven, Yale University Press, 2014) 32–33. 21  B Cardozo, The Nature of the Judicial Process (New Haven, Yale University Press, 1921) 60–61; also see AV Dicey, Lectures on the Relation between Law & Public Opinion in England during the Nineteenth Century (London, Macmillan, 1914) 362–63; JHH Weiler, The Constitution of Europe (Cambridge, Cambridge University Press, 1999). 22  cf F Schauer, ‘The Failure of the Common Law’ (n 13); N Duxbury, Elements of Legislation (Cambridge, Cambridge University Press, 2013) 32–36; J Waldron, “Can There be a Democratic Jurisprudence?’ (2009) 58 Emory LJ 675.

184 

Philip Sales

the whole of its field of action, may be contrasted with that of a national supreme court, operating alongside a democratic national legislature. The relatively open-textured nature of proportionality analysis makes the proportionality doctrine a vessel into which a wide range of background values can be poured, and in which different elements can achieve prominence depending on the values which are emphasised. This creates a centrifugal tendency in the application of the doctrine, undermining its status as a doctrine of uniform application. It matters to ask which court is applying the doctrine, for what purpose and in the context of which body of law. But at the same time, the formulation of the doctrine at a level of some abstraction allows for a distinct countervailing unifying tendency. Efforts are made to try to standardise the formulation and application of the doctrine, which helps leading courts (the CJEU, the ECtHR and superior national courts) to communicate in a common language in relation to areas of overlapping interest and allows a national court such as the UK Supreme Court to integrate and simplify its approach to proportionality analysis across the separate streams of law (including domestic public law, EU law, the law of the ECHR and, sometimes, public international law) which it may have to bring into account to decide a case.

II.  Proportionality in Strasbourg The doctrine of proportionality as applied in Strasbourg is inextricably bound up with the margin of appreciation, which has itself developed, with increasing concreteness, as a key concept in the jurisprudence of the ECtHR. The margin of appreciation governs the intensity with which the proportionality standard is applied in a given context. It has attracted considerable criticism. It is charged with being arbitrary and unclear in its application and as having no clear normative basis.23 But it is clearly not going to disappear. Indeed, as discussed below, it is assuming greater prominence in the ECtHR’s case law than ever before. Can a coherent and convincing account be given of its normative underpinnings, which has the capacity to explain the ECtHR’s case law and both guide its development and assist in predicting outcomes in particular cases? To an important extent, the intensity with which the proportionality doctrine is applied by the ECtHR reflects the tension of the rule of law and human rights tradition with the democratic tradition, and represents an effort to accommodate both of them.24 But this is not the whole picture. I suggest that the margin of appreciation may best be understood as operating along three axes, as a mechanism (i) to accommodate the tension between Convention rights with democratic 23  See, eg, G Letsas, ‘Two Concepts of the Margin of Appreciation’ (2006) 26 OJLS 705; Lord Lester of Herne Hill, ‘Universality versus Subsidiarity: A Reply’ (1998) 1 EHRLR 73. 24  Sales (n 7) above.

Proportionality and the Margin of Appreciation: Strasbourg and London 185 decision-making; (ii) to provide a space for the determinative application of local expertise or superior knowledge of relevant circumstances within the national judicial and regulatory systems; and, increasingly, (iii) to regulate the calls upon the ECtHR’s time and attention, in recognition of the role of the Court as, in effect, a sort of constitutional court of the European public order with responsibility for what is now a relatively mature human rights system with good penetration and understanding among national courts. Each axis has different implications for the operation of the proportionality doctrine at the national level. Draon v France25 provides a good example of a lower-level intensity of application of the proportionality doctrine in the face of recognised social disagreement and democratic resolution (axis (i)). The ECtHR recognises that the national authorities have direct democratic legitimation and are in principle better placed than an international court to evaluate local needs and conditions. In matters of general policy, on which opinions within a democratic society may reasonably differ widely, the role of the domestic policy-maker should be given special weight.

The need to accommodate the tension between human rights and democracy is present for a national court as well. Where the ECtHR provides guidance as to the width of the margin of appreciation along this axis in applying Convention rights, then it is reasonable to expect national courts to follow this lead, at least in ordinary circumstances.26 Axis (ii) is reflected in the notion of subsidiarity applied by the ECtHR;27 respect for factual decision-making by national authorities in cases requiring sensitive assessment of complex, interacting factual circumstances, such as those involving decisions whether to take children into care;28 and the so-called ‘fourth-instance

25  Draon v France (2006) 42 EHRR 40, GC, paras 106–08; also see Hatton v UK (2003) 37 EHRR 28, para 97; SAS v France (2015) 60 EHRR 11, GC, para 129. 26  See, eg, SRM Global Fund LLP v Commissioners of HM Treasury [2009] EWCA 788, [57]–[59], [73]–[78]; AXA General Insurance Ltd v HM Advocate [2011] UKSC 46, [2012] 1 AC 868. 27  See, eg, the Belgian Linguistics judgment (1979–80) 1 EHRR 252, para 10 (referring to the ‘subsidiary nature of the international machinery of collective enforcement established by the Convention’); Austin v UK (2012) 55 EHRR 14, GC, para 61; and the declarations which emerged from the Interlaken conference (2010), the Izmir conference (2011) and the Brighton conference (2012). The Brighton conference led to the promulgation of Protocol No 15, opened for signature in 2013, to insert a new recital into the Preamble of the ECHR to affirm that, ‘in accordance with the principle of subsidiarity’, the High Contracting Parties have the primary responsibility to secure Convention rights and freedoms, ‘and that in doing so they enjoy a margin of appreciation, subject to the supervisory jurisdiction of the [ECtHR]’. On a wide view of the principle of subsidiarity, the margin of appreciation can itself be said to be an aspect of it; but that tends to obscure the different strands covered by the margin of appreciation. 28 eg K and T v Finland (2003) 36 EHRR 18, GC, para 154; also see, eg, Chapman v UK (2001) 33 EHRR 18, para 92 (exercise of planning discretion); Eweida v UK (2013) 57 EHRR 8, para 99 (assessment of clinical safety); cf Jeunesse v Netherlands (2015) 60 EHRR 17, GC, para 120 (domestic authorities failed to consider and assess evidence as to the situation of children in an immigration case). As part of the proportionality doctrine and in exercise of its supervisory role, the ECtHR requires there to be ‘relevant and sufficient’ reasons adduced by the national authorities, including in relation to the evidence of any factual basis relied upon as a foundation for a measure which impinges on Convention rights: see, eg, Vogt v Germany (1996) 21 EHRR 205, para 52(iii); and Smith and Grady v UK (1999)

186 

Philip Sales

rule’, according to which the ECtHR adopts a very limited role when checking whether domestic law has been correctly interpreted and applied and facts correctly found by national courts,29 looking only to see whether the decisions of those courts are flawed by arbitrariness or are manifestly unreasonable.30 This aspect of the margin of appreciation seems to leave domestic courts to follow their established roles under domestic law. The third axis has recently assumed growing significance. Where superior national courts have directly applied Convention rights as part of their own national legal order, eg to balance Article 8 and Article 10, the ECtHR will generally respect the decisions they come to: Where … the superior national courts have analysed in a comprehensive and convincing manner the precise nature of the impugned restriction, on the basis of the relevant Convention case-law and principles drawn therefrom, this Court would need strong reasons to differ from the conclusion reached by those courts by substituting its own views for those of the national courts on a question of interpretation of domestic law.31

Despite the reference to interpretation of domestic law, this is different from application of the fourth-instance rule. In applying Convention rights, a national court is not more expert or better placed than the ECtHR. Rather, I suggest this development in the margin of appreciation is being driven by three factors. The first is the pressure imposed on the ECtHR by its heavy caseload. Proposals for far-reaching reform, to qualify the right of individual petition, have not been taken up, so the ECtHR has reacted at the level of articulation of doctrine to stand back more from direct substitution of decision-making by itself. This can be seen as an acknowledgement of its role as a supervising court, akin to a superior constitutional court within the European public order. As a superior constitutional court the ECtHR, especially as a court operating without any screening process in the form of permission

29 EHRR 493, paras 135–38; Parti-nationaliste Basque v France, ECHR 2007-II, para 46. However, the fourth-instance rule has been applied where complaints are made under Art 6 (right to a fair hearing) regarding the assessment of evidence by national courts: eg Koval v Ukraine, no 65550/01, judgment of 19 October 2006, para 115. 29  An issue of importance in relation to a number of Convention rights which depend upon action of national authorities being lawful or taken in accordance with the law: see, eg, SW v UK (1996) 21 EHRR 363. 30  See, eg, Anheuser-Busch Inc. v Portugal, no 73049/01, judgment of 11 October 2005, para 83; ­Garcia Ruiz v Spain (2001) 31 EHRR 22, GC, para 28; Beganovic v Croatia, no 46423/069, judgment of 26 June 2009, paras 78 and 85; Austin v UK (2012) 55 EHRR 14, GC, para 61 (regarding findings of fact by the domestic courts). 31  Roche v UK (2006) 42 EHRR 30, GC, para 120; MGN Limited v UK (2011) 53 EHRR 5, paras 144–50 and 155 (where the balancing exercise between competing Convention rights in Arts 8 and 10 had been undertaken by the national authorities in conformity with the criteria laid down in the Court’s case law, the Court ‘would require strong reasons to substitute its view’ for that of the majority in the House of Lords); Axel Springer AG v Germany (2012) 55 EHRR 6, GC, para 88; Von Hannover (No 2) v Germany (2012) 55 EHRR 15, GC, para 107; Palomo Sanchez v Spain (2012) 54 EHRR 24, para 57; Aksu v Turkey (2013) 56 EHRR 4, GC, paras 65–67.

Proportionality and the Margin of Appreciation: Strasbourg and London 187 to apply to it to focus its resources on cases of major importance,32 might be overwhelmed unless it could adopt a more hands-off, supervisory role in relation to the direct application of Convention rights than it did in the earlier stages of its operation. This withdrawal to a more supervisory role is underpinned by the second and third factors: the greater maturity of the ECtHR’s case law in terms of articulation of principles to bridge the gap between abstract Convention rights and application to concrete cases and the increasingly developed processes of acceptance of the ECtHR’s case law and enforcement of Convention rights at the national level. The first task of the ECtHR to articulate principles capable of winning acceptance by national courts is now giving way to a second phase, in which the ECtHR is able to adopt a more ‘hands-off ’ supervisory role even in relation to the direct application of Convention rights.33 There are also indications that the increasing ambit of the margin of appreciation along this axis is widening to cover not just decisions of superior national courts which display loyalty to the ECtHR’s case law and articulated principles but also measures taken by other national authorities, in particular national legislatures, which display similar loyalty. The standard formulation of the principle by the ECtHR refers to assessments made by the ‘national authorities’, not specifically national courts, though most cases have concerned assessments made by national courts and it is likely to be primarily in such cases that the ECtHR will find evidence of self-direction according to the ECtHR’s case law and will feel the greatest confidence in the quality of the assessment. However, in Animal Defenders ­International v UK the ECtHR attached ‘considerable weight’ to exacting reviews with reference to its case law in both the parliamentary process and the national courts as to how to strike the correct balance for Article 10 purposes in the regulation of broadcasting.34 It was significant that the case concerned a ‘general measure’ (rather than a prior restraint imposed on an individual act of expression), in relation to which the ECtHR had to assess the legislative choices available, where ‘[t]he quality of the parliamentary and judicial review of the necessity of the measure is of particular importance … including to the operation of the relevant 32  The admissibility screening process is not the same: it is simply a check whether an individual has a properly arguable case which should proceed to the merits stage. 33  There is an analogy with the approach adopted by the English Court of Appeal in relation to evaluative judgments in applying the law made by first instance courts: see, eg, Hadmor Productions v Hamilton [1983] AC 191; R (MH) v Secretary of State for the Home Department [2010] EWCA Civ 1112, [73]. Self-protective doctrines of this kind are important in achieving efficient use of the limited resources of the supervising court and minimising delays at that level. For a suggestive historical parallel with this arc of development of the influence of the ECtHR, compare the account in L Siedentop, Inventing the Individual: The Origins of Western Liberalism (Cambridge, MA, Harvard University Press, 2014) 226–29, of the assertion of papal/ecclesiastic jurisdiction in contest with secular jurisdiction in the course of the papal revolution in the 11th and 12th centuries, in which the availability of direct appeal to the pope reinforced the papacy’s claims to judicial authority over aspects of secular life, and when a degree of success had been achieved in terms of securing acceptance of these claims, and under the pressure of the flood of applications which grew as a result, the papacy introduced more graduated procedures under papal supervision. 34  Animal Defenders International v UK (2013) 57 EHRR 21, GC, see paras 113–17.

188 

Philip Sales

margin of appreciation’.35 This may make the processes of human rights scrutiny in Parliament, eg via the Joint Committee on Human Rights, particularly important,36 and can be regarded as a form of incentive for the legislature to take Convention rights into account when legislating.37 The implications of a widening margin of appreciation in Strasbourg along this axis for the approach domestic courts might take are unclear. In so far as it is applied in relation to decisions taken by superior domestic courts it supplies an incentive for the courts to respect and apply the case law and principles set out by the ECtHR, including the proportionality doctrine, at least in a general way. However, it appears to open up more space for the domestic courts to fashion their own detailed solutions to problems, and have them accepted by the ECtHR, provided they approach matters explicitly using the perspective of the ECHR and principles articulated by the ECtHR. In this way, the proportionality doctrine in Strasbourg appears to be sliding from one involving direct primary assessment of the proportionality of outcomes by the ECtHR to a more supervisory jurisdiction, at least in some contexts. In so far as a widening margin of appreciation along axis (iii) is applied in relation to measures taken by the legislature or the executive, it may afford a basis on which the domestic courts could accord greater respect to the judgments made by them if they have explicitly had regard to the Convention rights and principles drawn from the jurisprudence of the ECtHR when they act.38 On the other hand, there may be scope for the domestic courts to seek to fill the gap opened up by the withdrawal of the ECtHR from more intensive review with their own, more particularistic interpretations of Convention rights or by developing more intensive domestic public law doctrines. Some evidence that this is happening is referred to below. Adjustments of the margin of appreciation along each of the three axes involve striking a balance between procedural values and control of substantive outcomes. On axis (i), for example, the ECtHR is prepared to check to see if there has been effective debate in a democratic context, and if there has been, then to afford a wider margin of appreciation as to the acceptability of the result.39 But it appears that along axis (iii) the balance is now being struck in a different way 35  Ibid, paras 106–11. cf MGN Limited v UK (n 31) paras 198–220 (violation of Art 10 by reason of the Conditional Fee Agreement system in relation to legal costs; the flaws in the scheme, which were highlighted by later public reviews, meant that the state’s broad margin of appreciation was exceeded). 36  See the discussion in M Hunt, H Hooper and P Yowell (eds), Parliaments and Human Rights: Redressing the Democratic Deficit (Oxford, Hart Publishing, 2015). 37  Sales (n 7) 232–33. 38  Thereby reinforcing the incentive for democratic bodies to act in ways which explicitly take the human rights dimension into account: see, eg, R (Tigere) v Secretary of State for Business, Innovation and Skills [2015] UKSC 57, [32]; SS (Congo) v Secretary of State for the Home Department [2015] EWCA Civ 387, [13]–[17]. 39  James v UK (1986) 8 EHRR 123; cf Hirst v UK (No 2) (2006) 42 EHRR 41, the prisoner voting case, in which an absence of active review by Parliament of the long-standing ban on prisoner voting was a factor in the finding of violation of Art 3 of Protocol 1. This aspect of the Strasbourg jurisprudence potentially causes a difficulty for national courts bound by Art 9 of the Bill of Rights not to question proceedings in Parliament: see Wilson v First County Trust (No 2) [2004] 1 AC 816; but cf the divergent

Proportionality and the Margin of Appreciation: Strasbourg and London 189 than before, as under certain conditions the ECtHR is withdrawing from direct application of proportionality doctrine itself and choosing to use instead a more supervisory proportionality standard which involves review of judgments about proportionality made elsewhere, at national level. As the intensity of intervention by the ECtHR moves from direct judgments about what count as proportionate outcomes to more of a review function in respect of decisions made at national level, the domestic courts have to begin to consider whether to move to replace the ECtHR’s more direct role by their own doctrines or to follow it in allowing a greater margin of discretion to the legislature and executive where they have acted by taking Convention rights into account themselves. Assessments along the three axes may overlap and reinforce each other. For example, the ECtHR will respect the legislature’s judgment as to what is in the public interest unless manifestly without reasonable foundation,40 which reflects both democratic principle (axis (i)) and the legislature’s being better placed than the court to identify the public interest in the national context41 (axis (ii)); and the wide margin of appreciation allowed may be further supported if the legislature has acted in the context of a policy assessment informed by the ECtHR’s case law (axis (iii)). On the other hand, it may be that the ECtHR will not regard it as appropriate to defer to the national authorities’ assessment on all the elements inherent in an overall analysis of compliance with a Convention right; for example, it may prefer to make its own assessment of whether the quality of the relevant national ‘law’ is satisfactory for the purposes of Article 8(2) or Article 10(2). Further, indications in favour of a wide margin of appreciation along one or more of these axes do not necessarily have determinative impact; they may be counterbalanced by other factors tending to narrow the margin to be accorded, such as where the contracting state has interfered in a direct way with the core of an especially important Convention right.

III.  Proportionality in London The ECHR system supervised by the ECtHR is a dynamic legal system which is brought into a relationship with the English legal system, which has its own views among the justices in R (Nicklinson) v Ministry of Justice [2014] UKSC 38 and A Kavanagh, ‘Proportionality and Parliamentary Debates: Exploring Some Forbidden Territory’ (2014) 34 OJLS 443. For discussion of the other ways in which domestic courts balance procedural and substantive outcome factors in their approach to Convention rights, see A Kavanagh, ‘Reasoning About Proportionality Under the Human Rights Act 1998: Outcomes, Substance and Procedure’ (2014) 130 LQR 235. 40 

MGN Limited v UK (n 31) para 200. National authorities ‘are in principle better placed than the international judge’ to make assessments drawing on their ‘direct knowledge of their society and its needs’: see, eg, Hutten-Czapska v Poland ECHR 2006-VIII, para 165; see also A v UK (2009) 49 EHRR 29, regarding the national authorities’ assessment whether there was a public emergency threatening the life of the nation. 41 

190 

Philip Sales

­ rocesses of development, both via the Human Rights Act 1998 and in a more p indirect way, as a source of inspiration for domestic legal development. Changes in each system as they develop may produce effects in the other. There can be little doubt that application of the ECtHR’s case law, in addition to their working with EU law, has familiarised domestic courts with using the proportionality methodology. There are four particularly notable features of domestic English law at the moment which arise at the point of intersection between the ECHR and domestic legal systems. First, the domestic courts will not necessarily be deterred from applying their own interpretation of Convention rights at the national level just because a measure taken by the legislature would, if assessed by the ECtHR, be found to fall within the United Kingdom’s margin of appreciation.42 But it is unclear how far the national courts might be prepared to go down this controversial route. As Lord Justice Elias has written extra-judicially, in adopting this approach ‘[t]he judges are not simply “bringing rights home” as the White Paper introducing [the Human Rights Act] said was its purpose, but claiming the power to create home rights’.43 Even if the margin of appreciation allowed by the ECtHR as an international court applies to a state without, in itself, necessarily differentiating which organ in the state may exercise the discretionary power of choice thereby accorded by the ECHR, nonetheless within the UK’s own constitutional order there is a strong democratic imperative to respect the choices made by Parliament in legislation and by elected ministers accountable to Parliament. On the other hand, the willingness of the ECtHR to shift to more of a review function along axis (iii) above may encourage the Supreme Court further down this road. The fact that the Supreme Court has moved in this direction in the name of applying Convention rights indicates that it is likely that the proportionality doctrine will be applied even where the result may be different from that likely to be given by the ECtHR. Secondly, the Supreme Court has displayed a readiness to push back against Strasbourg jurisprudence, in particular at Chamber level, in a process of intercourt dialogue, where it is not satisfied that the ECtHR has fully understood the domestic legal context.44 This is likely to be most persuasive where the domestic courts can show that they have applied the structures of reasoning and, where appropriate, the proportionality methodology with which the ECtHR is familiar. However, the Supreme Court has accepted that there are limits to this process of dialogue, in particular where the Grand Chamber of the ECtHR has fully understood the national position and has made a clear decision as a matter of authoritative judgment under the ECHR.45 Nonetheless, the very process of 42 

In re G (Adoption: Unmarried Couple) [2009] 1 AC 173; Nicklinson [2014] UKSC 38, [63]. Justice Elias, ‘Are Judges Becoming too Political?’ (2014) 1 Cambridge Journal of International and Comparative Law 1, 19. 44  The dialogue between the domestic courts and the ECtHR in the Horncastle litigation is the prime example: R v Horncastle [2009] UKSC 14 leading to the Grand Chamber judgment of 15 December 2011 in Al-Khawaja v UK. 45  See, eg, R (Chester) v Secretary of State for Justice [2014] AC 271. 43  Lord

Proportionality and the Margin of Appreciation: Strasbourg and London 191 dialogue encourages the domestic courts to use the common proportionality approach. Thirdly, as widely noted in the literature, the Supreme Court has chosen to become more active in highlighting and developing conceptions of domestic constitutional principles and fundamental rights, both as an aspect of the principle of legality in statutory interpretation and in providing for more intensive domestic standards of judicial review in certain contexts.46 Where the domestic assessment of the context calls for heightened review, it is essentially the model of proportionality review which will be used.47 This means that common law judicial review is in these contexts capable of incorporating the same methodology and of giving the same answer as proportionality review under the ECHR or EU law. The assimilation of the methodology across these areas is welcome, in terms of simplifying the analysis which a domestic court may be called upon to carry out. Of course, a great deal may depend on identifying what qualifies as a relevant context for proportionality review in domestic law, in relation to which further guidance may be expected. All these developments have combined to embed and extend the application of the proportionality doctrine by the domestic courts. The other notable feature in leading domestic cases is the way in which the fourth limb of the proportionality template (what has been called proportionality in the strict sense) has tended to become the most powerful aspect of the doctrine, in the way it is used by the Supreme Court. This tendency is highlighted by Elias in his extra-judicial discussion48 of the leading cases of Lord Carlile of Berriew v Secretary of State for the Home Department,49 Nicklinson50 and R (SG) v Secretary of State for the Home Department.51 This emphasis on the fourth limb is something of a recent departure, since it was not originally included in the classic domestic statement of the

46  See, eg Bank Mellat (n 2); R (Osborne) v Parole Board [2013] UKSC 61; R (Faulkner) v Secretary of State for Justice [2013] UKSC 23; Kennedy v The Charity Commission [2014] UKSC 20; A v BBC [2014] UKSC 25; Pham v Secretary of State for the Home Department [2015] UKSC 19. For discussion, see, eg, R. Clayton, ‘The Empire Strikes Back: Common Law Rights and the Human Rights Act’ [2015] Public Law 3; R Masterman and S Wheatle, ‘A Common law Resurgence in Rights Protection?’ [2015] EHRLR 57. 47  Kennedy (n 46) [55]–[56] (Lord Mance); Pham (n 46) [94]–[96] (Lord Mance), [108]–[110] (Lord Sumption), [116]–[119] (Lord Reed). Although in R (Lumsdon) v Legal Services Board [2015] UKSC 40 a distinction was drawn between the proportionality analysis in domestic law (and the Human Rights Act) and in EU law, the substance appears closely similar: [26] and [33]–[34]. See also the formulation of the doctrine as operated in EU law, in section 5 of chapter 10 by AG Juliane Kokott and Christoph Sobotta in this volume. 48  Elias (n 43). 49  Lord Carlile of Berriew v Secretary of State for the Home Department [2014] UKSC 60, concerning a ban on entry of a controversial opponent of the Iranian regime, where this was judged necessary to prevent serious adverse reactions by Iran against British interests. 50  Nicklinson [2014] UKSC 38, concerning the compatibility with the ECHR of the statutory ban on assisting suicide. 51  R (SG) v Secretary of State for the Home Department [2015] UKSC 16, concerning the ­introduction of the benefit cap.

192 

Philip Sales

proportionality principle.52 It is also questionable whether the emphasis now given to it in the domestic cases finds much direct support in the Strasbourg case law. The ECtHR tends not to isolate the fourth limb in the same way when it applies the proportionality principle, although there are undoubtedly cases in which it has found measures to be disproportionate which can be fitted within this rubric. The willingness of the Supreme Court to resort to the fourth limb of the proportionality test poses problems. It is the part of the test which is least susceptible to rigorous legal-style analysis and most open to application of wide judgments by individual judges about what it is moral or legitimate for a state to do, and also judgments about what it is legitimate for judges to do. It is the part of the test where the judges may most clearly be making judgments, which could fairly be described as political, in areas which have moved well beyond those in which there is any identifiable community consensus, or even an identifiable consensus amongst judges and lawyers regarding the standards to be applied. This could have negative implications for both democratic principle and rule of law values, if the judges seek to by-pass democratic decision-making procedures for dealing with fraught and sensitive political and moral disagreements and do so without any identifiable normative consensus to guide them and which would (in the manner explained by Fallon) allow ordinary citizens to know what to expect. There have been substantial differences of view within the Supreme Court in each leading case reviewed by Elias, as well as departures from the reasoning of the judges in the lower courts. As Elias notes, although in his view ‘it is critical that the courts should be acutely aware of the significance of democratic accountability in a political democracy, of the limits of their own expertise, and of the narrow and potentially distorting focus of adjudication when compared with legislation’, nevertheless ‘[t]he cases demonstrate that some judges are more influenced by these considerations than others’.53 He observes that ‘the outcome of a case may well depend upon the constitution of the court’, and that the perception that judges are entering into the political arena may increase the pressure for political scrutiny of their appointments, concluding: ‘it would indeed be something of a paradox if the judicialisation of government were to lead to the politicisation of the judiciary’.54 The overall picture, from the domestic perspective, is that the Supreme Court is becoming more willing to engage in proportionality review at the same time as the ECtHR may be becoming more willing to leave the primary proportionality judgment to be made by the national authorities. In certain respects, the two movements are perhaps complementary. An extension of a unified proportionality methodology across cases involving Convention rights, EU rights and domestic

52 In De Freitas v Permanent Secretary of Ministry of Agriculture, Fisheries, Lands and Housing [1999] 1 AC 69, 80 (Lord Clyde). Elias (n 43) 8, traces the new emphasis on the fourth element to Huang v Secretary of State for the Home Department [2007] UKHL 11. 53  Elias (n 43) 25. A further recent example is R (Tigere) v Secretary of State for Business, Innovation and Skills [2015] UKSC 57. 54  Elias (n 43) 26.

Proportionality and the Margin of Appreciation: Strasbourg and London 193 fundamental rights is welcome, in the interests of simplicity and transparency of legal analysis. On the other hand, it is desirable for proportionality analysis to proceed hand in hand with a careful assessment of how far it is appropriate to intrude upon choices which have been made in central areas of governmental and legislative responsibility and how far it may detract from, rather than promote, rule of law values.

194 

12 Origins and Presentation of the Proportionality Principle in French Case Law YOAN SANCHEZ

I. Introduction A. The Origins of the Principle: The Case Law of the Council of State As a tool to supervise and control the business of public authorities, ­proportionality may be traced back to the early part of the twentieth century. Benjamin is often cited as the first example.1 In this case from 1933, a conference was organised in the city of Nevers by a journalist and public speaker named René Benjamin. Because of the controversial content of Benjamin’s writings, a number of trade unions threatened to organise public demonstrations against his visit. To avoid any risk to public order, the mayor decided to ban the conference. This decision was contested before the Council of State. The Council stated clearly that the public authority had to reconcile the protection of public order and the protection of rights and freedoms. In this case, the decision did not reconcile these two elements. The threat to public order was not so great that it was impossible for the mayor both to allow the conference and protect public order by relying on the police. Before this case, the requirement for public authorities to reconcile the protection of public order, on the one hand, and rights and freedoms, on the other, had been mentioned by the Council of State. In the Abbé Olivier in 1909 a mayor took a decision to forbid all religious events in his town.2 The judge declared that the mayor might restrict the local habits and traditions ‘only when strictly required by

1  2 

Benjamin [1933] rec 541 (Conseil d’État). Abbé Olivier [1909] rec 181 (Conseil d’État).

196 

Yoan Sanchez

the need to maintain public order’. As in Benjamin, we clearly see in Abbé Olivier the necessity for the public authority to reconcile public order and freedoms. This reconcilation is at the very heart of the proportionality control as it precludes excessively burdensome decisions.

B.  Use of the Proportionality Control by Other Courts This chapter will focus on the Council of State. But proportionality is also used by the Constitutional Council and the Cour de cassation. The former verifies whether measures provided by a statute are proportionate to the legitimate aim that Parliament pursues.3 For example, the Constitutional Council has recently declared that powers given to police forces because of the declaration of a state of emergency in France after the events of N ­ ovember 2015 mean that putting an individual under house arrest did not constitute a ‘­disproportionate infringement of the freedom of movement’.4 In this respect, the ­Constitutional Council has to protect constitutional rights and liberties from the most obvious infringement by Parliament. The Cour de cassation has to reconcile different private interests which might be contradictory.5 For the Cour de cassation, the proportionality control is mostly used in cases involving interests which are considered to be equal. This is because cases involve equal individuals. The reconciliation is made between horizontal interests, while by contrast for the Council of State the reconciliation is between vertical interests: interests of the state against interests of individuals.

C. Presentation of the Proportionality Principle in the Case Law of the Council of State Two points have to be noted when the proportionality principle is involved. The first one is that administrative judges do not apply it explicitly. The second concerns the aspect of the decision which is controlled by way of proportionality. This last development will help to demonstrate what proportionality is.

3  G Chetard, ‘La proportionnalité de la répression dans la jurisprudence du Conseil constitutionnel’ [2013] Revue de Science Criminelle et de Droit Pénal 51. 4  M. Cédric D [2015] no 2015-527 QPC (Conseil constitutionnel). 5  See eg J-B Seube, ‘Le contrôle de proportionnalité exercé par le juge judiciaire: Présentation générale’ (2009) 46 Les Petites Affiches 86; G Chetard, ‘Cassation pour “méconnaissance du principe de proportionnalité”’ [2015] Actualité Juridique Pénale 487; O Renaudie, ‘Conseil constitutionnel, liberté d’aller et venir et protection des personnes souffrant de troubles mentaux’ [2015] Revue de droit sanitaire et social 963; J-P Marguenaud, ‘La mise en œuvre du principe de “proportionnalité privatisée” par la première chambre civile de la Cour de cassation’ [2015] Revue Trimestrielle de Droit Civil 825.

Origins and Presentation of the Proportionality Principle in French Case Law  197  

(i)  Non-Explicit Use of Proportionality by Administrative Judges Even if proportionality control has a long history in France, it is not an explicit control. As revealed by Benjamin and Abbé Olivier, administrative judges seldom refer to the words ‘proportionate’ or ‘proportionality’. Nor are the three classical steps of proportionality control from the German formulation (adequacy, necessity and proportionality in the strict sense) used. So it is through the idea of reconciliation, or by the will to forbid excessively burdensome decisions, that proportionality control may be identified. In a previous article, I mentioned an evolution in the formulation of proportionality control in the administrative courts’ case law.6 As a matter of fact, the courts are less reluctant nowadays to use a threestep formulation of the test. For example by way of interim decisions delivered in the case D ­ ieudonné in 2014, the Council of State was ready to say that ‘limitations to the exercise of fundamental freedoms for the protection of public order must be necessary, adapted and proportionate’.7 Even though the control is not applied following the three steps, the formulation of the test is clearer. Sometimes, the Council of State is lies in between. A good example is the control exercised over administrative sanctions. In La Poste in 2015, the Council summarised the control by stating that judges have to verify ‘if the sanction chosen is proportionate to the seriousness of the faults’.8 If the word ‘proportionate’ is apparent here, the three steps are not.9 The proportionality control is clearer than in Benjamin and Abbé Olivier but less than in Dieudonné. One may briefly note that since 2008 the Constitutional Council has clearly applied the proportionality principle.10 It has structured its control in line with the three steps in one decision. The constitutional judge has thus verified that the bill was appropriate, necessary and proportionate in the strict sense. It is a very clear application of the proportionality principle that the Council of State should take as an example.

(ii)  To What Aspect of the Decision Is Proportionality Applied? This point is important. It is necessary clearly to distinguish two different ­questions: the question of the ‘nature’ of the control and the question of the intensity of the scrutiny. The first issue is related to elements of the decision controlled by way of the proportionality test. The second is about the degree of rigour with which the test is applied by judges. In short, it is a distinction between ‘what is controlled?’ and ‘how is it controlled?’. These two questions have often been ­confused

6  Y Sanchez, ‘Proportionality in French administrative law’ in S Ranchordas and B de Waard (eds), The Judge and the Proportionate Use of Discretion. A Comparative Study (London, Routledge, 2015) 43. 7  Dieudonné [2014] rec 1 (Conseil d’État). 8  La Poste [2015] rec 64 (Conseil d’État). 9  The three-step formulation is still used concerning police decisions. See Commune de Cournon d’Auvergnes [2015] rec 55 (Conseil d’État). 10  Loi relative à la rétention de sûreté [2008] no 2008-562 DC (Conseil constitutionnel).

198 

Yoan Sanchez

in France. It is not the purpose here to criticise this confusion.11 But to be as clear as possible, it is necessary to explain ‘what’ is controlled when we talk about proportionality. This point may usefully be discussed by taking the field of sanctions as an ­example. Two different aspects of a decision to impose a sanction have to be distinguished. The first one is the requirement of a fault. Without fault, no sanction is possible. So the court verifies if the facts (mostly the behaviour of the individual subject to the sanction) may be qualified as a fault. This stage is named ‘control of the qualification of facts’. If factual elements are qualified as a fault, the public authority may decide on a sanction. In the second step, administrative judges control the choice of the sanction. The control of proportionality is applied here. The question is the following: is the sanction proportionate to the seriousness of the fault? But if the question is simple, the answer may not be. The Council of State has a two-degree approach to this question. In some circumstances, the proportionality control is applied solely to quash manifest error. It means that the sanction will be illegal only if it is manifestly disproportionate. So a kind of ‘power to err’ is granted to the public authority. In some other circumstances, proportionality control is not limited as such and judges quash every error in the determination of the sanction. Here, the sanction must be strictly proportionate to the fault: that means that there is no ‘power to err’. This point will be developed below. This presentation is oversimplistic. Sometimes the words of the judge seem to reveal a limited scrutiny, whereas it is an intensive one. In the same way, proportionality control may be used according to a formula that reflects an intensive approach while in fact it is a limited one. So it is not exactly a satisfying presentation. But it is a very clear and helpful one for the current topic. It means that proportionality is about the control of a decision taken as an answer to a situation. The role of administrative judges in this respect is to verify that the public authority has not exceeded what is necessary to solve the ‘problem’. But if the control of the existence of this ‘problem’ and the control of the proportionality of the ‘solution’ have to be distinguished, there is a close link between them. In fact, the more important the ‘problem’, the more important is the solution’ authorised. It allows the reconciliation of the realisation of the legitimate goal pursued by the public authority (to solve the problem) and the protection of rights and liberties affected by the decision. So the role of proportionality is to permit at the same time the realisation of the goal pursued by public authorities and the protection of liberties. This does not mean that liberties always must prevail because it may be legitimate for public authorities to infringe upon them. But proportionality is a tool allowing judges to draw a line that the decision-maker cannot cross. This frontier is drawn according to a case-by-case approach since it depends on the importance of the goal pursued. The importance of the infringement of liberties has to be considered in the light of the importance of the legitimate aim. It is in that sense that proportionality prevents excessively burdensome 11 

For a study of this question, see Sanchez (n 6).

Origins and Presentation of the Proportionality Principle in French Case Law  199   decisions, ie those that infringe liberties more than can be permitted by reference to the importance of the aim.

II.  The Impact of EU Law on the Interpretation of the Proportionality Principle Determining the influence of EU law on national application of the proportionality principle is complex. Decisions by the European Court of Justice (ECJ) are indeed considered by administrative judges, but not always in the same way. The impact of EU law is for this reason highly variable. This conclusion is reinforced by the limited field of influence of EU law. Administrative judges, like the ECJ itself, have made clear that the European conception of proportionality is applicable only in matters ruled by EU law. Despite this affirmation, some national cases may reveal an influence outside the direct application of EU law. Before entering into these considerations, a note of precision is necessary. It is not the current purpose to explain how the proportionality principle is employed by the ECJ. Some elements will be considered, but only to explain how the principle influences national law. I shall not undertake a systematic study of proportionality control in ECJ case law.

A. EU Law Has a Variable Impact on the Proportionality Principle in French Law The proportionality principle is employed in several fields by the ECJ. The most relevant for current purposes is the use of the proportionality principle to control decisions taken by the state that restrict rights and liberties protected by the ­European treaties. To save space I will focus only on decisions restricting economic liberties. It is up to the national judges to control these decisions in the first instance. As such, the interpretation of proportionality by the ECJ has a clear impact.

(i) Consideration of the ECJ’s Interpretation of the Proportionality Principle by the Administrative Judges A lot of decisions taken by administrative judges rely on the proportionality principle as it is understood by the ECJ. For example, in a case concerning the question of the conformity with EU law of the monopoly of the Française des Jeux on gambling, the Council of State relied on three decisions of the ECJ to declare that such a monopoly has to be ‘justified by overriding considerations of public interest, … appropriate to guarantee the

200 

Yoan Sanchez

realisation of the aim pursued and … not beyond what is necessary to attain it’.12 And in this case, the monopoly was found not to infringe EU law because it ‘does not affect in a disproportionate way, with regard to the aim pursued, freedom to provide services and freedom of establishment’. In the same way, in the case Zeturf, about Pari Mutuel Urbain, which had a monopoly concerning bets on horses, the Council of State referred to two cases of the ECJ and stated that decisions which infringed the freedom of services are taken in accordance with EU law if they ‘do not go beyond what is necessary to attain aims pursued, if they are appropriate to guarantee their realisation and if they are executed by a coherent and systematic politic’.13 This shows that when the control of proportionality is applied in an EU context, the formulation of the test is not the same as in a purely national perspective. First, there is a clear influence of the ECJ here. The Council of State takes elements of the European case law to verify the proportionality of national decisions. Second, the control of administrative judges is much more developed and structured. It has been shown, eg in Benjamin, that the proportionality principle is expressed in one or two sentence(s).14 In an EU context, the Council of State is far less laconic. In the case concerning the Française des Jeux, the control is made in eight paragraphs; in the other case cited, it takes five paragraphs.15 It is even clearer when we look at the conclusions of the commissaires du ­Gouvernement. Concerning Zeturf, commissaire Jean-Philippe Thiellay relied heavily on the control made by the ECJ on this matter.16 He identified four points in the reasoning of the Court: the restrictions must be provided for in a national regulation; the national regulation must not create discrimination based on nationality; it is not the case that because a corporation is authorised in one Member State to organise bets that it has to be authorised in all others; and finally the regulation is authorised if it is justified by overriding considerations of public interest, necessary and applied in a coherent and systematic way. After a study of the application of all these elements by the ECJ, the commissaire applied it to the case in order to propose a solution. One point may reinforce the idea of an important influence of EU law on proportionality. If national judges are clearly competent to control the conformity of national decisions with EU law in the first place, the ECJ gives the interpretation of European norms.17 This means two things. First, when administrative judges have difficulties in determining if a national decision is compatible with EU law,

12 

Confédération française des professionnels en jeux automatiques [2000] rec 173 (Conseil d’État). Société Zeturf Limited [2008] Actualité Juridique Droit Administratif 961 (Conseil d’État). Benjamin (n 1). 15  Paragraphs are named considérant in French administrative law because each of them begins with the word considérant which can be translated by ‘considering’. Each considérant is constituted of several sentences. 16  J-P Thiellay, ‘Le monopole du Pari Mutuel Urbain et la liberté de prestation des services’ [2008] Revue Française de Droit Administratif 1043. 17  Case C-67/98 Zenatti [1999] ECR I-7289, para 26. 13  14 

Origins and Presentation of the Proportionality Principle in French Case Law  201   they may refer a question to the Court. The answer helps national judges to understand the control that has to be applied. This was done in Zeturf. Second, when national judges apply the proportionality principle they make several references to the appreciation of the ECJ. For example, a measure restricting free movement of goods has to pursue a ‘general interest’. But national judges do not give their own definition. They use what is considered by the ECJ or the treaties as a ‘general interest’. In the case Société Serc Fun Radio, concerning the implementation of obligations for radios to broadcast a determined percentage of French songs, the Council of State made clear that ‘the general interest linked to the valuation of [cultural] heritage is an overriding consideration, in the sense given by the ECJ, justifiying a limitation to the free movement of goods and freedom to provide services’.18 So even if national judges are those who control the proportionality of national measures, they do this under the influence of the ECJ.

(ii)  A Consideration Which Is Not Always Clear The influence of EU law on proportionality is clear, but sometimes it is also not completely satisfying. Two arguments will be made. First, the administrative judges do not always fully receive the ECJ’s views on proportionality. Second, sometimes, even if a clear answer is given to a problem by the ECJ, the Council of State decides not to follow it. The first argument is thus about the incomplete use by judges of the proportionality principle as expressed by the ECJ. In Française des Jeux, the reasoning of the Council of State, finally declaring that the monopoly is not a violation of EU law, was mostly based on the examination of the general interest justifiying the monopoly. The proportionality test, in itself, was stated peremptorily. The Council did not even verify the necessity or the proportionality in a strict sense. In Société Serc Fun Radio, the commissaire du Gouvernement Valérie Hubac referred to three steps in the control of proportionality, even if she combined the first two.19 The commissaire said that to be lawful a regulation needs to ‘permit an efficient protection of the general interest’; it is necessary that the national authority ‘does not have any other mean equally efficient but less restrictive for free movement of goods and services’; and the regulation has to be ‘proportionate to the aim pursued’. But the Council of State did not comply with this methodology and strangely stated that the regulation ‘is not disproportionate to the aim pursued since the percentage permits to guarantee its realisation and does not go beyond what is necessary to attain it’.20 This laconic statement fails to respect the three steps proposed by the commissaire du Gouvernement under the influence of the ECJ.

18 

Société Serc Fun Radio [1998] rec 138 (Conseil d’État). S Hubac, ‘La conformité au droit communautaire de l’obligation de diffuser un pourcentage minimal de chansons d’expression française’ [1999] Revue Française de Droit Administratif 194. 20  Société Serc Fun Radio (n 18). 19 

202 

Yoan Sanchez

But, to be fair, the ECJ is not always very clear about the proportionality control either. The formulation may change: the Court makes references to one, two or three element(s).21 This may be an explanation for the changing expression of the proportionality principle by the Council of State. In any case, however, the control of administrative judges is still too laconic when applying EU law. So even if, as observed above, the influence of EU law generally leads to a more structured control, it is not always true. The Council of State has still some efforts to make. Concerning the second argument, the fact that the ECJ is not always clear in the formulation of its control cannot be invoked to justify the solution of administrative judges. In Quintanel, in 2015, the Council of State concluded there was no discrimination in the system of pensions for civil servants.22 This system provides advantages for civil servants who have three children and who stop working in order to bring them up.23 Under the influence of EU law, the legislation had already been changed in 2003. It was previously restricted to women and so unlawfully discriminated between men and women. Since 2003, without consideration of the sex of the individual, a civil servant may invoke an interruption in their work to justify the right to some advantages. The ECJ was given the opportunity to rule on the matter and declared the new system unlawful as it still favours women and it found that this discrimination could not be justified by the realisation of the aim pursued, which was to compensate disadvantages that follow the interruption of work in order to bring up children. This part of the decision of the Court is not binding as it does not concern the interpretation of treaties. The Council of State had to answer the same question in Quintanel. The decision is clear: the system as it follows from the statute of 2003 is not a violation of EU law. For the Council, the discrimination is justified by a legitimate aim. Furthermore, the discrimination permits the realisation of this aim and is necessary. One identifies here two steps of the proportionality principle: adequacy and necessity. The Council of State provided a lot of factors to demonstrate the adequacy of the system, but none concerning its necessity. The Council has taken a clear position against the appreciation made by the ECJ. It is even clearer in the conclusions of the rapporteur public, Bertrand Dacosta, who had proposed to the judge, after a long presentation of the arguments of the ECJ, to ‘fully exercise its power as the judge responsible for the application of European Union law’ which means to ‘determine if the discrimination revealed by the Court is legitimate’.24 The French

21  D Simon, ‘Le contrôle de proportionnalité exercé par la Cour de justice des Communautés européennes’ (2009) 46 Les Petites Affiches 17; M Rifat Tinc, ‘Le contrôle de proportionnalité des mesures nationales restrictives des échanges. Un instrument de légalité au service de l’intégration’ [2010] Revue du Droit de l’Union Européenne 791. 22  Quintanel [2015] Actualité Juridique Droit Administratif 1761 (Conseil d’État). 23  For example, the right to retire early with pecuniary assistance. 24  B Dacosta, ‘Régime des pensions, égalité des sexes et droit de l’Union européenne’ [2015] Revue Française de Droit Administratif 550.

Origins and Presentation of the Proportionality Principle in French Case Law  203   judges have clearly decided to depart from the decision of the ECJ. This appears to be an obvious limitation on the influence of the ECJ on the national interpretation of the proportionality principle.

B.  EU Law Has a Limited Field of Influence The impact of EU law is real but, at the same time, limited. EU law applies only within the scope of application of the treaties. When a decision does not concern such matters, it is not possible to invoke EU law. In that sense, the proportionality principle can only be influenced by EU law when it is relevant to the case. This is not so when the situation is purely internal and does not affect individuals from other Member States.25 This also appears clearly in the Charter of Fundamental Rights of the European Union. It is stated in Article 51, which stipulates that this Charter is applicable ‘to the Member States only when they are implementing Union law’. However, it is possible to identify the influence of EU law outside its field of application. In Société Compagnie des bâteaux-mouches, the Council of State imposed a requirement to respect competition law in the case of all policy ­decisions.26 This field is not covered by EU law as the ECJ has stated that policy decisions do not necessarily have to respect competition law.27 So there was no obligation for the Council to take into account EU law. But, for the Council, the decision (involving the calculation of the minimum number of sailors with regard to the maximum capacity of the boat) was ‘necessary and proportionate’ because it ensures the objective of the security of passengers and no other means can. The formulation of the test is not an usual one in the case law of the Council. In this respect, it reveals the influence of EU law. A similar formulation is employed in RATP in 2012, which does not concern a policy decision.28 The Council of State made clear that to be lawful a decision which limits economic activies has to be justified by a general interest and must be proportionate. We may identify here the classical steps of the ECJ, even if the formulation is not exactly the same: an overriding consideration of public interest and a proportionality control. Even outside the field of the direct influence of EU law, the Council of State seems ready to accept some elements of the control of the ECJ. But in this situation, it has much more discretion to interpret each step as judges do not have to follow the ECJ’s point of view strictly.

25  Case C-212/06 Government of the French Community and Walloon Government v Flemish Government [2008] ECR I-1683, paras 32–42. 26  Société Compagnie des bâteaux-mouches [2009] rec 201 (Conseil d’État). 27  Case C-202/04 Cipolla [2006] ECR I-11421. 28  RATP [2012] rec 231 (Conseil d’État).

204 

Yoan Sanchez

III.  The Impact of the European Convention of Human Rights (ECHR) on the Interpretation of the Proportionality Principle A.  The ECHR has No Impact on the Formulation of the Control If one can identify an impact of EU law on the formulation of the proportionality control, the same conclusion cannot be reached concerning the ECHR. It is not the current aim to study the control exercised by the European Court of Human Rights (ECtHR). The aim is just to state briefly how this control is organised. The ECtHR applies a very structured control to any decision taken by a state which infringes a qualified right protected by the Convention. It is a three-step control. The Court verifies if the interference with the right was ‘prescribed by law’, if it pursues a legitimate aim and if it is ‘necessary in a democratic society’.29 In order to control this third aspect, the Court checks if there is a ‘pressing social need and, in particular, that is proportionate to the legitimate aim pursued’.30 So proportionality appears in this last step. For the ECtHR, a contracting state may not take a disproportionate or excessive decision. It has to respect the values of a democratic society, namely ‘pluralism, tolerance and broad-mindedness’. This structure cannot be found in decisions taken by administrative judges. When the ECHR is applied, and the proportionality of the decision is verified, it is not in this structured way. For example in Belgacem, the Council of State determined whether a decision to pronounce the deportation of an individual infringed Article 8.31 It was considered to be an infringement of this right. But the Council was very laconic. It stated that the individual had lived in France since his birth and had no particular link with his country of nationality; his family was in France and he had to take care of them; and finally even if he had been previously condemned for theft he had changed his behaviour. From all of this, the Council conclued that ‘the deportation taken against M Belgacem has, in consideration of the gravity of the infringement of his family life, exceeded what was necessary to the defence of public order’. The decision is disproportionate and it was thus quashed. But neither the structured control nor the word ‘proportionate’ are used by the Council. If one looks at the conclusion of the commissaire du Gouvernement Ronny Abraham in that case one can find references to the control of the ECtHR.32 But they are not very detailed. The commissaire gives no explanation

29 

Sunday Times v The United Kingdom, Series A no 30 (1979) no 6538/74 (ECtHR) §45 Pretty v The United Kingdom, Reports of Judgments and Decisions 2002-III (2002) no 2346/02 (ECtHR) §70. 31  Belgacem [1991] rec 152 (Conseil d’État). 32 ibid. 30 

Origins and Presentation of the Proportionality Principle in French Case Law  205   about the methodology of the ECtHR. It was solely made clear that the case law of the European Court required interference in the right to be necessary, ie not out of proportion to legitimate aims pursued. If one may find sometimes the word ‘disproportionate’ in the test applied by the Council,33 there is no evolution towards a control of proportionality similar to the control of the ECtHR.34 Concerning the formulation of the test, the ECHR seems to have no impact. But that does not mean that there is no impact at all on proportionality control from a French perspective. The concern now is to examine the field of proportionality control as it was influenced by the ECHR. In a later section it will also be shown how the ECHR has led to an increase in the intensity of the proportionality control.

B.  The ECHR Has an Impact on the Field of the Control Despite the lack of influence on the formulation of the test, due to the strikingly laconic approach of administrative judges in France, the ECHR is an important instrument permitting protection of rights and liberties in the national system. Belgacem is relevant again here. It represents the first time that the Council of State agreed to review a police decision concerning a foreigner in the light of the ECHR. This evolution has permitted the Council to control the proportionality of the decision. Indeed, in Touami Ben Abdeslem, in 1980, the Council had refused the application of Article 8 concerning a decision of the same nature as that in Belgacem.35 The proportionality test had not been applied. One author noted that on this matter the influence of the ECHR was very small because there were just a few cases in which a deportation had been set aside by the European Commission on Human Rights.36 So in Touami Ben Abdeslem, the Council had solely verified that the behaviour of the individual was correctly qualified as a threat to public order. For the judges here, it was sufficient to justifiy the deportation. This is not a proportionality control, but a control of the qualification of facts. To justify the evolution in Belgacem, commissaire du Gouvernement Ronny Abraham relied on recent developments since Touami Ben Abdeslem concerning the control exercised by the ECtHR over these decisions.37 So, as the influence of the ECtHR had increased, it was necessary to control the proportionality of the decision. The influence of the ECtHR may also be noticed in cases concerning sanctions imposed by public authorities on prisoners. For decades sanctions against prisoners have been qualified as mesures d’ordre intérieur (decisions of internal order).

33 

See eg Mme Babas [1991] rec 162 (Conseil d’État), delivered the same day as Belgacem. Section française de l’observatoire International des prisons [2014] req no 369692 (Conseil d’État). 35  Touami Ben Abdeslem [1980] JurisClasseur Périodique (1981) II. 196113 (Conseil d’État). 36  Bernard Pacteau, JurisClasseur Périodique (1981) II. 196113. 37  R Abraham, ‘La Convention européenne des droits de l’homme et les mesures d’éloignement des étrangers’ [1991] Revue Française de Droit Administratif 497. 34 

206 

Yoan Sanchez

This is a category of decisions which are not considered to have enough consequences to justify the control of a judge, because of the idea that de minimis non curat praetor (what is not important does not need to be controlled by the judge). There was no control at all. The position of the Council of State changed in Marie.38 It accepted that it could control sanctions against prisoners. The commissaire du Gouvernement Patrick Frydman gave several reasons to persuade the Council of State to change its position and to accept control of some internal decisions.39 One of them is the necessity to take into account the impact of that kind of decisions on rights and freedoms. This strong argument is reinforced, in the conclusions of Frydman, by the influence of the ECtHR. He said that the Council of State has to control some mesures d’ordre intérieur because of ‘the obligations, from some commitments signed by France and, more precisely, from the European Convention of Human Rights’.40 True, this is not the only reason used by the commissaire du Gouvernement, but it is an important one. The judicial movement towards an increase of the field of judicial control, and so a decrease in the category of mesures d’ordre intérieur, is not yet finished. In three cases delivered in 2007, the Council of State has accepted that it could control some other decisions affecting prisoners. It has also given the elements which are taken into account to allow access to the judge. In these three cases, Boussouar, Planchenault and Payet, the Council of State adopted a two-step reasoning.41 The first step permits it to verify if the decision has a negative impact on the prisoner. This is so, for example, if the decision is to transfer him to a high-security­ prison (Boussouar), a refusal of employment (Planchenault) or an increase in the level of control over a prisoner (Payet).42 The judge looks at the decision, in abstracto. The qualification depends on the intrinsic nature of the decision. The second step is applied if, in the first one, the decision is not considered in itself to have an impact on the rights and freedoms. If so, the court verifies whether the decision has, in an in concreto approach, such negative consequences for individuals. It may be so, for example, if a prisoner has to be transferred to a low-security prison which does not have the necessary medical facilities to treat his condition.43 Here again the ECHR has been used. In his conclusions in cases Planchenault and Boussouar the commissaire du Gouvernement Mattias Guyomar made it clear

38 

Marie [1995] rec 84 (Conseil d’État). Frydman, ‘Le contrôle juridictionnel des mesures disciplinaires dans les institutions fermées’ (1995) Revue Française de Droit Administratif 353. 40 ibid. 41  Boussouar [2007] rec 495; Planchenault [2007] rec 474; Payet [2007] rec 498 (Conseil d’État). 42 ibid. 43  Rogier [2008] rec 800 (Conseil d’État). The Council of State verified whether the decision to transfer a prisoner to a low-level security prison had resulted in diminishing the level of care needed by the prisoner. There was no such consequence in this case so the decision was qualified as a mesure d’ordre intérieur. 39  P

Origins and Presentation of the Proportionality Principle in French Case Law  207   that the solution proposed to the Council of State was greatly influenced by the jurisprudence of the ECtHR. Articles 6(1) (access to justice) and 13 (right to an effective remedy) of the European Convention are involved here. The general trend in the ECtHR case law is to extend the application of Article 6(1) to the field of prisoners by qualifying sanctions against them as criminal charges.44 Mattias Guyomar used both these articles in order to explain that the limitation of the category of mesures d’ordre intérieur is necessary to respect the European Convention. For example, in Ramirez Sanchez v France, the ECtHR condemned France because of the lack of remedy against a decision affecting a prisoner, which was qualified as a mesure d’ordre intérieur.45 The necessity of a remedy when a decision has consequences for an individual, even a prisoner, is apparent. So Mattias Guyomar stated in his conclusions that ‘by opening a remedy for all cases which involve fundamental rights of prisoners, you will echo the conventional requirements for an effective remedy when a right protected by the Convention has been violated’.46 The evolution proposed here has a clear purpose: to limit risks of future condemnations by the ECtHR or, in the words of the commissaire du Gouvernement, ‘to refuse the control of decisions involved today is like closing eyes waiting for the ECtHR open them for you’.47 The impact of the European Convention and ECtHR is very important in the decrease of mesures d’ordre intérieur. The Convention, like the Court, is the ‘engine of the judicial evolution’.48 But this is not solely about opening the judicial way to control a decision. There is another question: what kind of control does the judge have to apply? The answer is also given by Patrick Frydman. He said in his conclusions that the judge has to control the choice of the decision. There is no explicit reference made to either a proportionality test or to the ECHR on this point. But the control of the choice of the sanction is clearly a control of proportionality. The sanction is an answer to the problem created by the behaviour of the prisoner. In this respect, it has to be linked to the importance of the problem. That is exactly the object of the proportionality control. Even if the word is not used by Frydman, the idea may be identified. In Marie, the proportionality of the decision was not verified because the decision was quashed due to an error on the qualification of facts. Indeed, for judges here, no fault was committed by the prisoner. But the same day, the judgment in Hardouin was delivered concerning sanctions against members of the m ­ ilitary;49 as for prisoners, sanctions against them could not previously be the subject of control. Hardouin changed the position of the judges, as did Marie for prisoners. Both were delivered with the conclusions of Frydman. But here the soldier had committed a fault and, in the opinion of the judge ‘by deciding a punishment of 44  Ezeh and Connors v UK Reports of Judgments and Decisions 2003-X (2003) App No(s) 39665/98 and 40086/98 (ECtHR). 45  Ramirez Sanchez v France (2005) App No(s) 59450/00 (ECtHR). 46  Above n 41. 47  ibid, 494. 48  M Lascombe and F Bernard (1996) JurisClasseur Périodique II 22426. 49  Hardouin [1995] rec 82 (Conseil d’État).

208 

Yoan Sanchez

10 days of detention, the military authority has not committed a manifest error of appreciation’. So there is no disproportionality in the choice of the quantum of the sanction. This reveals that if in Marie the judge had accepted the existence of a fault, he or she would then have checked the quantum of the sanction chosen. So, in part thanks to the ECHR, the Council of State has increased the field of its control in the 1990s and 2000s and has spread the application of a proportionality test to cover sanctions against prisoners and members of the military.

IV.  The Question of the Intensity of Proportionality Review A.  Distinction between Two Different Kinds of Intensity As explained previously, the proportionality control is not applied always in the same way.50 Sometimes judges limit their examination to the more obvious disproportion; in other cases, every element of disproportionality is quashed. There is clearly a link between ‘manifest disproportion’ and the control of the ‘manifest error’. Sometimes this last expression replaces the first. A good example is Hardouin, discussed above, about sanctions against members of the military.51 For the Council of State, the sanction in this case did not reveal a ‘manifest error of appreciation’ by the military authorities. From the point of view of this inquiry, this is the expression of a proportionality control.52 But when this expression is employed, or when judges said that a decision is not ‘manifestly disproportionate’, it is clear that this does not mean that no error, or no disproportionality, is apparent; it means that if there is such a flaw, it is not enough to justify the quashing of the decision. That is why, as explained above, a ‘power to err’ is left to the public authority when judges use this kind of formulation of proportionality control. The concept of ‘manifest error’ is not a ground for review. It is much more a degree of scrutiny. When one refers to ‘manifest error’ one should ask: manifest error of what? The administrative judges often use the expression ‘manifest error of appreciation’. So this is a question of appreciation. But the question is still not answered: appreciation of what? It means that the words ‘manifest error’ are employed solely in the control of the qualification of facts and of proportionality. In this respect we can answer the question: there is a manifest error in the appreciation of the qualification of facts; there is a manifest error in the appreciation of the proportionality of the decision. The concept of ‘manifest error’ cannot be used in isolation.

50 

Above see I.C(ii). Above n 49. 52 ibid. 51 

Origins and Presentation of the Proportionality Principle in French Case Law  209   So how can it be explained that sometimes, as mentioned above concerning Hardouin, judges make references solely to a ‘manifest error’? Two explanations can be given. First, as explained, administrative judges are very laconic. As they do not give much detail about their control, sometimes they solely use this expression. Second, it may be also because administrative judges are sometimes not ready to express clearly that they exercise a deep control over the decision. To verify the qualification of facts or proportionality of a decision is, for judges, to go very far in controlling motives and the content of a decision. So in this respect, the concept of ‘manifest error’ allows judges to hide the very nature of their control. At the same time, it shows to the public authority that they only quash the biggest mistakes. The discretionary power of the decision maker is thus preserved. What is apparent in the case law of administrative judges, especially of the ­Council of State, is the strengthening of the control of proportionality. The increase in the intensity of control concerning sanctions is a clear and recent example of it. This trend is clearly influenced by the ECHR. But we need always to be aware that judicial control is very much influenced by circumstances. As such, the recent events of November 2015 in Paris and the declaration of a state of emergency may have a negative impact on the intensity of proportionality control on serious police decisions.

B.  The Impact of the ECHR on the Intensity of Review This examination will begin with sanctions generally, before turning to the most recent evolution, concerning sanctions against prisoners.

(i)  Sanctions in general Dahan, a case from 2013 concerning sanctions against a civil servant, is relevant for this discussion.53 An ambassador was forced to take retirement because of improper behaviour. One of the arguments of the ambassador was that the sanction was disproportionate. Previously, in the field of sanctions against civil servants, the Council of State used a limited proportionality control by verifing that the sanction was not manifestly disproportionate. But in Dahan, the Council of State checked whether the sanction was strictly proportionate. In the opinion of the Council, it was. In his conclusions on this case, the rapporteur public, Remy Keller, underlined some elements to justify this evolution. Two are relevant here. The first one is the case law of the ECtHR which gives an increasingly extensive scope to Article 6(1) of the Convention. In Vilho Eskelinen v Finland, the ECtHR enlarged the application of Article 6(1) to all civil servants.54 Before this case, the

53 

Dahan [2013] rec 279 (Conseil d’État). Vilho Eskelinen v Finland, Reports of Judgments and Decisions 2007-II (2007) App No(s) 63235/00 (ECtHR). 54 

210 

Yoan Sanchez

position of the Court was stated in Pellegrin v France.55 It excluded at that time any benefit deriving from this Article for civil servants who had functions relating to the sovereignty of the state. By application of the latter case, ambassadors could not invoke Article 6(1). It is now available to them. This therefore means that in Dahan the ambassador had a right to a fair trial, and more interesting for this inquiry, a right to be judged by a court which has ‘full jurisdiction’.56 This important phrase means that the judicial body has to control all aspects of the decision. The ECtHR has already made it clear that to have full jurisdiction, in order to respect Article 6(1), the judicial body has to control the strict proportionality between the sanction and the fault.57 It is thus apparent that the evolution of the Strasbourg case law is a key element in the evolution of the position of the Council of State in Dahan. The second element taken into account by Remy Keller was the evolution in the case law of the Council of State before 2013. The rapporteur public mentioned national cases illustrating the trend towards a more intensive proportionality control. But what is also very interesting is the influence of the ECtHR on those developments. Two of the cases cited by Remy Keller serve as examples. The first one is Arfi, concerning sanctions imposed by a public authority on a professional.58 In his conclusions on this case, the commissaire du Gouvernement Mattias G ­ uyomar undertook a full study of the requirements of Article 6(1). He made clear that it requires control by a judicial body with ‘full jurisdiction’, as defined by the ECtHR. At the end, he proposed to the Council of State a solution to comply with it, which is to ‘exercise a full control of proportionality on the sanction’. The second case quoted by Remy Keller is Société ATOM.59 The rapporteur public in this case, Camille Legras, clearly stated that: [A] recours pour excès de pouvoir is sufficient to satisfy the requirement of the full jurisdiction involved by article 6 since the control of the judge on the sanction is full, which means that it is not limited to a control of manifest error.

In conclusion, it appears that to propose a strict control of proportionality in Dahan, the rapporteur public relied directly and indirectly on the ECtHR. To conclude, an old case also deserves to be mentioned. In 1981, in Le Cun, the Council of State chose to control the strict proportionality of a sanction against an agent of the Stock Exchange. Several reasons were given by the commissaire du gouvernement, Maryvonne de Saint Pulgent, to justify an increase in the proportionality control. In fact, at that time, the relevant case about the control of sanctions was Lebon in which the Council of State applied a low level of p ­ roportionality

55  Pellegrin v France Reports of Judgments and Decisions 1999-VIII (1999) App No(s) 28541/95 (ECtHR). 56  Albert and Le Compte v Belgium, Series A no 58 (1983) App No(s) 7299/75 and 7496/76 (ECtHR). 57  Diennet v France, Series A no 325-A (1995) App No(s) 18160/91 (ECtHR). 58  Arfi [2007] rec 263 (Conseil d’État). 59  Société ATOM [2009] rec 25 (Conseil d’État).

Origins and Presentation of the Proportionality Principle in French Case Law  211   control. Thus, the commissaire du gouvernement had no precedent on which she could base her proposition to use a strict proportionality control. One of the arguments used was the case law of the ECtHR concerning the requirements of ­Article 6. What is really interesting here is the will of commissaire de Saint Pulgent to make a clear distinction between sanctions against a civil servant, imposed by the immediate superior, and sanctions against a professional, imposed by a public authority vested with a regulatory mission. It was in order to draw a clear line between these two kinds of sanctions that the European Convention was used. In 1981, sanctions against a civil servant were not yet submitted to Article 6, while sanctions against professionals were controlled, because of their impact on their civil rights. So two strong arguments were used here to justify a strict proportionality test for the latter kind of sanctions. First, a sanction against a professional has an impact on civil rights because it concerns his or her job; secondly, the distinction between sanctions against civil servants and against professionals is accepted by the ECtHR which submitted the latter to a strict proportionality test because of Article 6. This demonstrates that the influence of the EctHR is not only important. It is also a comparatively old one.

(ii)  Sanctions on Prisoners In this field, the influence of the ECtHR is not as clear as it should be. The relevant case is Boromée, from 2015.60 In this case, a prisoner was sentenced to twenty-five days in a disciplinary cell. The Administrative Appeal Court (AAC) stated that the sanction was correctly decided by the public authoriy which had made no manifest error of appreciation. In this respect, the AAC relied on the control as stated previously by the Council of State.61 This decision of the court was contested before the Council of State. It decided that the AAC was wrong in so far as it solely required the public authority not to make a manifest error in the choice of the sanction. The Council said expressly that the AAC needed to ‘seek if the sanction was proportionate to the gravity of the fault’. In using a control based on manifest error, the AAC committed an error of law. It appears clear that the Council of State has changed its position. In light of previous developments, the impact of the ECtHR should be important to explain the evolution of 2015. But, in a remarkable way, it is not. In her conclusions on Boromée, Aurélie Bretonneau, the rapporteur public, said that a sanction against a prisoner does not fall within the requirements of Article 6(1) because it is not a criminal charge and the decision is taken by an administrative authority. No reference is made to the requirement of Article 13. So, to explain the evolution proposed, Bretonneau relied on three arguments.

60  61 

Boromée [2015] Actualité Juridique Droit Administratif 1596 (Conseil d’État). Letona Biteri [2011] rec 246 (Conseil d’État).

212 

Yoan Sanchez

First, the previous decisions of the Council of State concerning the intensive control of proportionality are used as good precedents. Before Boromée, sanctions against prisoners were the only field with a limited control of proportionality. The second argument is the necessity to protect prisoners, who are vulnerable persons needing judicial protection of their rights and freedoms. This is all the more so since it appears that an increase in the intensity of the proportionality test does not lead to the inefficiency of public authorities. So both protection of prisoners and administrative efficiency may be preserved. In her third, and last, argument the rapporteur public stated that this evolution is encouraged, not by the ECtHR, but by the case law of the Constitutional Council. All these arguments are based on internal considerations as if European law had no impact. Three criticisms have to be made about the position of the rapporteur public in this case. First, the study of the ECHR is not complete. As we have noted, no reference was made to the Article 13. Moreover, it is obvious that the evolution noted by Bretonneau in other fields of administrative law toward a greater intensity of control is in an important respect the result of the evolving jurisprudence of the ECtHR. The rapporteur public gave Dahan as an example of the change in the control exercised by the Council of State. This case was clearly influenced by the ECtHR. So the evolution in Boromée is indirectly influenced by the ECtHR. Finally it seems that the rapporteur public made some mistakes about the requirements of the ECtHR. In fact, she stated that Article 6(1) was not applicable because the sanction was imposed by a public authority and because the sanction could not be qualified as a criminal charge. Then, a decision of the Council of State of 2012 is quoted to highlight that, because of its administrative nature, the authority which takes the disciplinary decision in prisons cannot be submitted to the requirements of Article 6(1).62 But this argument appears erroneous in two ways. First, the application of this Article is not dependent of the nature of the primary decision-maker. In some ways, eg in town planning, the administrative authority has to take decisions in a way that is compatible with Article 6(1).63 The nature of the authority is taken into account when it is necessary to define the requirements of that Article. The second mistake is related to the role of the court in the course of ensuring the respect of this Article. What is clear in the case law of the ECtHR, and which is not open to question in the Council of State’s case of 2012, is the necessity for courts to fill the gap left by the decision-maker in respect of Article 6(1). So in Segame v France, quoted by the rapporteur public, the ECtHR clearly states that ‘the respect of Article 6§1 of the Convention means indeed that the decision of an administrative authority which does not fulfil itself

62  Section française de l’Observatoire International des Prisons [2002] Gazette du Palais no 222 (Conseil d’État). 63  Bryan v the United Kingdom, Series A no 335-A (1995) App No(s) 19178/91 (ECtHR).

Origins and Presentation of the Proportionality Principle in French Case Law  213   the ­requirements of this article had to undergo the later control of a judicial body with “full jurisdiction”. ’64 The question is not: does the public authority have to respect all the requirements of Article 6(1)? The question is: is the decision involved in the case submitted to the demands of Article 6(1), and if so, does there exist a judicial body with full jurisdiction which controls the decision? The 2012 case decided by the Council of State does not say anything else. Indeed, the judge argued that the primary decision-maker did not have to respect the requirements of Article 6(1). But this does not mean that this Article does not define the control of the court over the decision or the decision-maker. If the judicial body controls all aspects of the decision, which encompasses a strict proportionality control, the requirements of Article 6(1) are respected. Now, the question is: does Article 6(1) apply to discplinary sanctions against prisoners? As Bretonneau said, disciplinary sanctions are not qualified by the ECtHR as criminal charges. But, in some aspects, a sanction may have an impact on ‘civil rights’, and if so, lead to the application of Article 6(1). For example, in Albert and Lecompte v Belgium, the ECtHR has stated clearly that this Article is applicable to a decision to suspend an individual from his work.65 In France, the disciplinary authority has the power to suspend a prisoner from his work for a few days, as a sanction.66 The authority may also decide, as in the 2015 case, to put the prisoner in a disciplinary cell, away from other prisoners, for a maximum of 30 days. This decision forbids the prisoner the right to buy things and to work; it therefore has a clear impact on the detention conditions of the individual. For all these reasons, it appears that disciplinary sanctions in prison have to be considered as subject to the requirements of Article 6(1). For Boromée, it does not seem to change the decision, because the Council of State decided to apply a strict proportionality control. But it is not satisfying to distance the European Convention from this evolution, from an intellectual as well as from a practical point of view. The Council of State needs to take into account the case law of the ECtHR also in this field, otherwise France risks being condemned in Strasbourg. This point may be demonstrated by studying the case law about the evolution of the control of decisions concerning the organisation of life in prisons. The prison authorities need to take preventive measures to maintain order. ­Systematic and full-body searches are designed in order to pursue this aim. In 2000, in Frérot, the Council of State declared lawful a circular requiring full-body searches. In the opinion of the judges, no other means could achieve the legitimate purposes specified by the minister, which were to limit risks of assaults and ­smuggling.67 It was clear in the words of the court that the circular was not disproportionate. So the control seems to be a strict one, because there is no reference to the idea of ‘manifestly disproportionate’. This appears also from the conclusions 64 

Segame v France Reports of Judgments and Decisions 2012 (2012) App. No(s) 4837/06 (ECtHR). Albert and Le Compte v Belgium (n 56). 66  Art 57-7-34 of the Code de Procédure Pénale. 67  Frérot [2000] rec 589 (Conseil d’État). 65 

214 

Yoan Sanchez

of the commissaire du Gouvernement Rémy Schwarts.68 He said that the court had to ‘make a proportionality control between interests involved’, which means ‘on the one hand, security and public order, and on the other, the necessary respect of human dignity’.69 Clear references were made in the conclusions to the requirements of the ECtHR’s jurisprudence on this matter. So it seems to be in line with the European Convention. But this case was the subject of challenge before the ECtHR. France had been condemned in 2007. In Frérot v France, the ECtHR declared full-body searches a justified method of control in prisons. But, regarding the facts of the case, nothing really justifies the great frequency of the searches.70 The control of the Council of State was here condemned because it was only conducted in abstracto. In the opinion of the ECtHR, it is necessary to go beyond this and control, in concreto, the reasons for and the frequency of the searches. It is not enough to declare them as a legitimate tool to preserve public order in prison. Their use has to be justified.71 It was not until 2013 that the position of the French administrative judges changed, in two cases decided on the same day, ME and Section Française de l’Observatoire International des Prisons.72 In the latter case, the judges declared illegal a decision of the director of a prison to institute for three months systematic full-body searches after visiting hours for all prisoners. The Council of State argued that this would only be lawful for a prisoner who was clearly dangerous. These two cases show the evolution in the control of the Council of State which is now in concreto. In both cases, one can find that ‘the necessity of proportionality in the methods of organisation of full-body searches requires them to be strictly adapted not solely to the objectives pursued but also to the personalities of prisoners who are concerned’. It explains why in Section Française de l’Observatoire International des Prisons, a general decision to conduct systematic full-body searches is declared disproportionate. It is too general, and lacks adaptability and justification. The Council stated that the public authority, acting wrongfully, ‘does not plan the possibility to modulate its application by taking into account the personality, behaviour and frequency in the utilisation of visiting room of the prisoner’. But an individual decision to run these searches, as in ME, is proportionate because the decision is based on facts which justify it, namely the dangerousness of the prisoner. The impact of the ECtHR is manifest here. It is also necessary to mention that another authority has played an important part in this ­evolution: Parliament. A statute of 2009 provides that full-body searches have to be

68 (2001) 69 ibid. 70 

Les Petites Affiches no 28, 16.

Frérot v France (2007) App No(s) 70204/01 (ECtHR) §46–48. See also Khider v France (2009) App No(s) 39364/05 (ECtHR) §123–32. ME [2013] Actualité Juridique Droit Administratif 1191 (Conseil d’État); Section Française de l’Observatoire International des Prisons [2013] Actualité Juridique Droit Administratif 1191 (Conseil d’État). 71  72 

Origins and Presentation of the Proportionality Principle in French Case Law  215   decided by taking into account the personality of the prisoner but also have to be ‘strictly adapted’ to the prisoner’s personality.73 Both requirements drawn from the ECtHR’s jurisprudence are apparent: a decision has to be taken in concreto and has to be proportionate to the situation. Sometimes, statutes may be a useful instrument to change the jurisprudence of the Council of State in order to respect the ECHR’s requirements. So, even when control seems to be a full proportionality test, this does not mean that it is sufficient to satisfy the requirements of the European Convention.

C. The Impact of the Declaration of the State of Emergency: What Intensity of Control? Following the terrorist attacks in Paris in November 2015, a state of emergency was declared in France, under the 1955 Act. It gives police authorities more powers in order to counter the terrorist threat. It is possible to put an individual under house arrest if there are some elements of evidence against him proving he is a threat to public order and security. The decision is taken by the Home Office. The state of emergency infringes liberties, especially freedom of movement, in an important way. In the opinion of the Constitutional Council, the proportionality of a house arrest decision has to be controlled by the administrative judges. It means that judges have to verify if the decision is ‘adapted, necessary and proportionate to the goal pursued’.74 This control is one of the arguments used by the Constitutional Council to declare that the mechanism of house arrests is constitutional. Administrative judges thus have a constitutional mission in verifying their proportionality. The Council of State does not fulfil this function. In seven interim decisions taken the same day, the Council of State declared legal seven house arrest decisions without reference to a proportionality test.75 The three steps of control are not even mentioned, whereas they appear clearly in the decision of the Constitutional Council. This is even more questionable since police decisions are the perfect field to apply this test explicitly. The fact that those decisions are interim judgments cannot explain the absence of an explicit proportionality control. In some other interim decisions, administrative judges have made references to the three steps.76 The difference between them cannot be explained by the nature of the judicial procedure. The only possible explanation lies in the emergency situation. If this

73 

Loi pénitentiaire [2009] Art 57. M Cédric D (n 4). 75  MJ Domenjoud [2015] Actualité Juridique Droit Administratif 2404; ML Gauthier [2015] req no 394990; MC Verrier [2015] req no 394991; MP Boilleau [2015] req no394992; Mme M Satier [2015] req no 394993; Mme S Crochet [2015] req no 395002; MC Domenjoud [2015] Actualité Juridique Droit Administratif 247 (Conseil dÉtat). 76  Dieudonné [2014] rec 1 (Conseil d’État). 74 

216 

Yoan Sanchez

conclusion is right, it means that the proportionality of house arrest decision cannot be controlled. But if the three steps are not mentioned in the Council of State’s decisions, one can, however, find the idea of reconciliation. In fact, the Council stated that the police authority had to take a decision which reconciles the respect for liberties and the protection of public order. As the idea of reconciliation is at the heart of the proportionality test, we might expect a control of this nature by the judge. We may also add that the three-step control is mentioned in the conclusions on these seven decisions prepared by rapporteur public Xavier Domino.77 But it does not really seem to be the intention of the Council to assert such a control. The seven house arrest decisions controlled in December 2015 did not involve terrorism suspects, but rather extreme left-wing individuals. Police decisions had as their purpose to prevent them from participating, maybe in a violent way, in public demonstrations during the United Nations conference on climate change in Paris at the end of the year. This purpose is remote from the reasons justifying in the first place the declaration of a state of emergency. But the Council of State declared them legal. The rapporteur public proposed a similar decision. Can one say that these decisions really reconcile protection of public order and respect for liberties? It is in fact highly questionable to allow the police authority to take such decisions with a purpose which is not directly linked to the declaration of a state of emergency. The Council of State justified its position by suggesting that the police forces were mostly being employed to counter the terrorist threat and to protect the international conference. The argument here seems to be the risk of a lack of police forces if extreme left-wing persons were released from house arrest and able to organise and participate in large demonstrations. This argument is not convincing since limiting the liberties of seven persons has little positive consequence for the protection of public order. These seven house arrest decisions cannot diminish the threat to public order so as to release police forces in order to use them for the purpose of preventing terrorist threats. If protection of public order wins a little, respect for liberties loses a lot. No reconciliation is in operation here. In this respect, these decisions are far from the requirements of Benjamin.78 Are these decisions adapted? Necessary? Proportionate? It seems not. Seven ­persons in a public demonstration do not create such a risk to public order. This is also revealed by the conclusions of the rapporteur public. Xavier ­Domino stated that the proportionality control applies to the obligations imposed on an individual who is subject to house arrest. A contrario, this control does not apply to the decision itself to put an individual under house arrest. Concerning the seven decisions, Domino was very laconic and concluded, with no real argumentation, that ‘they appear to us necessary, adapted and proportionate to the aim pursued’.79 77  X Domino, ‘Assignations à résidence en état d’urgence’ [2016] Revue Française de Droit Administratif 105. 78  Benjamin (n 1). 79  Domino (n 77).

Origins and Presentation of the Proportionality Principle in French Case Law  217   But this is very questionable since the rapporteur public noted that the government had forbidden public demonstrations during the most sensitive day of the international conference. The seven decisions are not necessary as another means to protect public order that is less restrictive to individuals exists and has already been used by the governement. In this respect, the house arrest decisions seem superfluous and thus disproportionate. The intensity of judicial control is thus very low. The judge does not check the proportionality of house arrest decisions. The declaration of a state of emergency gives a lot of powers to police authorities without any effective counterbalance. This is a highly questionable position taken by the Council of State. First, because the Constitutional Council expressly said that administrative judges have to verify the proportionality of this kind of decisions. In not following this position, the supreme administrative judge is refusing to perform his constitutional role.80 ­Second, it reveals that in times of trouble, administrative judges can be really generous to public authorities. But it is exactly in that kind of situation that a powerful judge is needed. The judicial control, by way of proportionality test, has to draw a line which police decisions cannot cross. This is missing in the decisions studied here. Because of the necessity of public order, the administrative judge gives too much margin of appreciation to police authorities. In my opinion, this is clearly reflected by the first decision to suspend a house arrest decision pronounced by the Council of State, in January 2016.81 The decision was based on three reasons: the individual had taken pictures of the house of a person who was protected by the police; he was an Islamist; and he had participated in the trafficking of cars linked to a terrorist movement. But it has been made clear by the Council of State that none of these three reasons was true or properly proved by the Home Office. This means that the only possibility to declare unlawful such a decision is to show that there is no reason at all that justifies the decision. It is not a proportionality control. A contrario, it means that the only thing that police forces are required to find and prove is a legitimate basis for the decision. If this is done, it is hard to imagine how the judge may quash a properly reasoned decision without using a proportionality test. To conclude concerning house arrest decisions, one final case deserves mention. At the beginning of January 2016, the Council of State declared excessive the conditions imposed by a house arrest decision concerning a woman.82 The judge clearly stated that the decision to put her under house arrest was, in itself, legal; but the conditions imposed on the woman are not. In fact, a person who is subjected to such a decision has obligations to comply with these conditions. For example, in this case, the woman concerned had to go to the police office three times daily, every day of the week. The office was in a town 10 km from her house, although 80  The seven decisions are previous to the decision of the Constitutional Council. But the day after this decision, the Council of State confirmed its position: MAB [2015] req no 395229 (Conseil d’État). 81  MB [2016] Actualité Juridique Droit Administratif 127 (Conseil d’État). 82 Mme BEC [2016] Actualité Juridique Droit Administratif 11 (Conseil d’État).

218 

Yoan Sanchez

there was one where the claimant lived. She has three children; two have to go to school and the last one is too young to go to; she has no car so she has to take a bus and a train to go to the police office. All these elements placed great difficulties on her as she tried to to comply with the obligations of the police decision. The Council took all these factors into account and on this point declared the decision illegal as it infringed excessively on the right to respect for family life and the protection of the superior interest of the child. I identify a proportionality test here as, for the Council, the police authority had to choose another way to ensure the conditions would be met by the woman. For the judge in this case, the police authority had to choose the office that was closer to her home, even if its opening days did not exactly match the conditions imposed by the house arrest decision. By verifying all these elements (bus and train timetables, schedules of the school, etc), the Council took an accurate look at the case in order to verify the proportionality of the burdens imposed on family and children’s rights. This is a strict proportionality control. It is this control that is mentioned by Xavier Domino in the conclusions studied at the beginning of this section. It is a very welcome decision and an important use of proportionality as a tool to protect the rights of persons who are under house arrest. It appears, then, that if a decision to impose house arrest has as its main object to restrict freedom of movement, it cannot impose heavy burdens on some other rights such as the right to respect for family life. It is a normal consequence of the decision to infringe these rights. But it has to be as limited as possible—ie it must be proportionate. The case reinforces the critical view presented above concerning the control of the decision itself. If a strict proportionality control of the conditions imposed by the decision is possible, what are the reasons not to apply a similar test to the decision to put someone under house arrest?

D. Do Judges Control Merits of Decisions by Using Proportionality Control? Before the developments described above, the control of proportionality was more flexible and left public authorities with a kind of ‘power to err’. Now, with control being more intensive, there is no longer ‘power to err’. So this kind of reflection may lead to the idea that if the sanction is not the correct one, then from the point of view of the Council of State it will be quashed. But it is not so easy. The increasing intensity does not mean that public authorities have no more discretionary power. Here is the difference. There is no more ‘power to err’ but there is still a discretionary power. It is clear, for example, in the conclusions of Mattias Guyomar in Arfi. The commissaire du Gouvernement asserted that proportionality is useful to determine what sanctions are allowed in a case: the sanction cannot be too low or too high, but between these limits

Origins and Presentation of the Proportionality Principle in French Case Law  219   the ­public authority is free to choose the sanction it prefers.83 The increase in the intensity of proportionality control must not ‘give the impression that to punish a specific fault, only one sanction is legal’. The exact same idea is defended by Rémy Keller when he said in his conclusions in Dahan that ‘it is only about strengthening the bounds of legality’, and not to ‘deprive the public authority of any flexibility. It is only if the sanction has crossed the bounds that it shall be quashed’.84 The evolution of proportionality control is limited by the necessary respect for the discretionary power of public authorities. Sometimes judges are not the best decision-makers. This is because they lack technical competences or because the matter is a political one. Sometimes it is far more necessary and convenient to leave discretionary powers to public authorities. In this vein, judges must not state what the best decision is, as they are not the best institution to decide. Even if proportionality control operates at its maximum degree of intensity, it cannot be used to decide the merits or opportuneness of a decision. Proportionality has to be used as a tool to quash decisions that are not acceptable, and so illegal. It is not an instrument allowing the judge to decide what the ‘good decision’ is. The distinction between what is legal and what is good is an important one. Administrative judges, using proportionality, decide what is legal. When the intensity of control increases, it enables the judge to define more precisely what legal decisions are, but not what the better or best one is. This ultimate determination is for the public authority. The opportuneness of the decision is an important limit to the proportionality control, a limit that needs to be respected. Because of it, it is not possible to say that the Council of State puts itself in the shoes of the public authority. However, the distinction is not always very clear. This may be demonstrated by two points. First, it is the judge who draws the line between these two concepts. This is clearly revealed by the increase in the intensity of proportionality control. When judges quash decisions only where they are manifestly disproportionate, it is, in part, because judges do not want to cross the line toward a control of merits. But it is crossed by the transition to strict proportionality control. Does this mean that judges are now ready to control the merits? No, because the increase in the intensity of control serves only to push back the borders of legality. So it leads to a circular reasoning: the court has to control the legality, not the merits of a decision; the legality of a decision is constituted by all the elements controlled by the court; so it means that what is controlled by judges is what they decide to control. The second point focuses on the possible decision left to the public authority after a strict control of proportionality. Concerning particularly the field of sanctions, it is not very unusual to find that a statute makes a limited number of sanctions available to the primary decision-maker. When a fault is committed, the

83  84 

Arfi (n 58). Dahan (n 53).

220 

Yoan Sanchez

authority has to choose one of them. But if, after judicial control, there is only one sanction that could satisfy the requirements of the decision of the judge, the line between legality and merits cannot be clearly drawn. Nothing of the discretionary power of the authority remains. This argument has more weight when the administrative judge quashes a decision because it is an insufficient one.

V. Conclusion Proportionality control is a complex topic. The Council of State is not very clear in its formulation and use. Although European laws may influence this control and increase its intensity, the current position is far from revealing a systematisation of the principle by the judge. It is very questionable. A clearer application of the proportionality principle, as has been achieved by the Constitutional Council, would be welcomed. It is also vital to be very careful about how proportionality control will be applied to decisions taken under the state of emergency. If public order has to be preserved, the price must nevertheless not be too expensive. The proportionality principle, in this respect, has a role to play. It is up to judges to let it perform this role.

13 An Italian Perspective on the Principle of Proportionality GIUSEPPE MARTINICO AND MARTA SIMONCINI*

I. Introduction This chapter aims to explore the origins and the application of the proportionality principle in Italy in the areas of administrative and constitutional law. By ­analysing how the principle has evolved in the domestic legal framework according to the influence of EU law, the chapter aims to investigate whether a general principle of proportionality under public law exists in Italy and how it is articulated. In an article published in 2010 Moshe Cohen-Eliya and Iddo Porat1 explored the concepts of balancing and proportionality, stressing the differences existing

*  ‘Giuseppe Martinico is Associate Professor in Comparative Public Law, Scuola Superiore Sant’Anna, Pisa. He can be reached at: [email protected]. Marta Simoncini is FWO Postdoctoral F ­ ellow at the University of Antwerp and King’s College, London. She can be reached at: marta. ­[email protected]. Paper presented at the international conference ‘General Principles of Law: European and Comparative Perspectives. Celebrating 20 Years of the Institute of European and Comparative Law’, Oxford, 25–26 September 2015. Giuseppe Martinico wrote sections I, III and V while Marta Simoncini wrote paras II, IV and VI. Many thanks to Giuseppe Bianco, Daniel Sarmiento, Giacomo Delledonne and Cesare Pinelli for their comments. 1 ‘Indeed, the differences are so evident that they outweigh the similarities. For instance, ­proportionality was originally developed in administrative law, and was related only tangentially (if at all) to private law, while balancing arose in private law and was only later extended to public law. Moreover, proportionality was created as part of an attempt to protect individual rights against a background of little textual support for such protection, whereas balancing was created for the exact opposite purpose—to check the overzealous (libertarian) protection of rights by the US Supreme Court based on an excessively literal reading of the constitutional text. And, finally, proportionality was developed in the course of the formalistic and doctrinal jurisprudence of the German administrative courts and was not part of an anti-formalistic legal philosophy, whereas balancing was part of the antiformalist revolution of the US progressives, and a leading aspect of this revolution. We believe, therefore, that the historical account shows that we have an interesting phenomenon confronting us: two legal principles that began very differently but came to a point where, today, it seems natural to ­discuss the two together.’ M Cohen-Eliya and I Porat, ‘American Balancing and German ­Proportionality: The Historical Origins’ (2010) 8 International Journal of Constitutional Law 263, 266.

222 

Giuseppe Martinico and Marta Simoncini

between the two, reconstructing their genesis and the importance of the American and German contexts in this story.2 As we will try to show, in Italy national judges seem not to emphasize too much the difference between the two concepts either from a linguistic or a substantive point of view. They often regard these concepts as fungible and rarely articulate the proportionality test in three or four steps according to the traditional Prussian model. This is perhaps more evident in the case law of the Italian Constitutional Court. The remainder of this chapter is divided into four parts. In section II we shall recall the roots of the proportionality principle in Italian administrative law. This part is crucial to help the reader understand how EU law has influenced Italian law. In section III, we shall look at proportionality in the EU, stressing the ­chameleonic nature of this concept in EU law as the terminology employed by scholars shows: proportionality is described both as a principle and a test, where test c­orresponds to the criteria used in judicial review to identify and assess the principle in a concrete case. Tests differ depending on the concrete substantive area characterising the case pending before courts. In section IV, we shall see the impact that this way of conceiving proportionality has had on the Italian judges, presenting an overview of the case law of some administrative judges and of the Italian ­Constitutional Court. The last part will offer some final remarks.

II.  Origins of the Proportionality Principle in Italian Administrative Law The concept of proportionality identifies the reach of authoritative public ­measures by setting the limits to the suitable amount of public intervention in the private sphere. Lord Diplock has clearly explained the idea of proportionality as the prohibition against using a ‘steam hammer to crack a nut, if a nutcracker would do’.3 It is not by chance that the principle emerged in the context of national measures governing police powers, as a restriction on their coercive use. German scholarship has described proportionality as the interpretative rule preventing police from ‘shoot[ing] sparrows with cannons’4 and it identified in proportionality a further expression of the general principle of legality.5

2  On the concept of balancing, see J Bomhoff, Balancing Constitutional Rights: The Origins and Meanings of Postwar Legal Discourse (Cambridge, Cambridge University Press, 2013) and R Alexy, ‘Balancing, Constitutional Review, and Representation’ (2005) 3 International Journal of Constitutional Law 4, 572. For a general comparison between balancing and proportionality, M Tushnet, Advanced Introduction to Comparative Constitutional Law (Cheltenham, Elgar, 2015) 71. 3 See R v Goldsmith [1983] 1WLR 151, para 155 4  See F Fleiner, Institutionen des Deutschen Verwaltungsrechts (Tübingen, Mohr Siebeck, 1912) 354. 5  See HP Ipsen, Europäische Gemeinschaftrecht (Tübingen, Mohr Siebeck, 1972) 512.

An Italian Perspective on the Principle of Proportionality

 223

Every legal order in Europe has been developing the idea of proportionality according to its own legal understanding of the fair limits to the concrete exercise of public powers. The reading of the proportionality principle depends on the balance between individual rights and public authority that is considered fair in a specific legal order and it is shaped by the conception of fundamental rights and public interests. The application of the proportionality principle can therefore show to what extent a specific legal order wants to limit the exercise of public competences and how it justifies public interference with the exercise of individual rights. In German law the proportionality principle has found its most defined ­formulation through the elaboration of a three-step test aimed at identifying firm limits to public powers. Proportionality is assessed on the different grounds of suitability, necessity and proportionality in the strict sense or ­adequacy.6 ­ Suitability means that the measure should be appropriate for the purpose of facilitating or achieving the pursued goal; necessity suggests that the concerned public authority has at its disposal no other instrument to pursue its goals which is equally effective and less restrictive of freedom; proportionality in the strict sense provides that the measure may not impose an intolerable burden on the ­individuals that it affects. In administrative law, it is generally accepted that proportionality consists of the reasonable adjustment of public measures to the pursued public goals. The principle provides that a public authority may not impose obligations on citizens except to the extent to which these are strictly necessary in the public interest to attain the purpose of the measure.7 The proportionality test is a means to check that individual rights are not over-restricted by administrative action, but in doing so it also assesses whether that administrative power is correctly implemented. In fact, the choice of the instrument which is less intrusive for private parties relies on complete and accurate fact-findings. This commits public administration to the correct understanding of the different public and private interests at stake. Under the proportionality test the lens of individual rights sets the ­criterion for controlling the correct exercise of administrative discretionary powers. Therefore, proportionality concerns the exercise of administrative competences. The connection between the accuracy of fact-findings and the ­proportionality of administrative decisions is shown by the origins of the principle in ­Italian administrative law. It took a while before the principle could be recognised as an

6  See DU Galetta, Principio di proporzionalità e sindacato giurisdizionale nel diritto amministrativo (Milano, Giuffrè, 1998) 14; A Sandulli, La proporzionalità dell’azione amministrativa (Padova, Cedam, 1998) 365; JH Jans, ‘Proportionality Revisited’ (2000) 27 Legal Issues of Economic Integration 239, 240; J Snell, Goods and Services in EC Law. A Study of the Relationship Between the Freedoms (Oxford, Oxford University Press, 2002) 196. 7  See J Schwarze, ‘The Principle of Proportionality and the Principle of Impartiality in European Administrative Law’ (2003) Rivista trimestrale di diritto pubblico 1, 53, who considers proportionality as the principle ‘set[ting] material limits to the interference of public authorities into the private sphere of citizens’.

224 

Giuseppe Martinico and Marta Simoncini

autonomous ground of review. Originally, in Italy the principle emerged within the ultra vires review of administrative competences and was developed as a sub-set of the test of reasonableness. The early Italian scholarship on administrative law suggested that the function pursued by public administration should be accomplished with the minimum sacrifice of the private sphere: in the absence of clear-cut criteria about how to settle competing public and private interests and in the presence of a lacuna legis which could not be addressed differently, the public good should prevail over the private one to the minimum detriment of private property and liberty.8 The case law of the highest administrative court, Consiglio di Stato, has slowly included this idea of proportionality within its judicial review of administrative discretion. Alongside the so-called symptomatic figures aimed at detecting the excès de pouvoir,9 the control over proportionality has been interpreted as a further check over the legitimacy of administrative action through the lens of the private sacrifice.10 It paired the reasonableness test of administrative action by ­introducing the concrete review of the necessity of administrative action. The principle of reasonableness allows the scrutiny of the abstract consistency of the administrative decision with the rule of law, the internal coherence of the administrative decision, and its external coherence with comparable administrative decisions. Reasonableness tells that theoretically, from a standpoint of pure logic, no contradictions exist in the implementation of the administrative action. By stating how a reasonable decision shall balance the different interests at stake, proportionality confers on reasonableness a more concrete take, so as to obtain also a fair decision. The principle of reasonableness controls only the suitability of the 8  See GD Romagnosi, Principi fondamentali del diritto amministrativo onde tesserne le istituzioni (Prato, Guasti, 1815) 14; S Spaventa, ‘Discorso per l’inaugurazione della IV sezione del Consiglio di Stato’ (1909) Rivista di diritto pubblico 308; F Cammeo, Commentario delle leggi sulla giustizia amministrativa (Milano, Vallardi, 1911-12) 130; VE Orlando, Primo trattato completo di diritto amministrativo italiano, vol 1 (Lodi, Società editrice libraria, 1897) 159; C Vitta, Diritto amministrativo, 2nd edn (Torino, Unione tipografico-editrice torinese, 1937) 78; see also R Alessi, Sistema istituzionale del diritto amministrativo italiano (Milano, Giuffrè, 1958) 181; V Spagnuolo Vigorita, ‘Eccesso di potere per sproporzionata gravosità dei vincoli imposti alla proprietà privata’ (1958) 1 Rivista giuridica dell’edilizia 626; V Spagnuolo Vigorita, ‘Sulla necessaria proporzione fra area espropriata e utilità ­pubblica’ (1961) 4 Rivista giuridica dell’edilizia 825. 9  Administrative judges developed a series of cases which are treated as symptoms of the excess of power in the exercise of administrative discretion, so that the detection of these symptoms shall imply the judicial annulment of the administrative provision due to excès de pouvoir. Symptomatic figures include, for instance, insufficient investigation in the fact-finding, the manifest unfairness of the disposition, the alteration of facts, the contradiction with previous provisions, the contrast between the motivation and the disposition, and also the lack of motivation. 10  See Consiglio di Stato, Plenary Session, 6 February 1993, n 3, which grounded the proportionality review on the reasonableness of the administrative decision. The clear distinction between the two principles emerged in the case Consiglio di Stato, V, 18 February 1992, n 132, but it has been consistently developed only later by TAR Lombardia Milano, III, 5 May 1998, n 922. In the literature, see Sandulli (n 6) 322; A Sandulli, ‘Il principio di proporzionalità dell’azione amministrativa’ in G Pasquini and A Sandulli (eds), Le grandi decisioni del Consiglio di Stato (Milano, Giuffrè, 2001) 588; A Sau, La proporzionalità nei sistemi amministrativi complessi. Il caso del governo del territorio (Milano, Franco Angeli, 2013) 56-62.

An Italian Perspective on the Principle of Proportionality

 225

administrative measure, whereas proportionality also informs about the necessity and the fairness of the burden on private parties.11 As has been correctly observed, this means that the judicial annulment of an unreasonable decision allows public administration to reiterate the same kind of measure with different motivations, whereas the annulment of a disproportionate decision precludes the adoption of another measure of the same kind.12 Justification of the decision is fundamental for the proportionality test, whereas the reasonableness test focuses more on the formal implementation of the guarantees of administrative procedure.13 As long as the decision is justified on the grounds of the findings of the administrative proceedings, the judicial review of the proportionality principle builds upon the one of reasonableness. It particularly requires the participation of rights’ (and interests’) holders in the administrative proceedings and an evidence-based ­justification to the restriction of individual rights and interests. In Italian administrative justice, the peculiarity of these different principles is not straightforward. Sometimes the two principles are ambiguously addressed in the case law and the judgment of reasonableness is conducted, which in ­practice means a judgment of proportionality.14 The origin of proportionality in the domain of reasonableness still affects the judicial understanding of the principle. However, Italian administrative justice clearly perceives the potential of the ­proportionality test in terms of intensity of judicial review. Unlike the reasonableness test, the very idea of proportionality is able by itself to shift judicial review from the control of the legality of administrative measures to their opportunity, ie from a legitimacy control to a control of the merit.15 As the general principle of separation of powers prescribes that any confusion and substitution between administration and judiciary shall be avoided, judicial review on the merit is very limited in Italian administrative law. It concerns only s­ pecific

11 

See Galetta (n 6) 207. V Parisio, ‘Principio di proporzionalità e giudice amministrativo italiano’ (2006) 15 Nuove autonomie 717, 718-19. 13  See R Ferrara, Introduzione al diritto amministrativo. Le pubbliche amministrazioni nell’era della globalizzazione (Bari: Laterza, 2005) 192; Sau (n 10) 53–54. 14  See Consiglio di Stato, IV, 18 October 2002, n 5714, which did not assess the breach of proportionality requested by the plaintiff and limited the review to the accuracy and the completeness of the fact-finding; and Consiglio di Stato, IV, 26 September 2008, n 4648, which reviewed the excès de pouvoir on the grounds of the suitability and necessity of the administrative decision, but considered it as an assessment of reasonableness and non-arbitrariness. See also Consiglio di Stato, IV, 6 February 2009, n 699; Consiglio di Stato, VI, 13 February 2009, n 1348; Consiglio di Stato, VI, 2 March 2009, n 1191; Consiglio di Stato, VI, 6 March 2009, n 1348; Consiglio di Stato, VI, 4 June 2009, n 3446; Consiglio di Stato, IV, 8 June 2009, n 3501; Consiglio di Stato, IV, 12 June 2009, n 3723; Consiglio di Stato, VI, 11 November 2010, n 14. In the literature, see G Lombardo, ‘Il principio di ragionevolezza nella giurisprudenza amministrativa’ (1997) Rivista trimestrale di diritto pubblico 4, 939; Galetta (n 6) 209–10; Sau (n 10) 62–63. 15  On the concerns about the possible interference of the proportionality review with the principle of separation of powers, see P Craig, ‘Theory and Values in Public Law: A Response’ in P Craig and R Rawlings (eds), Law and Administration in Europe: Essays in Honour of Carol Harlow (Oxford, Oxford University Press, 2003) 38. 12  See

226 

Giuseppe Martinico and Marta Simoncini

cases identified in the legislation where the exercise of administrative powers may involve too broad a discretion which may overlap with the legislative power itself and may also infringe upon fundamental liberties and values recognised in the Constitution.16 When applying the proportionality test in a judicial review of the legitimacy of administrative measures, therefore, the judge shall restrict his ­judgment only to those aspects which do not fall within the merits. The ­standard of judicial review has therefore been adjusted so that the judgement should concern only the consistency of fact-finding as it emerged in the motivations of administrative decisions.17 Nonetheless, the shift from the reasonableness test to the proportionality test affects also the articulation of the administrative process, by contributing to moving its focus from the reasonableness of the administrative decision as such to the fairness of the relation between citizens and public administration in the course of the administrative action.18

III.  Proportionality in European Union Law Before moving to the influence EU law has over the activity of national judges, it is necessary to recap the main features that the concept shows in EU law.19 As anticipated in the introduction to this chapter, in EU law (and, probably, not only there) proportionality is a multifaceted concept which is described, at the same time, as a principle, as a test and as a technique of dialogue.20 Indeed, as we will see in this section, the Court of Justice (CJEU) frequently delegates the outcome of the proportionality test to national judges after having given them some guidelines.21 This results in a real interchange (if you prefer, dialogue) between 16  Art 134 of the Italian Code of the Administrative Process (D Lgs 2 July 2010, n 104 and sm) identifies the peremptory areas where judicial review covers also the merit of administrative decisions. 17  See Consiglio di Stato, IV, 18 October 2002, n 5714, which limited judicial review over the criteria to identify the most economically advantageous tender in the area of public procurement to their congruity and the absence of contradictions in the fact-findings and the motivation; see also Consiglio di Stato, IV, 5 August 2005, n 4168, Consiglio di Stato, IV, 31 January 2006, n 325, and Consiglio di Stato, IV, 24 April 2009, n 2536, which limited judicial review of sanctions to the manifest error, so as to avoid unjustified substitutions in the exercise of administrative discretion. 18  See Parisio (n 12) 733. 19  Some elements of the proportionality test were already present in the Fédéchar case (Case 8/55 Fédéchar. v. High Authority [1954–56] ECR 245), where the Court argued that ‘in accordance with a generally accepted rule of law such an indirect reaction by the High Authority to illegal action on the part of the undertakings must be in proportion to the scale of that action’, and in Mannesmann (Case 19/61 Mannesmann AG/ECSC High Authority [1962] ECR 357). 20  F Cafaggi, M Moraru, F Casarosa, F Fontanelli, N Lazzerini, M Mataija, G Martinico, K Podstawa and C Pitea, Final Handbook Judicial Interaction Techniques—Their Potential and Use in European Fundamental Rights Adjudication (Fiesole, Centre for Judicial Cooperation, 2014), available at www.eui.eu/ Projects/CentreForJudicialCooperation/Home.aspx. 21  C Román Vaca, ‘EU Citizenship before the CJEU: On the Importance of the Application of the Proportionality Principle’ (2013) 5 Perspectives on Federalism 1, 1. See also T Tridimas, ‘Constitutional Review of Member State Action: The Virtues and Vices of an Incomplete Jurisdiction’ (2011) 9 International Journal of Constitutional Law 737.

An Italian Perspective on the Principle of Proportionality

 227

national judges and the Luxembourg Court and this of course has favoured the circulation of interpretative patterns and the mutual influence between national and supranational law. This point is key in order to understand if and how EU law has changed the traditional argumentative patterns and techniques employed at national level. To start with, one could say that proportionality is a general principle of EU law, and, as many other general principles, it has been shaped (and reshaped) by the case law of the CJEU over the years. It is fair to say that the principle of proportionality in use at the Luxembourg Court was arguably ‘borrowed’ from the ­German legal tradition, although the classic three-step test elaborated by the German judges is rarely followed by the CJEU.22 According to some scholars, the transplant of the German principle into the supranational context has also ­benefited from the experience of the French Conseil d’État, which elaborated on the use of the bilan avantages–coûts (costs–benefit analysis).23 The paradigm of proportionality in the EU, considered in theory, follows the German model: Abstract discussions of ‘proportionality’ in EU law always describe a three-part test: (1) Does the regulation contribute to the achievement of some legitimate purpose? (2) Is there is any alternative way to achieve the purpose to the same degree with less damage to whatever Community value(s) the regulation threatens? (3) Even if the regulation is the least-damaging way to achieve whatever it achieves, is it worth it? Does the national benefit outweigh the damage to the relevant Community value(s)?24

However, as Petursson argued on the basis of a very detailed analysis provided with statistical data, the reality is quite different: the CJEU rarely reaches the third step of the test and this means that ‘the matter is settled by going through the levels of appropriateness (suitability) and necessity’.25 This conclusion reflects contradictory case law of the CJEU concerning the structure of the proportionality test. For instance, sometimes the Court seemed to follow the three-level division. In Fedesa: The Court has consistently held that the principle of proportionality is one of the general principles of Community law. By virtue of that principle, the lawfulness of the prohibition of an economic activity is subject to the condition that the prohibitory measures are appropriate and necessary in order to achieve the objectives legitimately pursued by

22  Case C-96/03 and C-97/03 A Tempelman and Coniugi THJM van Schaijk v Directeur van de ­Rijksdienst voor de keuring van Vee en Vlees [2005] ECR I-1895. 23 D Galetta, ‘Il principio di proporzionalità comunitario e il suo effetto di “spill over” negli ­ordinamenti nazionali’ (2005) 2 Nuove Autonomie 541. 24  DH Regan, ‘An Outsider’s View of Dassonville and Cassis de Dijon’ in M Poiares Maduro and L Azoulai (eds), The Past and Future of EU Law (Oxford, Hart Publishing, 2009) 465. 25 ‘Two step proportionality assessment is the most dominant method applied by the Court of ­Justice. This means that effectively the third level of the proportionality assessment, stricto sensu, is not reached.’ GT Petursson, The Proportionality Principle as a Tool for Disintegration in EU Law—Of Balancing and Coherence in the Light of the Fundamental Freedoms (PhD thesis, University of Lund, 2014) 310.

228 

Giuseppe Martinico and Marta Simoncini

the legislation in question; when there is a choice between several appropriate measures recourse must be had to the least onerous, and the disadvantages caused must not be disproportionate to the aims pursued.26

In other cases, instead, the Court of Justice followed a two-step approach, as ­exemplified by Fromançais: In order to establish whether a provision of community law is consonant with the principle of proportionality it is necessary to establish, in the first place, whether the means it employs to achieve its aim correspond to the importance of the aim and, in the second place, whether they are necessary for its achievement.27

The ‘natural’ ambiguity of proportionality is somehow amplified in the EU law context:28 as Harbo pointed out, it is very difficult to find a coherent approach in the CJEU’s use of proportionality.29 A recent confirmation of this difficulty is in the recent study of Reich,30 who identified at least four different versions of the proportionality test: 1. an autonomous balancing approach;31 2. a ‘margin of discretion’ approach;32 3. a ‘fundamental rights’ approach;33

26  Case C-331/88 The Queen v Minister of Agriculture, FEDESA Fisheries and Food and Secretary of State for Health, ex parte: Fedesa and Others [1990] ECR 4023. 27 Case 66/82 Fromançais SA v Fonds d’orientation et de régularisation des marchés agricoles (FORMA) [1983] ECR 395. 28  N Emiliou, The Principle of Proportionality in European Law: A Comparative Study (London, ­Kluwer Law International, 1996); E Ellis, The Principle of Proportionality in the Laws of Europe (Oxford, Hart Publishing, 1999). For a very original perspective see also Petursson (n 25). 29  ‘It is, in my opinion, nevertheless plausible to suggest that there are enough cases to underpin an assumption that the court is interpreting the proportionality principle in a variety of different ways, ie that the different results in cases where the proportionality principle is applied are not merely due to the cases’ different facts. The diverging interpretations of the proportionality principle are determined by the different areas of law in which it is applied, and the substance of the conflicting interests at stake.’ TH Harbo, ‘The Function of the Proportionality Principle in EU Law’ (2010) 16 European Law Journal 158, 180. 30  N Reich ‘How Proportionate is the Proportionality Principle? Some Critical Remarks on the Use and Methodology of the Proportionality Principle in the Internal Market Case Law of the CJEU’ (2011), available at: www.jus.uio.no/ifp/forskning/prosjekter/markedsstaten/arrangementer/2011/ free-movement-oslo/speakers-papers/norbert-reich.pdf (now published in HW Micklitz and B de Witte, The European Court of Justice and the Autonomy of the Member States (Antwerp and Oxford, Intersentia, 2012) 81). 31  Case C-275/92 HM Customs and Excise v Schindler [1994] ECR I-1039; Case C-208/00 Überseering BV v Nordic Construction Company Baumanagement GmbH [2002] ECR I-9919; Case C-322/01 DeutscherApothekerverband eV v 0800 DocMorris NV and Jacques Waterval [2003] ECR I-14887; Cases C-155 and 157/08 X et al v Staatssecretaris van Financien [2009] ECR I-5093. 32  Case C-405/98 Konsumentenombudsmanen v Gourmet International products [2001] ECR I-1795; Case C-350 KattnerStahlbau GmbH v Maschinenbau- und Metall- Berufsgenossenschaft [2009] ECR I-1513; Case C-42/07 Liga Portuguesa de Futebol Professional & BwinInt v Departamento de Jugos da Santa Casa da Misericordia de Lisboa [2009] ECR I-7633. 33 Case C-368/95 VereinigteFamiliapressZeitungsverlags- und Vertriebs GmbH v Heinrich Bauer Verlag [1997] ECR I-3689; Case C-71/02 Herbert Karner v Troostwij GmbH [2004] ECR I-3025; Case C-60/00 Mary Carpenter v Secretary of State for the Home Department [2002] ECR I-6279.

An Italian Perspective on the Principle of Proportionality

 229

4. a ‘quasi-legislative’ approach ‘where the Court uses the “proportionality” principle to impose measures of a legislative nature on Member states, even though formally the Court is only “interpreting” EU law and thus seemingly not invading the legislative competences of Member states’.34 Another important factor of ‘incoherence’ is the oscillating use of proportionality by the CJEU, depending on whether the measure under review was taken by EU institutions or by a Member State. In the first scenario, the CJEU rarely declares the illegitimacy of the measures, and generally acts as a third party arbiter at best. Instead, with regard to the Member States, the Court adopts a stricter stance, and often derives additional duties from the interests of integration, declaring the violation of the ‘duty of loyal cooperation’ set by the Treaties.35 The history of the proportionality test in EU law is strongly connected with the progressive extension of the scope of what is now Article 34 TFEU, prohibiting non-tariff barriers to trade in goods,36 which has been fundamental for the construction of the common market and for the ‘negative integration’ approach endorsed by the CJEU.37 Article 36 TFEU38 gives a list of legitimate aims that may justify derogations from Article 34 (and Article 35) TFEU.

34  Reich (n 30) 14. Also at 20: ‘This approach is “quasi-legislative” insofar, as certain elements of a state measure may be excessive if seen in isolation, but may still be justified in a complex policy area where different public interests are at stake and choices have to be made respecting democratic procedures which will however not be accepted by the Court, for instance on arguments of consistency as mentioned above. The Court has been criticized for substituting the national legislators’ evaluation (and solution!) of a certain problem area by its own understanding of the legislative process. I have in mind those areas where Member state actions may indeed be controversial and differ from a strictly liberal economic model, but where careful analysis will find some “hidden” yet reasonable justification which however was discarded by the CJEU on proportionality reasons.’ See Case C-81/87 R v Treasury and Commissioners of Inland Revenue ex parte Daily Mail and General Trust [1988] ECR 5483; Case C-212/97 Centros Ltd v Erhvervs- og Selskabsstyrelsen [1999] ECR I-1459. 35  ‘Proportionality in the narrow sense—stricto sensu—is, according to the claim, applied whenever the court finds it suitable in order to promote the desired outcome. Accordingly, the proportionality analysis conducted by the court is not objective in the sense that it is value-neutral. On the contrary, the analysis is informed by a very strong substantial bias, namely that of promoting European integration.’ Harbo (n 29) 172. 36  For the first case law on this provision see, above all Case 8/74 Procureur du Roi v Benoît and ­Gustave Dassonville [1974] ECR 837; Case 104/75 Officier van Justitie v De Peijper [1976] ECR 613; Case 120/78 Cassis de Dijon [1979] ECR 649. 37  ‘After the consolidation of the CJEU’s constitutional doctrines of supremacy and direct effect, the emergence of proportionality balancing as a master technique o… … [referring to what is now Art 34 TFEU] No other provision of the Rome Treaty has been more implicated in market-building, and in defining the relationship between the scope and authority of European law, on the one hand, and the regulatory autonomy of the Member States, on the other.’ A Stone Sweet and J Mathews, ‘Proportionality Balancing and Global Constitutionalism’ (2008) 47 Columbia Journal of Transnational Law 72, 141–42. 38  Art 36 TFEU reads: ‘[T]he provisions of Articles 34 and 35 shall not preclude prohibitions or restrictions on imports, exports or goods in transit justified on grounds of public morality, public policy or public security; the protection of health and life of humans, animals or plants; the protection of national treasures possessing artistic, historic or archaeological value; or the protection of industrial and commercial property. Such prohibitions or restrictions shall not, however, constitute a means of arbitrary discrimination or a disguised restriction on trade between Member States.’

230 

Giuseppe Martinico and Marta Simoncini

In the area of freedom to provide services, as shown in Gebhard,39 the CJEU— when reviewing the compatibility of a restriction to the fundamental freedoms with EU law—takes into account the following factors: 1. 2. 3. 4.

the restriction must be applied in a non-discriminatory manner; it must be justified by imperative requirements; it must be suitable for achieving the pursued purpose; it must not go beyond what is necessary to achieve it.

If the first peculiarity of the proportionality test in EU law is its ‘changing nature’ (depending on the matter to which it applies, and on the subjects involved), another noteworthy aspect is its bearing on the specific interpretative cooperation dynamics that take place between judges in the EU, notably in respect of the use of the preliminary ruling machinery. This device is set up by Article 267 TFEU and has had a crucial role in the ‘judicial construction of Europe’.40 Scholars have usually devoted their attention to the ‘upward phase’ of this mechanism, ie when the question (formulated by the referring judge) is referred to the CJEU, and the CJEU issues the ruling thereafter. In the literature, it has also been pointed out that the hypothetical division of labour between national judges and the CJEU (the application of EU law is for the former, while interpretation is for the latter)41 has known many exceptions in practice. Proportionality, instead, offers a quite interesting example of ‘downward’ cooperation between the CJEU (which rules on the interpretation/validity of an EU act) and the referring judge (who receives from Luxembourg the applicable instruction to resolve the main proceedings). In many cases, indeed, the CJEU, after ‘constructing’ the test of proportionality, hands it over to the ordinary judge. This is because, in the logic of judicial subsidiarity, the referring judge is the real master of the proceedings. He knows the factual background of the dispute, as well as the national context and sensitivity in which the allegedly discriminatory or restrictive measure operates. The application of the proportionality test in the field of fundamental rights offers a graphic example of this delegating practice. In Schmidberger42 the CJEU distinguished between two groups of fundamental rights: absolute rights (which admit of no restriction) and other fundamental rights.43 With respect to the 39  Case C-55/94 Gebhard v Consiglio dell’Ordine degli Avvocati e procuratori di Milano [1995] ECR I-4165. 40  A Stone Sweet, The Judicial Construction of Europe (Oxford, Oxford University Press, 2004). 41  G Bebr, ‘Preliminary Rulings of the Court of Justice: Their Authority and Temporal Effect’ (1981) 18 Common Market Law Review 475, 480. 42  Case C-112/00 Schmidberger [2003] ECR I-5659. 43  Case C-112/00 Schmidberger [2003] ECR I-5659, para 80: ‘Thus, unlike other fundamental rights enshrined in that Convention, such as the right to life or the prohibition of torture and inhuman or degrading treatment or punishment, which admit of no restriction, neither the freedom of expression nor the freedom of assembly guaranteed by the ECHR appears to be absolute but must be viewed in relation to its social purpose. Consequently, the exercise of those rights may be restricted, provided that the restrictions in fact correspond to objectives of general interest and do not, taking account of the

An Italian Perspective on the Principle of Proportionality

 231

second category of rights, the Court of Justice admitted the necessity to evaluate through a case-by-case approach the proportionality of their possible restrictions. More recently, the Court confirmed this approach in Laval44 and Viking.45 In these cases the Court recognised the fundamental right to collective action as an integral part of EU law. This right can justify restrictions on the fundamental freedom of establishment or on the freedom to provide services guaranteed under the EU Treaty, when it is exercised to protect workers and their conditions of employment. Significantly, the Court limited itself to declare that collective action restricting economic freedoms is legal ‘only if it pursues a legitimate aim such as the protection of workers’ and left the final decision on the legitimacy of the conduct challenged for the national courts to decide, as they are better positioned to do the balancing between the rationale of market integration and the value of the protection of workers’ rights. The Viking judgment is emblematic in this sense, as it concerned a dispute hinging on the balance between one of the fundamental freedoms established by the EU treaties and those interests pursued by collective action-negotiation. In this case, the Court reconstrued the proportionality test, only to hand it over to the referring judge, observing that ‘it is ultimately for the national court, which has sole jurisdiction to assess the facts and interpret the national legislation, to determine whether and to what extent such collective action’ (i) has a legitimate aim; (ii) is effective and (iii) is not excessively detrimental for economic freedoms. As seen, although the supranational idea of proportionality was derived from German law the Court of Justice does not always follow the approach of the ‘German masters’ to the structure of the test. Moreover, EU law proportionality has acquired a very diversified physiognomy depending on the area touched by the Luxembourg Court. We shall come back to these considerations when verifying the influence of supranational proportionality over the national legal system. We shall start with the administrative judges’ approach and move to the case law of the Italian Constitutional Court.

IV.  The Influence of EU Law over National Administrative Justice The application of the proportionality principle in Italian administrative law has been considerably influenced by the development of this principle in EU law and aim of the restrictions, constitute disproportionate and unacceptable interference, impairing the very substance of the rights guaranteed.’ See to this effect also Case C-62/90 Commission v Germany [1992] ECR I-2575, para 23 and Case C-404/92P X v Commission [1994] ECR I-4737, para 18. 44 

Case C-341/05 Laval un Partneri Ltd/Svenska Byggnadsarbetareförbundet e a [2007] ECR I-11767. Case C-438/05 The International Transport Workers’ Federation and The Finnish Seamen’s Union [2007] ECR I-10779. 45 

232 

Giuseppe Martinico and Marta Simoncini

especially in the case law of the CJEU. The Europeanisation of protection played a key role not only in the emergence of proportionality as an autonomous ground of review, but also when shaping the proportionality test. As EU case law elaborated on the three-step test developed in German law, the evaluation of suitability, necessity and proportionality in the strict sense passed throughout national legal orders and also influenced Italian administrative law in the application of EU law.46 The steady assimilation of EU law into the domestic legal order contributed to distinguishing the principle of proportionality from the principle of reasonableness, so that the three-step test has been increasingly applied in the judicial review of administrative action.47 This especially happened in conjunction with the completion of internal market, when EU law started to affect concretely the structure of the domestic market in areas of administrative action such as the regulation of public utilities and public procurement. In these domains, national administrative law has been steadily exposed to the implementation of the internal market and its concepts and tools are oriented to the protection of operators in the internal market. Since the 1990s, the principle that impartiality shall guide fact-finding and shape the administrative decision-making process has somehow embedded the concept of competition in the internal market. Originally, in the General Accounting Law of the Italian state the principle of impartiality was aimed to protect the (economic) interest of the concerned public administration and only the principle of par condicio intended to protect private interests during the administrative action. EU law has emphasised the need to protect private parties competing on the market. Moreover, national administrations have also to deal with the principles of non-discrimination and competition when applying the principle of impartiality. In this context, during the 1990s, proportionality emerged as a principle which was able to protect those private interests exposed to administrative action more effectively than the reasonableness test.48 Unlike the reasonableness test, the three-step test of proportionality focuses more on the burden imposed on private parties when carrying out administrative activities. Setting the threshold of public action therefore started to involve a proportionality test which shall

46 

On the diffusion of the EU proportionality principle in the Member States, see Galetta (n 23). See in particular TAR Toscana, Firenze, I, 6 March 2001, n 381; TAR Puglia, Bari, II, 7 June 2001, n 2405; TAR Lazio, III, 2 February 2007, n 777; Consiglio di Stato, IV, 1 October 2004, n 6410; Consiglio di Stato, VI, 17 April 2007, n 1746; Consiglio di Stato, VI, 8 February 2008, n 424; Consiglio di Stato, VI, 10 March 2009, n 1420; Consiglio di Stato, IV, 11 January 2010, 19. In some other cases, however, the proportionality review has concerned only the suitability and the necessity of administrative decisions; see Consiglio di Stato, IV, 22 June 2004, n 4381; Consiglio di Stato, IV, 5 October 2004, n 6490; Consiglio di Stato, IV, 28 February 2005, n 702; Consiglio di Stato, IV, 22 March 2005, n 1195; Consiglio di Stato, V, 14 April 2006, n 2087. In the literature see DU Galetta, ‘Il principio di proporzionalità’ in M Renna and F Saitta (eds), Studi sui principi del diritto amministrativo (Milano, Giuffrè, 2012) 409–10. 48 See DU Galetta, ‘El principio de proporcionalidad en el Derecho Público italiano’ (1998) 5 Cuadernos de Derecho Publico 299. 47 

An Italian Perspective on the Principle of Proportionality

 233

take into account the reach of the public limitation of private rights. In 2000 in a case concerning the selling of the state’s shareholding in an airport-managing company, the Consiglio di Stato recognised that when EU law applies, proportionality as a general principle of the EU, based on the evaluation of these three criteria, shall apply as a limit to the exercise of administrative powers also to the Italian administration.49 Subsequently, as an almost inevitable consequence of the constitutional principle of equality, the proportionality principle has been applied also to those cases of domestic law that did not concern the implementation of EU law.50 Administrative judges therefore permeated the entire domain of administrative law with the proportionality principle designed under EU law. However, the Consiglio di Stato also stressed that proportionality exists as a general principle of Italian administrative law alongside its nature as an EU law principle.51 As soon as Italian administrative judges applied the principle to domestic cases, the legislator also started to include proportionality among the general principles which shall inspire administrative action when implementing not only EU law, but also national law.52 Against the EU law backdrop, the principle of proportionality started to consolidate its appearance in shaping Italian administrative law. The reform of administrative procedure law in 2005 gave a further impulse to the implementation of the proportionality principle under Italian administrative law. According to Article 1 of the reformed law, Italian public administrations shall comply with the principles of EU law for administrative measures whether these are based on EU law or not.53 As a consequence, the proportionality principle as conceived in EU law shall inspire both administrative action and judicial review by administrative courts. This reform has formally admitted the proportionality principle among the general principles which shall guide any administrative action and has contributed to the partial renovation of the principles of Italian administrative law.54 Since 2005, the application of the proportionality principle 49  See Consiglio di Stato, VI, 1 April 2000, n 1885. In the literature see DU Galetta, ‘Una sentenza storica sul principio di proporzionalità con talune ombre in ordine al rinvio pregiudiziale alla Corte di giustizia’ (2000) 2 Rivista italiana di diritto pubblico comunitario 396. See also Consiglio di Stato, V, 14 April 2006, n 2087, which recognised proportionality as a general principle of Italian administrative law alongside its nature as an EU law principle. In the literature see DU Galetta, ‘La proporzionalità quale principio generale dell’ordinamento’ (2006) 10 Giornale di diritto amministrativo 1106, 1107–11. 50  See Galetta (n 6) 223; Galetta (n 49) 1110. In the early case law, see in particular TAR Lombardia, Milano, IV, 1 December 2004, n 6095 and 6096; TAR Puglia, Lecce, II, 31 March 2004, n 2216; TAR Sicilia, Palermo, II, 18 November 2004, n 2590; TAR Veneto, Venezia, III, 10 March 2005, n 850, 851 and 852. 51  See also Consiglio di Stato, V, 14 April 2006, n 2087. In the literature see Galetta (n 49) 1107. 52  See in particular Art 178(3), d Lgs 3 April 2006, n 152 (the environmental law code) and Art 23, L 28 December 2005, n 262, on the protection of consumers in financial markets. In this regard, see DU Galetta, ‘Il principio di proporzionalità’ in MA Sandulli (ed), Codice dell’azione amministrativa (Milano, Giuffrè, 2011) 125–26. 53  See Art 1 (1), L 7 August 1990, n 241 as amended by L 11 February 2005, n 15. 54  See Consiglio di Stato, VI, 17 April 2007, n 1736, which extended the application of the proportionality principle from the content of sanctions to all the measures limiting individual rights, and interpreted it on the basis of the three-step test.

234 

Giuseppe Martinico and Marta Simoncini

has entirely covered the course of administrative action and it entails the choice of the means that are least burdensome for private parties, as well as the limitation to the exercise of administrative discretionary powers. The proportional protection of individual rights in the course of administrative action therefore becomes a key principle of administrative law which affects the substance of administrative decisions and requires adequate procedural arrangements in order to be implemented. A clear example is offered by recent case law of Lazio Regional Administrative Court (so called TAR, Tribunale Amministrativo Regionale), which questioned highly discretionary administrative decisions of the Rome municipality on the grounds, among others, of the proportionality principle. TAR Lazio accepted the pleas of an association for consumer protection, Codacons, which complained about the lack of reasonableness and consistency in the discretionary choices of the administration when setting tariffs for reducing urban traffic. In a case about the increase of tariffs for parking, TAR Lazio annulled decision 48/2014 of the Rome municipality to the extent that it did not fix the tariff according to the parking demand and different categories of consumers.55 Similarly, in another case about the thresholds of the tariffs for Limited Traffic Zones, the administrative judge annulled decision 119/2014 of the municipality which increased the tariffs with an act of general application as far as it did not satisfy the proportionality principle, as well as because the procedure of its adoption was not grounded on adequate and accurate fact-findings.56 Even if according to Italian law, acts of general application do not require formal motivation, TAR Lazio observed that the administration did not provide sufficient information to the court about how it proceeded when making the decision and how it implemented the proportionality principle. The administration only alleged that tariffs in Rome were lower than in other cities in Italy and abroad (particularly, Milan and London), without supporting data showing an accurate comparison of the structural conditions of traffic regimes and urban mobility and without demonstrating the fairness of its choice through adequate fact-finding. Even if in these cases TAR Lazio did not engage in a detailed three-step test to assess the proportionality of administrative acts of general application in question, it nevertheless showed that the idea of proportionality of administrative action shall permeate all kinds of administrative action and applies also to discretionary acts which implement political goals of the administration. By asking for further data about the legitimacy of political choices concerning urban traffic management, TAR Lazio somehow expanded the area of judicial control through the proportionality test, touching lightly upon a merit control. The application of the proportionality test shows its potential to force and shape also the more ­political action of public administrations.

55  56 

See TAR Lazio, II; 17 March 2015, n 4233. See TAR Lazio, II, 3 December 2015, n 3666.

An Italian Perspective on the Principle of Proportionality

 235

V.  The Case Law of the Italian Constitutional Court Proportionality is also employed by the Italian Constitutional Court quite frequently, but when doing so this Court somehow absorbs the proportionality test in the idea of reasonableness or conceives it as a part of the balancing exercise. This point has been confirmed by Justice Cartabia, among others, in a speech given in a trilateral colloquium organised by the Italian, Spanish and Portuguese constitutional courts in 2013.57 The fact that the Constitutional Court has relied on proportionality when reviewing national legislation should be seen as part of the ‘demythologisation of power’. The fact that even the will of the people expressed by a democratically elected Parliament may be reviewed by judges is one of the building blocks of modern constitutionalism. Public power (including the legislator) may not discriminate or do wrongs. When doing so it will be subject to the scrutiny of the Constitutional Court. More in general, according to this scheme, public power has to give reasons or, better, has to make reasonable choices. The Italian case does not depart from these general considerations applicable to all the post-totalitarian constitutional experiences. This premise permits one to understand the strong connection between reasonableness and non-discrimination present (also) in the Italian constitutional case law. It is not by coincidence, in fact, that reasonableness was originally conceived as connected to the principle of non-discrimination but has evolved even beyond the parameters of Article 3 of the Constitution, which contains the principles of non-discrimination and equality.58 In a nutshell, one of the main features of the proportionality test in the case law of the Italian Constitutional Court is that it tends to conflate proportionality and reasonableness. Even in this case one can detect a terminological ­uncertainty (‘incertezza terminologica’).59 As Justice Cartabia pointed out, there has been ‘a promiscuous use of terms like rationality, reasonableness and proportionality, but also adequacy, consistency, not arbitrariness, relevance and

57  M Cartabia, ‘I principi di ragionevolezza e proporzionalità nella giurisprudenza costituzionale italiana’, Trilateral colloquium, Palazzo della Consulta, Rome 24–26 October 2013, available at www. cortecostituzionale.it/documenti/convegni_seminari/RI_Cartabia_Roma2013.pdf. 58  On this evolution, see G Scaccia, Gli strumenti della ragionevolezza nel giudizio (Milano, Giuffrè, 2000) and A Morrone, Il custode della ragionevolezza (Milano, Giuffrè, 2001). See also L Paladin, ‘Ragionevolezza (principio di)’ in Enciclopedia del diritto (Milano, Giuffrè, 1997) vol 1, ‘aggiornamento’ 899; G Zagrebelsky, ‘Su tre aspetti della ragionevolezza’ in Il principio di ragionevolezza nella giurisprudenza della Corte costituzionale. Riferimenti comparatistica (Milano, Giuffré, 1994) 179, M Sandulli, ‘Il principio di ragionevolezza nella giurisprudenza costituzionale’ (1975) 3 Diritto e Società 561. 59  M Cartabia, ‘I principi di ragionevolezza e proporzionalità nella giurisprudenza costituzionale italiana’, Trilateral colloquium, Palazzo della Consulta, Rome 24–26 October 2013, available at www. cortecostituzionale.it/documenti/convegni_seminari/RI_Cartabia_Roma2013.pdf.

236 

Giuseppe Martinico and Marta Simoncini

many ­others, as they were synonymous’.60 An emblematic example of this is given by decision 2/1999 whereby the Court argued that: ‘[T]he automatism of the disciplinary sanction is unreasonable, contrasting with the principle of proportionality, which is at the basis of the rationality which shapes the principle of equality.’61 When comparing the ‘toolbox’ of the Italian Constitutional Court with that of other Supreme or Constitutional judges, one must make clear that the case law of the Corte Costituzionale does not really distinguish between the proportionality and reasonableness tests. The Court made this explicit in decision 220/1995 by defining proportionality (there named ‘principio di proporzione’) as a ‘direct expression of the general canon of reasonableness’.62 The reasonableness concept itself been defined by the Constitutional Court both as a form of ‘practical rationality’63 and as a sort of legislative excès de p­ ouvoir,64 but the latter approach is not univocal and has been excluded by the Court itself in other decisions.65 However, these definitions reveal one of the most frequent uses of reasonableness in the constitutional case law, since this concept is often invoked to tackle those ‘legislative automatisms’66 where the legislator fails to distinguish between situations that must be treated differently. Over the years, the reasonableness test has changed its nature in the case law of the Italian Constitutional Court67 and today works as a pillar in the case law of the Constitutional Court combined with all the other principles of the ­Italian ­Constitution.68 Because of that, reasonableness has acquired a ­‘pervasive ­character’69 in the case law of the Corte Costituzionale, being invoked by the Court

60  Cartabia, ibid. In Italian: ‘Nella giurisprudenza costituzionale italiana si nota un uso p ­ romiscuo di termini come razionalità, ragionevolezza, proporzionalità, ma anche adeguatezza, coerenza, ­congruenza, non arbitrarietà, pertinenza e molti altri, quasi si trattasse di sinonimi.’ 61 Italian Constitutional Court (Corte Costituzionale), decision no 2/1999, available at: www. cortecostituzionale.it. 62  Italian Constitutional Court (Corte Costituzionale), decision no 220/1995, available at: www. cortecostituzionale.it 63 Italian Constitutional Court (Corte Costituzionale), decision no 172/1996, available at: www.cortecostituzionale.it 64  Italian Constitutional Court (Corte Costituzionale), decision no 313/1995, available at: www. cortecostituzionale.it 65 Italian Constitutional Court (Corte Costituzionale), decision no 37/1969, available at: www. cortecostituzionale.it. See the reconstruction offered by M Fierro, ‘La ragionevolezza nella giurisprudenza costituzionale italiana’ (2013) available at: www.cortecostituzionale.it/documenti/convegni_ seminari/STU_249_Relazioni.pdf 66  Cartabia (n 59). Justice Cartabia recalls the following decisions of the Italian Constitutional Court: decisions no 303/ 1996, 240/ 1997, 329/ 2007, 31/ 2012, all available at: www.cortecostituzionale.it. 67  On its vagueness, see L Paladin, ‘Esiste un “principio di ragionevolezza” nella g ­ iurisprudenza ­costituzionale?’ in Il principio di ragionevolezza nella giurisprudenza della Corte costituzionale. ­Riferimenti comparatistici (Milano, Giuffré, 1994) 163. 68  Cartabia (n 59). 69 ibid.

An Italian Perspective on the Principle of Proportionality

 237

when ­balancing fundamental rights and interests.70 This way the Court makes clear the fundamental rights are not absolute, in the sense that they limit each other and have to coexist. For instance, in decision 264/2012 the Court clarified that the protection of fundamental rights must be ‘systematic and not fragmented into a series of rules that are uncoordinated and potentially conflict with one another’.71 More recently, in decision 83/2013 it added that: All fundamental rights protected by the Constitution are mutually related to one another and it is thus not possible to identify any one of them in isolation as prevailing absolutely over the others. … If this were not the case, the result would be an unlimited expansion of one of the rights, which would ‘tyrannise’ other legal interests recognised and protected under constitutional law, which constitute as a whole an expression of human dignity.72

Even when the Italian Constitutional Court employs the word proportionality, the test is not carried out in three or four steps. In order to stress this point Cartabia argued that ‘there is no sign of this argumentative pattern’ in the case law of the Italian Constitutional Court.73 Before moving to the interaction between EU law and the use of proportionality by the Italian Constitutional Court, it should be made clear that the European Court that seems to have influenced the Corte Costituzionale to a greater extent in this ambit is not the Luxembourg one but, rather, the Strasbourg one, especially after the openness shown by the Italian Court in decisions 348/2007 and 349/2007.74 As Justice Gallo, former President of the Italian Constitutional Court, wrote in a text prepared for a public meeting in Brussels held on 24 May 2012, recently the exchange of views between the Italian Constitutional Court and the Strasbourg

70  On the idea of ‘bilanciamento’ in the Italian case-law, see R Bin, Diritti e argomenti: il b ­ ilanciamento degli interessi nella giurisprudenza costituzionale (Milano, Giuffrè, 1992); A Vespaziani, Interpretazioni del bilanciamento dei diritti fondamentali (Padova, Cedam, 2002). 71  Italian Constitutional Court (Corte Costituzionale), decision no 264/2012, available at: www. cortecostituzionale.it 72 ‘As is the case under other contemporary democratic and pluralist constitutions, the Italian Constitution requires that an ongoing reciprocal balance be struck between fundamental principles and rights, and that none of them may claim absolute status.’ Italian Constitutional Court (Corte Costituzionale), decision no 83/2013, available at: www.cortecostituzionale.it. 73  Cartabia (n 59). 74  Italian Constitutional Court (Corte Costituzionale), decisions no 348 and 349/2007, available at www.cortecostituzionale.it. For a detailed analysis of the judgments, see F Biondi Dal Monte and F Fontanelli, ‘The Decisions No 348 and 349/2007 of the Italian Constitutional Court: The ­Efficacy of the European Convention in the Italian Legal System’ (2008) 9 German Law Journal 889; O Pollicino, ‘Italy—Constitutional Court at the Crossroads between Constitutional Parochialism and Co-operative Constitutionalism. Judgments No 348 and 349 of 22 and 24 October 2007’ (2008) 4 European C ­ onstitutional Law Review 363. On the influence of the case law of the Strasbourg Court on the Italian administrative judges, see Sau (n 10) 36. More in general, see also R Conti, Diritto di proprietà e CEDU. Itinerari giurisprudenziali europei. Viaggio fra Carte e Corti alla ricerca di un nuovo statuto proprietario (Roma, Aracne, 2012).

238 

Giuseppe Martinico and Marta Simoncini

Court has become more and more frequent.75 In principle this gives an added value to the protection of fundamental rights in Europe. However, as Justice Gallo pointed out: ‘[T]he work of transposition of the case-law of the ECtHR into the national legal order has not been easy.’76 These words represent a certain tension that has emerged over recent years. Since 2007, the Italian Constitutional Court has acknowledged that a particular importance attaches to the interpretative function performed by the Strasbourg Court and this was a consequence of the particular status of the ECHR recognised by the Corte Costituzionale.77 Some of these disagreements have been also due to a different approach to striking the balance and to handling the proportionality principle.78 It is more difficult to find explicit exchanges between the Constitutional Court and the Luxembourg Court concerning the different ways to understand balancing and proportionality, although the argument as to whether the type of balancing exercise before a national constitutional court differs from that struck by supranational courts could be recalled even in this ambit. This can be explained considering the different understanding of principles that are, on paper, shared by the national and supranational order: the principle of non-discrimination is perhaps the most telling example in this respect. This is also confirmed by the cases on reverse discrimination where the case law of the Italian

75  F Gallo, ‘Rapporti fra Corte costituzionale e Corte EDU’, Brussels, 24 May 2012, available at www. cortecostituzionale.it/documenti/relazioni_internazionali/RI_BRUXELLES_2012_GALLO.pdf. 76 ibid. 77  In a nutshell, according to the Italian Constitutional Court, the Convention has a super-primary value (ie its normative ranking is halfway between statutes and constitutional norms). This is confirmed by the fact that, in some cases, the ECHR can serve as an ‘interposed parameter’ for the constitutional review of primary laws, since the conflict between them and the ECHR can entail an indirect violation of the Constitution, namely of its Art 117, para 1, which reads: ‘Legislative power belongs to the state and the regions in accordance with the Constitution and within the limits set by European Union law and international obligations.’ Since Art 117 recalls international obligations, a conflict between a national piece of legislation and the ECHR can be solved by considering the ECHR as an external part of the standard employed by the Corte Costituzionale to review the constitutionality of domestic norms. In other words, by interposed norm (norma interposta) scholars mean those norms that are sub-constitutional from a formal point of view but somehow indirectly recalled by a Constitutional provision and thus able to ‘complement’ the Constitutional text, in a sense that their violation by a legislative provision can amount to an indirect violation of the Constitution. Since Art 117 of the Constitution recalls that the legislator has to respect international obligations, a breach of the ECHR may lead to an indirect violation of Art 117. The constitutional favour accorded to the ECHR implies the obligation to interpret national law in light of the ECHR’s norms. At the same time, this does not imply that the ECHR has a constitutional value; on the contrary, the ECHR has to respect the Constitution. As we will see when dealing with the second judicial practice, according to the Italian Constitutional Court the ECHR cannot be treated domestically like EU law. 78  See eg Italian Constitutional Court (Corte Costituzionale), decision no 264/2012, available at: www.cortecostituzionale.it. On this see B Randazzo, ‘Il sindacato sulla ragionevolezza della legge e lo scrutinio di proporzionalità sul margine di apprezzamento riservato allo Stato in rapporto a misure generali aventi natura legislativa. Aspetti problematici del dialogo tra le Corti’ (2013), available at: www.cortecostituzionale.it/documenti/convegni_seminari/STU_249_Relazioni.pdf.

An Italian Perspective on the Principle of Proportionality

 239

Constitutional Court on the principle of equality79 has been influenced by the jurisprudence of the Luxembourg Court.80 Moving back to proportionality understood stricto sensu, an interesting example of the use of proportionality by the Italian Constitutional Court is given by the decision on the law (69/2005) implementing the Council Framework Decision on the European arrest warrant.81 The case concerned the different treatment of Italians and other EU citizens. In this case, the Corte Costituzionale declared the partial unconstitutionality of Article 18(1)(r) of law 69/2005 ‘insofar as it does not provide that nationals of another Member State of the European Union who are lawfully and effectively resident or staying in Italy should also not be surrendered, in order to enforce the custodial sentence in Italy in accordance with national law’.82 There is a short passage in this decision where the proportionality test is recalled, but not fully articulated according to the three-step model:83 The prohibition on discrimination on the grounds of nationality does permit different treatment between the nationals of one Member State of the Union and those of another Member State. However, the difference in treatment must have a legitimate and reasonable justification, and be subject to a rigorous proportionality test as to the objective pursued.84

In light of this consideration and taking into account the case law of the CJEU, the Italian Constitutional Court declared the unconstitutionality of Article 18, because of the unreasonable (and thus ‘non proportional’,85 according to the terminology of the Corte Costituzionale) justification of the choice made by the legislator.

79 See eg Italian Constitutional Court (Corte Costituzionale), decisions no 249/1995, 61/1996, 443/1997, available at: www.cortecostituzionale.it. 80 Case 35-36/82 Morson and Jhanjan [1982] ECR 3723; Case 229/83 Leclerc [1985] ECR 1; Case 355/85 Cognet [1986] ECR 3231; Case 90/86 Zoni [1988] ECR 4285; Case 33/88 Allué I [1989] ECR 1591; Case C-259/91 C-331/91 and C-332/91 Allué II, [1993] ECR I-4309. 81  Italian Constitutional Court (Corte Costituzionale), decision no 227/2010, available at: www. cortecostituzionale.it. On this decision, see F Fontanelli, ‘Commento alla sentenza della Corte Costituzionale n. 227 del 2010 in tema di mandato di arresto europeo’ (2011) Giornale di diritto amministrativo 47. 82  Italian Constitutional Court (Corte Costituzionale), decision no 227/2010, available at: www. cortecostituzionale.it. 83  Cartabia (n 59). 84  Italian Constitutional Court (Corte Costituzionale), decision no 227/2010, available at: www. cortecostituzionale.it 85  ‘In particular, a requirement for residence for a continuous period of five years for non-nationals was held by the Court of Justice not to go beyond what is necessary to attain the objective of ensuring reintegration into the executing Member State (Wolzenburg judgment, paragraph 73). However, in contrast to the Dutch law transposing the Framework Decision on the EAW, which was at issue in the above case, the provision contested here does not restrict the equal treatment of nationals of another Member State of the Union with Italian nationals with regard, for example, to the duration of the residence aut similia, but precludes at root the possibility that the national of another Member State may benefit from the refusal to surrender, and accordingly the enforcement of the sentence in Italy. This amounts to subjective discrimination against the citizen of another Member State of the Union

240 

Giuseppe Martinico and Marta Simoncini

This is perhaps the most important case showing the influence of the case law of the CJEU on the use of the proportionality test in the jurisprudence of the Italian Constitutional Court.

VI.  Final Remarks The Italian route to the proportionality principle seems to be paved with an endogenous understanding of the concept of proportionality and its derived understanding from exogenous legal orders, namely the EU (and, as mentioned in the last paragraph, the ECHR). If in Italy the concept is originally associated with the principle of reasonableness, the European influence contributed to emancipating the original meaning and to developing an autonomous construction of the proportionality principle. This stratified nature of the principle, however, is still apparent and it is reflected in the application that the proportionality test has in Italian case law. As this chapter has illustrated, administrative and constitutional judges have elaborated and interpreted the principle in different ways. Administrative courts seem to be more exposed to the application of the principle as interpreted in the CJEU’s case law as a graduation in the use of public powers, whereas the Constitutional Court has developed a more autonomous interpretation which relies more heavily on the reasonable balance of rights. Such a difference is probably due to the different approach of national judges to EU case law.86 Administrative judges have to take into account supranational case law as long as public administrations operate in the competitive internal market and cannot alter competition rights without infringing EU law. Impartiality of administrative action therefore comes close to protection of competition in the internal market, which requires the subordination of the achievement of public goals to the proportionate protection of individual rights on the market. Making compliance with the principles of EU law mandatory law for national public administrations goes exactly in this direction. Historically, the Italian Constitutional Court has had a tendency to protect the autonomy of its judgment. This is still true even if more recently the Corte on the grounds that he is a foreign national which, absent a reasonable justification, is not proportional.’ Italian Constitutional Court (Corte Costituzionale), decision no 227/2010, available at: www. cortecostituzionale.it 86  There are also broader cultural reasons which can explain this difference. After all, the law—la legge—(the sole object of the review of constitutionality realised by the Italian Constitutional Court) is still culturally conceived as the will of the people and this can explain why constitutional courts still tend to leave a certain margin to parliaments. This is somehow connected to the ‘counter majoritarian difficulty’ explained by Bickel. In this respect the control over the activity of public administration is still perceived as qualitatively different because the element of the counter-majoritarian difficulty does not seem to be present in this case. See AM Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of Politics 16 (Indianapolis, Bobbs-Merrill, 1962) 18.

An Italian Perspective on the Principle of Proportionality

 241

Costituzionale has shown an increasing openness towards the preliminary ­ruling mechanism governed by Article 267 TFEU,87 accepting direct and formal dialogue with the Luxembourg Court.88 This is not peculiar to the Italian Constitutional Court, but it is a consolidated strategy to protect judicial—and national—sovereignty against supranational law experiences. This autonomy somehow is expressed also in the elaboration of the general principles and in the interpretation of the proportionality test. The Italian Constitutional Court has remained more faithful to the original understanding of the proportionality test in Italy as a sub-set of the reasonableness test and has been less influenced by the supranational development that the principle faced. Even if administrative and constitutional courts seem to go in different directions one cannot deny the existence of a common understanding of the concept of proportionality which is deeply rooted in the general principle of reasonableness. In fact, as seen, administrative judges sometimes mix up the two principles in question and conduct a reasonableness test instead of a proportionality one. When applying the three-step test following the German model, the difficulty in ­distinguishing the single steps can be due to the autonomous conception that ­permeates judicial understanding under national law as well as to the approach of the Constitutional Court which continues to show the viability of the original interpretation. The intensity of the proportionality test varies according to the more or less Europeanised character of judicial review and the policy framework of reference is not unrelated to the judicial choice. In this regard, proportionality confirms its kaleidoscopic nature as it is able to swing between a less intrusive check of the public measures based on reasonableness and a stronger check which is also able to touch upon the merit of public policies. This makes proportionality a ­strategic tool in the hands of judges, who can articulate their decision in ­context-specific ways which make the intensity of the judicial review less predictable. The ­ stratified nature of the principle—built upon a national tradition and supranational ­ ­ influences—operates as a double-edged sword that is able to ­support different standards of judicial review, which nevertheless have equal ­legitimacy in Italian law.

87 Italian Constitutional Court (Corte Costituzionale), decisions no 102/2008 and 103/2008, available at: www.cortecostituzionale.it. The preliminary reference was raised during principaliter proceedings. More recently the Italian Constitutional Court extended its revirement to incidenter proceedings, see: Italian Constitutional Court (Corte Costituzionale), decision no 207/2013, www.governo.it/ Presidenza/CONTENZIOSO/comunicazione/allegati/ordinanza_207_2013_completa.pdf. 88  Although the preliminary ruling mechanism does not exhaust all the possible ways of communication between constitutional courts and the CJEU: see G Martinico, ‘Judging in the Multilevel Legal Order: Exploring The Techniques of ‘Hidden Dialogue’ (2010) 21 King’s Law Journal 257.

242 

14 The Application of Proportionality in Denmark in the Light of European Legal Integration HELLE KRUNKE1

I. Introduction The purpose of this chapter is to draw a picture of the meeting between three legal systems in the field of proportionality: the Danish legal system, the EU legal ­system and the legal system of the European Convention of Human Rights (ECHR). The idea is to identify differences and tensions between (i) the EU principle of proportionality and the Danish principle of proportionality and (ii) the ECHR principle of proportionality and the Danish principle of proportionality. Further, the impact of the EU principle of proportionality and the ECHR principle of proportionality on the Danish legal system will be studied. Finally, trends in the future application of the principle of proportionality in light of European legal integration will be discussed. In the conclusion, observations from Member State level will be applied to a more general discussion of which way EU integration is moving. As regards delimitation, it is not the aim to provide a detailed and exhaustive analysis of the differences of the three principles of proportionality and the impact of the European principles of proportionality on Danish law. It would be possible to write several articles on these matters. Rather, the ambition is to provide the reader with an overview and to give a broader picture of differences, impact and future trends of European integration in the field of proportionality. Furthermore, the main focus is not on the relationship between the EU principle of proportionality and the principle of proportionality in the ECHR.

1  The author would like to thank research student Jens Christian Dalsgaard for assistance with s­earching for legal sources and researching information for the chapter, and research student Anah Maskell Knudsen for house style assistance.

244 

Helle Krunke

II.  The Supreme Court and National Legal Principles in Light of EU Integration Lately the Danish Supreme Court has shown great concern as to whether certain interpretations by the European Court of Justice in the field of labour law will result in judgments that violate national legal principles such as rule of law, legal foreseeability and legal certainty when applied in the national context.2 This is clear in recent case law from the Danish Supreme Court and in a recent preliminary question sent to the European Court of Justice.3 Furthermore, the former President of the Danish Supreme Court, Børge Dahl, has on several occasions expressed such views—among others at the FIDE conference in Copenhagen in 2014: I have to admit that over the years the development of law through the practice of the European Court of Justice have gone beyond the limitations drawn by the notion of legal foreseeability and certainty under Danish law. Time and again, we are confronted with European judgments finding European harmonisation to have gone further than our legislator and courts had thought. Time and again, we find ourselves bound by EU law through European judgments beyond our understanding and expectations at the time of our commitment. Time and again, I find it rather difficult to foresee the decisions made by my honourable colleagues in Luxembourg. And I know from talks with fellow justices from various countries that this is a matter of growing concern in the supreme courts of the Member States.4

Whereas concern is expressed by the Supreme Court as regards possible violations of the Danish legal principles of rule of law, legal foreseeability and legal certainty, concern is never expressed as regards the Danish principle of proportionality. One might wonder what the reason for this is. The answer could be that the principle

2  This is not an isolated Danish concern. Case law from the United Kingdom shows the same concern of general legal principles being violated by interpretations of the European Court of Justice. See eg R (HS2 Action Alliance Ltd) v The Secretary of State for Transport [2014] EWHC 2759 (Admin) and on appeal [2014] EWCA Civ 1578. 3  U 2014 914 H, C-441/14 Dansk Industri (DI), acting on behalf of Ajos A/S v Estate of Karsten Eigil Rasmussen in which the Supreme Court decided to ask the European Court of Justice for a preliminary ruling and Case 15/2014 6 December 2016. On the mentioned development, see also J Kristiansen, ‘Grænser for EU-konform fortolkning i lyset af nyere højesteretspraksis’ (2014) 96 Juristen 17; J Kristiansen, ‘Når dansk ret (måske) er uforenelig med EU-retten: Er der en fast linje i højesteretspraksis?’ (2015) 97 Juristen 225; J Kristiansen, ‘Ajos-dommen og Højesterets dilemma’ (2016) U.2016B.301; T Baumbach and H Krunke, ‘The Role of the Danish Constitution in European and Transnational Governance’ in A Albi and S Bardutzky (eds), The Role and Future of National Constitutions in E ­ uropean and Global Governance (The Hague, Asser Press, 2017); H Krunke, ‘Courts as Protectors of the P ­ eople: Constitutional Identity, Popular Legitimacy and Human Rights’ in M Aksenova, H Krunke and M Scheinin (eds), Judges as Guardians of Constitutionalism and Human Rights (Cheltenham, Edward Elgar Publishing, 2016). 4  B Dahl, ‘Keynote Address by Mr Børge Dahl’ in U Neergaard and C Jacqueson (eds), Proceedings: Speeches from the XXVI FIDE Congress, Congress Publications, vol 4 (Copenhagen, Djøf Forlag, 2014) 26, 28.

Proportionality in Denmark in the Light of European Legal Integration  245 of proportionality has a stronger position in EU law than in the traditional Danish legal system, especially as regards constitutional rank, and that it offers the individual a stronger protection.

III.  The EU Principle of Proportionality: Differences, Tensions and Impact on the Danish Legal System The principle of proportionality is not codified in the Danish Constitution. In general, Denmark does not have a strong tradition of codifying general legal principles in its Constitution.5 According to legal literature, the constitutional provisions must be interpreted in the light of the principle of proportionality when possible.6 This is possible in relation to the following human rights: freedom of religion (Article 67), the right to property (Article 73), freedom of expression (Article 77), maybe the freedom of association (Article 78), and freedom of assembly and the prohibition of discrimination based on belief or descent (Article 70).7 The right of political parties to run for elections and the right to vote (Article 31) are other examples.8 Although it has been discussed in the legal literature whether the Danish principle of proportionality has developed into a general constitutional principle, this literature has never provided a clear conclusion that this should be the case. This is underlined by the fact that the Ministry of Justice, which reviews all Danish bills for, among others, constitutionality, does not perform a review of whether bills fulfil a general constitutional principle of proportionality.9 On the other hand, the Ministry of Justice reviews whether Danish bills are in conformity with Denmark’s international obligations, including EU law and the ECHR, and as part of this review it might be relevant to consider whether bills fulfil the EU principle of proportionality and the ECHR principle of proportionality. This means that whereas the principle of proportionality is an EU constitutional principle, the traditional Danish principle of proportionality is not a general legal principle of constitutional rank.10 Hence, whereas the EU legislator must respect the principle of proportionality in general, the Danish legislator is only bound by the principle as regards specific constitutional provisions, primarily on the protection of human rights and especially in the field of the political freedoms. In Danish law the principle of proportionality is primarily relevant in relation to the 5 

Baumbach and Krunke (n 3) bullet 2.1.1. Jensen, ’Proportionalitetsprincippet i forfatningsretlig belysning’ (1994) Ugeskrift for ­Retsvaesen 335, U 1994B 335. 7  JE Rytter, Individets grundlæggende rettigheder, 2nd edn (Copenhagen, Karnov Group, 2016) 90. 8  Jensen (n 6). 9  Justitsministeriet, ’Vejledning om lovkvalitet’ (København, 2005). 10  Jensen (n 6) and JT Pedersen, Proportionalitetsprincippet (PhD thesis, Copenhagen, Djøf Forlag, 2000). 6 MH

246 

Helle Krunke

executive power in the field of administrative law and criminal law (from which it originates). There exists an unwritten principle of proportionality in administrative law. This principle prescribes necessity as regards choosing the least intrusive legal measure within the pursued purpose and proportionality as regards whether the measure is proportionate with the pursued purpose.11 The question of constitutional rank is not the only difference between the EU principle of proportionality and the Danish principle of proportionality. The content of the EU principle also differs from the content of the Danish principle,12 especially as regards the criterion on whether the measure imposes a burden on the individual that is excessive in relation to the objective sought to be achieved (proportionality stricto sensu). This criterion leaves it to the courts to balance relevant considerations and allow a discretion that in Denmark would normally be up to the administration—or if it was a constitutional law case, the legislator— to perform.13 This is quite revolutionary, seen in light of the traditional Danish approach to the relationship between the courts and the political institutions, in which courts show great restraint. In conclusion, the EU principle of proportionality in general seems to offer individuals a stronger protection than the Danish principle of proportionality. What is the impact of this? It can create an asymmetric legal space in the sense that in cases which are covered by EU law the protection of individuals is stronger than in cases which are not covered by EU law. In other words, a situation of reverse discrimination might occur. Reverse discrimination occurs when national law in a Member State provides national citizens with a less favourable treatment than the treatment provided when a comparable situation falls within EU law. As a result of this, private parties might in some cases have an interest in claiming that their case is covered by EU law simply to gain a stronger protection. Such behaviour is not a new phenomenon. It can, for instance, be observed in the field of reverse discrimination caused by EU law in the field of citizenship. Reverse discrimination is a problem for European integration because free movement rights might be abused by citizens of Member States with stricter national legislation (a behaviour which the Council has criticised).14 Reverse discrimination might also create a feeling of unfairness among the citizens of the Member States, which can lead to dissatisfaction with EU integration (or national politicians). Finally, there seems to exist a social dimension of reverse discrimination, since primarily the elite in

11 

eg Jensen (n 6). For a thorough analysis, see N Fenger, ‘EU-rettens proportionalitetsprincip’ (2014) Ugeskrift for Retsvaesen 227, U 2014B 227. 13 ibid. 14 Council of the European Union, ‘Press Release, 2962nd Council meeting, Justice and Home Affairs’, 13467/09 (Presse 271) (Brussels, 21 September 2009); P v Elsuwege and D Kochenov, ‘On the Limits of Judicial Intervention: EU Citizenship and Family Reunification Rights’ (2011) 13 European Journal of Migration and Law 443. 12 

Proportionality in Denmark in the Light of European Legal Integration  247 society will be able to trigger the protection vested in EU law by establishing a cross-border element.15

IV.  The Principle of Proportionality in the ECHR: Differences, Tensions and Impact on the Danish legal system As mentioned, the principle of proportionality is not a general constitutional principle in Denmark. However, a number of constitutional provisions reflect proportionality and should therefore, according to legal literature, be interpreted in light of the principle. This includes, among others, the provisions on the protection of the political freedoms (freedom of expression, freedom of assembly and freedom of association), the protection of the right to property and the protection of freedom of religion. According to recent constitutional literature it is natural to ask approximately the same standard of proportionality as regards violations of the political freedoms, the right to property and the right to freedom of religion as is to be found in the ECHR.16 Accordingly, it seems to follow from case law that the Supreme Court will probably use the same proportionality test as regards the Constitution as is used in relation to the ECHR, at least as regards the political freedoms—that has been the case as regards freedom of assembly17 and freedom of expression.18 However, at the same time, certain judges from the Supreme Court have emphasised that case law does not reflect the notion that the Danish Constitution should be interpreted according to the ECHR as regards the substance of rights (as claimed by some scholars).19 Furthermore, in the Iraq case20 the Supreme Court has recently emphasised the dualistic principle (which had been challenged by some scholars). Danish courts, in general, act in a restrained manner. They interpret the Constitution in a restrictive way, leaving ‘political’ choices to the political actors. This is a tradition far from the dynamic purpose-based interpretative style of the European Court of Human Rights (and the European Court of Justice).

15  H Krunke and F Schulyok, ‘National Citizenship and EU Citizenship: What Actual Competence is Left for the Member States in the Field of Citizenship?’ in T Giegerich, O Gstrein and S Zeitzmann (eds) The EU Between ‘an Ever Closer Union’ and Inalienable Policy Domains of Member States, Schriftenreihe des Arbeitskreises Europäische Integration eV 80 (Baden-Baden, Nomos, 2014) esp 137. 16  Rytter (n 7). 17  U 1999 1798 H. 18  U 2010 1859 H. 19  eg JP Christensen, ‘Internationale konventioners betydning for Højesterets grundlovsfortolkning’ (2013) Ugeskrift for Retsvaesen 15, U 2013B 15. 20  U 2010 1547 H.

248 

Helle Krunke

In order to understand the impact of the differences between the two legal systems as regards proportionality, the following features of the Danish constitutional system are important. First, the Danish Constitution stems from 1849 with a last revision in 1953. It has altogether only been revised four times since 1849.21 The provisions on human rights are quite few and quite old. Second, as mentioned, the Supreme Court interprets the Constitution in a restrictive manner. When the Supreme Court performs constitutional review, it will only set aside legislation as unconstitutional in cases of clear violations. In practice this has only happened once (in the so-called ‘Tvind case’).22 If we then turn to the impact of the differences between the two legal systems in the field of proportionality, in light of the constitutional features just mentioned, several observations can be made. In general, the ECHR will play an increasingly important role in the human rights protection in Denmark (and this way ‘outmatch’ the Constitution). The Convention might be said to reflect a modern perception of human rights protection and it covers areas that the Danish Constitution does not. Hence, in court, lawyers will often base their arguments on ECHR and not on the Constitution in human rights cases. In relation to the principle of proportionality this means that the proportionality test in the ECHR will gain a stronger platform in the human rights protection in Denmark. This way one might fear that the Constitution will slowly lose its practical impact and importance.

V.  The Future Trends of the Application of the Principle of Proportionality I believe we can identify three possible tendencies towards developing a common European understanding of proportionality. First, while EU law is expanding into new areas and the EU Charter of Fundamental Rights is now legally binding, the area for purely internal Danish cases will gradually become smaller. We have been able to witness such a development in other areas of EU law where, for instance, free movement has been an engine in broadening the scope of EU law and diminishing the area for purely internal cases. Second, at the same time old, outdated and unclear constitutions can strengthen the impact of the ECHR. As explained, the ECHR plays a greater and greater role in Danish human rights protection, slowly ‘outmatching’ the Constitution.

21  On the procedure for amending the Constitution, see H Krunke, ‘Formal and Informal ­Methods of Constitutional Change in Denmark’ in X Contiades (ed), Engineering Constitutional Change: A Comparative Perspective on Europe, Canada and the USA, Routledge Research in Constitutional Law (Abingdon and New York, Routledge, 2013). 22  U 1999 841 H.

Proportionality in Denmark in the Light of European Legal Integration  249 Third, the judicial dialogue between national and international courts could contribute to a common European understanding of proportionality. This could in principle also be the case in Denmark in the future if European international courts and other European national courts contribute to a different Danish approach to, among others, the principle of proportionality. However, the latter does not seem to be the case; at least not yet. This can be explained by the strong Danish tradition for restrained courts afraid of overstepping legal borders and entering into what might be considered a ‘political space’. This is a tradition which builds upon a specific view of democracy and court legitimacy. Namely that whereas politicians have a democratic mandate, courts do not. This might of course change if the Danish courts can gain new kinds of legitimacy, eg based on public trust in the courts. In an article in Global Constitutionalism, Niels Petersen showed how the German Federal Constitutional Court started to apply the principle of proportionality in a more active and direct way as the Court gained a stronger platform, built on, among other things, public trust.23 The Danish courts are clearly not even close to such a stage (and the Danish historic context of course differs from the German). However, it is interesting that the former president of the Danish Supreme Court, Børge Dahl, has actually referred to the public trust in the Danish Supreme Court at several occasions and also in relation to upholding legal principles, such as the rule of law: ‘In the European Union Denmark is number one on the list of the Member States when looking at the confidence of the population in the courts and the judicial system.’24 And further, ‘It is crucial that the population has confidence in the courts as the guardians of the rule of law as fortunately is still the case in Denmark.’25 Similar points were raised in a speech by the former President of the Danish Supreme Court at the celebration of 350 years of the Danish Supreme Court in 2011: Det er vigtigt, at Højesteret har et omdømme, som understøtter den nødvendige tillid… Vi kan heldigvis glæde os over, at vi lever i et land, hvor domstolene nyder befolkningens tillid. Ifølge undersøgelser er Danmark blandt de lande i verden, hvor befolkningens tillid til domstolene er størst. Det kan vi kun være taknemmelige for. Og det er en tillid, vi må gøre os al umage for at bevare. Det er en nødvendig ballast—ikke blot i de få højtprofilerede grundlovssager, men også i de mangfoldige sager fra dagliglivet om, hvad ret og retfærdighed byder i dagens Danmark.26 It is important that the Supreme Court has a reputation which supports the necessary confidence… Luckily, we live in a country in which the courts have the confidence of 23  N Petersen, ‘Balancing and Judicial Self-Empowerment: A Case Study on the Rise of Balancing in the Jurisprudence of the German Federal Constitutional Court’ (2015) 4 Global Constitutionalism 49. 24  Dahl (n 4), 26. 25  ibid, 30. 26 B Dahl, ‘Højesteretspræsident Børge Dahls tale ved højtideligheden på Københavns Universitet i anledning af Højesterets 350 årsdag’ (Danish Supreme Court 350th Anniver­ sary, Copenhagen, 14 February 2011) available at www.hoejesteret.dk/om/artikler/Documents/ Jubil%C3%A6umstale.B%C3%B8rge%20Dahl.pdf, 13–14.

250 

Helle Krunke

the population. According to surveys, Denmark is among the countries in the world in which the population has the highest confidence in the courts. We can only be thankful for this. And it is a confidence that we must exert ourselves to maintain. It is a necessary foundation—not only in the few high profiled constitutional cases, but also in the many cases from daily life on law and justice in a Denmark of today.27

Whereas sociological court legitimacy in the form of public confidence in the courts is a quite well established way of thinking about court legitimacy in the United States,28 this is not (yet) the case in the European context. However, in the citations above, public confidence in the Supreme Court is attached considerable weight. The rhetoric used reminds us of the rhetoric used by Aharon Barak in the Israeli Supreme Court in the 1980s and the beginning of the 1990s.29

VI. Conclusion Although it might take a while, one could in theory predict that, with time, we will end up with common European standards of proportionality, since the different legal systems, including the national legal systems, the EU legal system and the legal system of the ECHR, might eventually create such a common European proportionality principle. We have already taken steps in that direction. At the national level, reverse discrimination could, for instance, push even further in that direction. So can Constitutions with an old and not very detailed regulation of human rights. Also, judicial dialogue among national European courts, the ­European Court of Justice and the European Court of Human Rights could be mentioned. Furthermore, areas of EU law such as free movement and the Charter of Fundamental Rights must be mentioned as possible fields for extending the EU version of the principle of proportionality. Seen from a global constitutional point of view some might even argue in favour of a more global understanding of ‘constitutional’ principles—among them the principle of proportionality. However, at the same time we can observe factors pulling in the opposite direction, eg at EU level with Opinion 2/13.30 At the national level, the chapters in this section of the book show that the meaning and application of proportionality (even the EU principle of proportionality and the principle of proportionality in the ECHR) still, to a certain extent, differ from Member State to Member State in 27 

My translation. O Bassok, ‘The Supreme Court’s New Source of Legitimacy’ (2013) 16 Journal of Constitutional Law 153; O Bassok, ‘The Changing Understanding of Judicial Legitimacy’ in M Aksenova, H Krunke and M Scheinin (eds), Judges as Guardians of Constitutionalism and Human Rights ­(Cheltenham, Edward Elgar Publishing, 2016). 29  Bassok, ‘Changing Understanding’ (n 28). 30  Opinion 2/13 of the Court (Full Court) of 18 December 2014 on Accession of the European Union to the European Convention for the Protection of Human Rights and Fundamental Freedoms [2014] OJ C65/2. 28  eg

Proportionality in Denmark in the Light of European Legal Integration  251 the European Union. This way, we are reminded that there is still some way to go in order to reach a common understanding and application of proportionality. At the same time, these observations to some extent question current fundamental theories on how to understand the EU legal system. The identified differences do not entirely seem to support theories which explain the relationship between EU law and national law as a pluralistic multilevel constitutional system with one legal system composed of two complementary constitutional layers—an EU layer and a national layer—in which fundamental values at the EU level and at the national Member State level over time approach each other and will end up as common ­values/principles.31 Or perhaps we have just not reached this stage yet. Furthermore, the important assumption that EU law is applied and enforced equally and in the same way in all the EU Member States is slightly shaken by the identified differences in the application of the principle of proportionality.32

31  Although many slightly different theories exist in this field, Ingolf Pernice is considered the father of multilevel constitutionalism in the just mentioned variation. See eg I Pernice, ‘Multilevel Constitutionalism in the European Union’ (2002) 27 European Law Review 511 32  At the conference on which this book is based, Gráinne de Búrca emphasised the importance of studies of how EU Law is actually applied in the Member States.

252 

Part III

Private Autonomy and Protection of the Weaker Party

254 

15 Private Autonomy and Protection of the Weaker Party STEPHEN WEATHERILL, STEFAN VOGENAUER AND PETRA WEINGERL

I.  Introduction: The Tension(s) between Private Autonomy and Protection of the Weaker Party The chapters contained in this section of the book address private autonomy and protection of the weaker party. In suggesting a clash between freedom of the individual and intervention by a regulator they carry a rich historical resonance and, crucially, one that confronts and challenges the specious simplification that appeal to ‘freedom’ tends to convey. The intellectual shape of the debate and the principal philosophical contributions are traced with great insight by Hector MacQueen and Stephen Bogle in their chapter. One may choose to begin with a guiding notion of preference for freedom from constraint, but they demonstrate that this rapidly collides with the harsh realities of life. The weaker party, in particular, will for countless reasons enjoy no such freedom from constraint in the unregulated market or in society more generally. They show how autonomy has several distinct implications within the development of contract law and they engage with the rich literature that assesses justifications for laws that may appear to constrain freedom. In this inquiry they touch on matters addressed from a jurisprudential perspective by Dori Kimel. In his chapter he demonstrates that regulatory intervention may be misunderstood as an intrusion into autonomy, because it may in truth operate as a means to promote the autonomy of the weaker party whose economic (or other) disadvantage prevents them bargaining effectively. In this sense he argues that the promotion of personal autonomy and the protection of weaker parties are not polar opposites—in fact he shows that the very reverse is true. From these perspectives flows a readiness for the law to look hard at an apparent contractual agreement. Some contracts are unenforceable, as was already recognised in (inter alia) Stair’s Institutions of the Laws of Scotland, first published in 1681, to which MacQueen and Bogle pay close and illuminating attention in their chapter. The detail alters over time: contracts made by married women were not then, but are now, enforceable. But the basic notion that the law does not permit

256 

Stephen Weatherill, Stefan Vogenauer and Petra Weingerl

an unconfined and unquestioned space for private autonomy endures. MacQueen and Bogle show that Stair offered protection to the weaker party. A party lacking the power of reason—‘infants, idiots [and] furious persons’ as well as those affected by ‘fear’, ‘drunkenness’, ‘disease’ or ‘err[or] in the substantials of what is done’—could not conclude a valid contract. The control they mention over ‘the charging of unreasonable prices for food and drink in common hostelries’ has a ready contemporary resonance, in the light of the development of control of unfair terms in consumer contracts. MacQueen and Bogle show how the nineteenth century saw a decreased concern to intervene in order to protect the weaker party, but that this was followed by rising readiness in the twentieth century to adopt sectorspecific rules protecting tenants, employees and consumers. Kimel too elucidates from a jurisprudential perspective how understandings of the theoretical basis for (and against) shaping contract law in the light of the value of personal autonomy have mutated over time and remain contested. And one important impulse in this trend is the role of the European Union. The EU is facing up to its own quest to shape an understanding of the proper relationship between private autonomy and protection of the weaker party, and it is doing so without any clearly stated Treaty mandate. It would be fanciful to suppose that the EU’s judicial and political institutions can make full use of the rich range of sources provided by MacQueen and Bogle and by Kimel as they piece together the legal architecture of the internal market; equally it would be a shame and a loss were their activities to develop without any connection at all to Europe’s intense philosophical tradition reaching all the way back to ancient Greece. The chapters contained in this section of the book are designed to push research endeavours gently in this enriching direction.

II.  Protection of the Weaker Party as a General Principle of EU Law Norbert Reich, in his book General Principles of EU Civil Law published in 2014, made a case for treating the protection of the weaker party as a general principle of EU law.1 His assembly of legislative measures and their judicial interpretation led him to argue that: EU civil law has emerged not so much as a body of rules with an objective of enabling citizens to use their autonomy for purposes … to be determined by themselves, but rather as a body of provisions that tries to protect the weaker party and to combat discrimination.2

1  2 

N Reich, General Principles of EU Civil Law (Cambridge, Intersentia, 2014). ibid, 38.

Private Autonomy and Protection of the Weaker Party

 257

The material on which he draws to develop an approach to the protection of the weaker party as a general principle covers EU social policy, especially in the matter of employment, and EU law concerning consumer protection. He also draws on fields such as transport, energy and telecommunications, where patterns of privatisation have brought civil law, most of all contract law, to the fore as public regulation has been relaxed. It is, of course, questionable how far one should seek to extract general principles from such material at all, especially given the EU’s predominant focus on the project to complete the internal market. The urge to systematise is itself methodologically contested. Moreover even if one is prepared to operate at such a level of abstraction, whether one can really go so far as to extract this general principle from the EU’s legislative acquis is open to dispute too. But there is, for sure, a rich seam of material to be mined. Provisions of the Charter such as Article 31 (fair and just working conditions), Article 36 (access to services of general economic interest) and Article 38 (consumer protection) today frame the debate about the EU’s place in protecting the weaker party, but the Charter, which acquired binding effect only in 2009 as a result of the entry into force of the amending Treaty of Lisbon, has arrived late at the party. Secondary legislation adopted with the constitutional purpose of improving the functioning of the internal market has a lengthy track record in addressing the protection of the weaker party.

III.  The Weaker Party in the Internal Market The principal legal base used for the purpose of legislative action to protect the weaker party is Article 114 TFEU (and its functionally equivalent predecessors Article 95 EC and Article 100a EEC). Article 114 TFEU provides for the ­adoption of measures ‘for the approximation of the provisions laid down by law, regulation or administrative action in member states which have as their object the establishing and functioning of the internal market’. ‘Harmonisation’, rather than ‘approximation’, has become the more commonly used noun, but the key constitutional point is that it may not be pursued for its own sake, but rather only in order to achieve the objective of creating an internal market within the EU. This is how the limiting constitutional principle that the EU enjoys only the competence conferred on it by its Treaties, contained in Article 5 TEU, is given concrete expression in application to Article 114. It is market-making harmonisation that is at stake, not harmonisation per se. It is moreover a market-making project that is backed by basic economic understandings of the virtues of international trade, as the chapter written by Fernando Gómez and Mireia Artigot explains. The concern for the protection of the weaker party found in the EU’s legislative acquis is a classic instance of the re-regulatory function of harmonisation. Article 114 is not explicitly devoted to pursuit of the protection of the weaker party, but because diversity among national laws, inter alia those that address the

258 

Stephen Weatherill, Stefan Vogenauer and Petra Weingerl

protection of the weaker party, is treated as causing competitive distortion within the internal market the EU acquires competence to act in these areas in order to achieve a better functioning internal market. Put another way, the EU reaches the weaker party instrumentally, through the internal market—but it does reach the weaker party. Harmonisation counts as a response to legislative diversity at national level and its effect is to transfer to EU level responsibility to select an appropriate—common—regulatory standard. So, for example, Directive 2009/72 is a measure of harmonisation which establishes common rules for the internal market in electricity, and it includes provision for universal service and the protection of vulnerable consumers.3 It is the diversity among national approaches to such matters that generates an EU competence to introduce common rules, and those common rules reflect and absorb the social priorities that have provoked intervention at national level. That the EU’s legislative acquis addresses universal service and consumer vulnerability is a demonstration of how protection of the weaker party, not simply autonomy and choice in the market, has become part of the EU’s harmonised pattern of re-regulation. A cascade of such measures of harmonisation, most of them Directives, stretches across consumer law, labour market regulation, transport law, public procurement and far beyond.4 This is not the direct check on enforceability envisaged over 300 years ago by Stair: rather, it is a question of setting boundaries within which contractual autonomy shall be practised. It is felicitously described as ‘regulated autonomy’ in the EU’s internal market.5

IV.  Anti-Discrimination Law Lately a strand of EU intervention has developed in a manner that is independent of the focus on the internal market. This is the anti-discrimination acquis. Here too Reich has argued for a general principle of non-discrimination in EU civil law.6 Article 19 TFEU provides for a competence to take appropriate action to combat discrimination based on sex, racial or ethnic origin, religion or belief, disability, age or sexual orientation. This is not tied to or limited by the internal market. Amplification by secondary legislation includes most prominently Directive 2000/43 on equal treatment between persons irrespective of racial or ethnic ­origin7 and Directive 2000/78 establishing a general framework for equal

3 

[2009] OJ L211/55. See S Weatherill, Contract Law of the Internal Market (Cambridge, Intersentia, 2016) esp ch 4. 5  G Comparato and H Micklitz, ‘Regulated Autonomy between Market Freedoms and Fundamental Rights in the Case Law of the CJEU’ in U Bernitz, X Groussot and F Schulyok (eds), General Principles of EU Law and European Private Law (Alphen aan den Rijn, Wolters Kluwer, 2013) ch 4. 6  Reich (n 1) ch 3. 7  [2000] OJ L180/22. 4 

Private Autonomy and Protection of the Weaker Party

 259

t­ reatment in employment and occupation.8 The unilateral practices of an undertaking which does not hold a position of dominant economic power are beyond the supervision of the EU’s competition rules and the same is probably true of the free movement rules governing goods, persons and services.9 But such respect for autonomy to select a contracting partner (or not) is set aside where the matter falls within the protective scope of the anti-discrimination Directives. The chapter written by Alina Tryfonidou explores the contribution made by EU law to the protection of the rights of a particular set of individuals vulnerable to discriminatory treatment: lesbian, gay, bisexual and trans individuals (LGBT) and same-sex couples. She shows the Court’s initial—rather nervous and certainly institutionally cautious—approach to use of the general principles of equality and non-discrimination in this context. It is Treaty revision, most significantly achieved by the Treaty of Amsterdam (with effect from 1999) and the Treaty of Lisbon (with effect from 2009) that paved the way for a firmer basis of legal protection in this area, moreover one that is not focused exclusively on the economic dimension. The grant of binding force to the Charter of Fundamental Rights, a major element of the Lisbon reforms, is central to this. This serves to emphasise that the development of protection of the weaker party is a task shared among several actors within the EU system, and also by the Member States acting to change it from the outside. It shows too that the Court may lead, albeit sometimes hesitantly, but that it may need, and may attract, support from other sources. This should certainly not be taken to mean that a fully satisfactory pattern has been concocted. As Tryfonidou explains, gaps in legal protection against discrimination are still to be found, prominent among them the limited material scope of Directive 2000/78. But her examination of the post-Lisbon case law persuades that the Court is intent on an interpretative approach which goes some distance towards filling these gaps.

V.  Harmonisation, Integration and Re-Regulation before the Court Alina Tryfonidou’s essay points to a truth that transcends the particular context of her chapter. It is of the highest significance to grasp that the Court has been an important actor in this pattern of legal evolution. In the first place it has made very clear that Article 114 TFEU is to be interpreted as a grant of a generously

8 

[2000] OJ L303/16. rather than certainly, because on a small number of occasions the Court has fleetingly suggested that not only collective but even unilateral acts of private parties affecting movement of persons may be caught: Case C-281/98 Angonese [2000] ECR I-4139; Case C-94/07 Andrea Raccanelli [2008] ECR I-5939; Case C-379/09 Maurits Casteels v British Airways [2011] ECR I-1379. 9  ‘Probably’

260 

Stephen Weatherill, Stefan Vogenauer and Petra Weingerl

wide ­legislative competence to the EU. Article 114 TFEU may be validly relied on to adopt a measure of legislative harmonisation: where there are differences between national rules which are such as to obstruct the fundamental freedoms and thus have a direct effect on the functioning of the internal market; or where such differences cause significant or appreciable distortions of competition; or where it is likely that such obstacles will emerge in the future, because the Member States are about to take divergent measures with respect to a product or a class of products.

A body of case law has accumulated over the last twenty years in which the Court has consistently adopted these tests as its starting point when it is asked to determine whether a challenged instrument of harmonisation meets these governing criteria and is consequently valid as a matter of EU law.10 So Article 114 is in principle a limited legislative competence, as all EU legislative competences are in principle limited. In practice, however, it is broad, for it is driven by the functional demands of the completion of the internal market. It is hugely difficult for those opposed to EU action to find a constitutionally effective objection. In fact, the famous ruling in the first Tobacco Advertising case, in which the Court found that the limits of EU competence had been violated, is high profile precisely because it is so atypical.11 Most attempts to persuade the Court that a measure of harmonisation lacks a valid constitutional underpinning have failed. Moreover in its accumulated case law the Court has confirmed the re-regulatory paradigm whereby the EU acquires a responsibility not simply to adopt a common rule but also to address the quality of that common rule. That the decisive factor in the choice of re-regulatory regime is public health or consumer protection is not at all an objection that carries constitutional force. Quite the reverse. Provided the necessary contribution to the functioning of the internal market is present, it is proper that the quality of the regime be taken into account in the legislative process.12 If what is being subjected to the discipline of legislative harmonisation is national rules governing public health or consumer protection, then the consequence of

10 Case C‑376/98 Germany v Parliament and Council [2000] ECR I‑8419; Case C‑377/98 Netherlands v Parliament and Council [2001] ECR I‑7079; Case C‑491/01 British American Tobacco (Investments) and Imperial Tobacco [2002] ECR I-11453; Case C‑210/03 Swedish Match [2004] ECR I‑11893; Joined Cases C‑154/04 and C-155/04 Alliance for Natural Health and Others [2005] ECR I-6451; Case C-217/04 United Kingdom v Parliament and Council [2006] ECR I‑3771; Case C‑380/03 Germany v Parliament and Council [2006] ECR I‑11573; Case C‑301/06 Ireland v Parliament and Council [2009] ECR I-593; Case C-58/08 Vodafone Ltd and Others v Secretary of State for Business, Enterprise and Regulatory Reform [2010] ECR I-4999; Case C-270/12 UK v Council and Parliament judgment of 22 January 2014; Case C-398/13 P Inuit judgment of 3 September 2015. See Weatherill (n 4) esp ch 3. 11  Case C-376/98 (n 10). 12  eg Case C-376/98 (n 10) para 88; Case C-380/03 (n 10) para 38; Joined Cases C‑154/04 and C-155/04 (n 10) para 30.

Private Autonomy and Protection of the Weaker Party

 261

­ armonisation is EU rules governing public health or consumer protection— h albeit public health or consumer protection in the internal market. This model explains how the EU reaches the protection of the weaker party—in the internal market. This ‘dual function’ of harmonisation, as both a means to integrate and to re-regulate the internal market, is commanded by the Treaty, especially Articles 12, 114(3), 168(1) and 169(1) TFEU, and it is anchored too in the Charter of Fundamental Rights, especially Articles 35, 37 and 38. These provisions dictate a constitutional linkage between protection and the definition and implementation of all other EU common policies, which evidently includes linkage to the project to complete the internal market through legislative harmonisation. In this fashion re-regulatory values are embedded in internal market law.

VI.  The Court’s Interpretation of the Legislative Acquis on the Protection of the Weaker Party The Court also interprets the harmonised rules that have been adopted. In doing so it pursues a principled approach which reveals its own understanding of the protective ethos motivating EU rule-making. Here too it draws on Articles of the Treaty and the Charter, including those directing pursuit of a high level of worker and consumer protection. Article 12 TFEU and Article 38 of the Charter commit the EU to a high level of consumer protection, and Article 151 TFEU and Title IV of the Charter perform a similar function in the matter of protection of the worker. There is no better example of the combination of legislative initiative and judicial interpretation in the protection of the weaker party than that which is provided by Directive 93/13 on unfair terms in consumer contracts.13 Directive 93/13 asserts in its Preamble that disparities between national laws of this type cause distortions of competition in the internal market. Harmonising rules prohibiting unfair terms will, it is claimed, generate intensified cross-border competition and a better functioning internal market. The re-regulatory dynamic is clear: the EU adopts a common rule in the service of a more tightly integrated market, but the content of that common rule is dictated by the concern to guard against use of unfair terms in consumer contracts. National law is thereby adjusted by the EU’s harmonisation programme and more generally the EU becomes a factor in the pattern of law-making provoked by scepticism about the virtue of permitting unrestricted contractual autonomy in relationships that are e­ conomically imbalanced. Article 6 of the Directive directs that unfair terms used in a contract concluded with a consumer by a seller or supplier shall not be binding on the consumer. 13 

[1993] OJ L95/29.

262 

Stephen Weatherill, Stefan Vogenauer and Petra Weingerl

Ordinary courts within the Member States are expected to decide whether a term is unfair, in so far as the dispute in question falls within the scope of Directive 93/13. The Court has devised a formula to explain the purpose of the Directive on which it now relies routinely. It holds that Article 6 is a mandatory provision which, taking into account the weaker position of one of the parties to the contract, aims to replace the formal balance which the contract establishes between the rights and obligations of the parties with an effective balance which re-establishes equality between them.14 This, then, is treated as a direct check on the enforceability of contractual terms which is motivated by concern to tackle the imbalance that flows from permitting contractual freedom free rein in the economically asymmetric relationship that normally, albeit not always, prevails between the trader and the consumer. Judged on these terms the Directive is not a threat to freedom of contract, but rather aims to promote the autonomy of weaker parties. The emphasis on addressing the weaker party in the light of economic imbalance chimes with the chapter written by Dori Kimel. He shows how there is nothing neutral about non-intervention in contractual freedom when the consequence of that choice is to reinforce the power of strong parties at the expense of the weak. The Court’s case law on the interpretation of Directive 93/13 is precisely in this vein. Fernando Gómez and Mireia Artigot too show how asymmetries between contracting parties may taint the contracting process and invite legal intervention: they explain that the presence of weaker parties is just one such motivating asymmetry. They are concerned too to counsel that private law is not the best instrument to address inequalities which operate at a macro level, for example, in the structure of the market. There is abundant case law in which the Court has been invited by national courts to interpret Directive 93/13 on unfair terms.15 That it is sensitive to the concern to protect the weaker party is visible in rulings in which the Court has chosen to promote the effective application of the Directive before national courts by treating procedural matters that are beyond the Directive’s explicit terms as matters that nonetheless fall within the scope of EU law. In Elisa María Mostaza Claro v Centro Móvil Milenium SL (‘Mostaza Claro’) the Court was asked by a referring Spanish court to deal with an arbitration agreement which was contaminated by inclusion of an unfair contractual term.16 The problem was that under a­ pplicable

14 eg Case C-168/05 Mostaza Claro v Centro Móvil Milenium [2006] ECR I-10421, para 36; Case C‑137/08 VB Pénzügyi Lízing [2010] ECR I-10847, para 47; Case C-618/10 Banco Español de Crédito v Joaquín Calderón Camino judgment of 14 June 2012, paras 40, 63; Case C-415/11 Aziz j­udgment of 14 March 2013, para 45; Case C-470/12 Pohotvost Pohotovosť s r o v Miroslav Vašuta judgment of 27 February 2014, paras 39–41; Case C-169/14 Sánchez Morcillo judgment of 17 July 2014, para 23; Case C-8/14 BBVA SA judgment of 29 October 2015, para 17. 15 For examination in depth, see H Micklitz and N Reich, ‘The Court and Sleeping Beauty: The Revival of the Unfair Contract Terms Directive’ (2014) 51 Common Market Law Review 771; O Gerstenberg, ‘Constitutional Reasoning in Private Law: The Role of the CJEU in Adjudicating Unfair Terms in Consumer Contracts’ (2015) 21 European Law Journal 599; Weatherill (n 4) ch 4.4.3. 16  Case C-168/05 [2006] ECR I-10421.

Private Autonomy and Protection of the Weaker Party

 263

Spanish procedural law the consumer had failed to plead that the arbitration agreement was invalid during the arbitration proceedings and it was now too late for her to raise the matter in a subsequent action for annulment. Spanish law was clear: the arbitration award against her had to stand, even though that decision disrespected Directive 93/13’s prohibition of unfair terms. The Court accepted that ‘it is in the interest of efficient arbitration proceedings that review of arbitration awards should be limited in scope and that annulment of or refusal to recognise an award should be possible only in exceptional circumstances’,17 but it found that these circumstances were exceptional. Spanish law had to yield, for otherwise the ‘regime of special protection established by the Directive would be definitively undermined’.18 The Court emphasised the Directive’s purpose as a means to protect the consumer who is in a relatively weak position.19 That this is the leitmotif is hammered home by appreciation that, absent the presence of a consumer, the Court is much readier to uphold the autonomy of national procedural law even where its application may frustrate the proper integration of EU law’s requirements into national litigation. In Van der Weerd the applicants unsuccessfully challenged an order concerning the slaughter of their animals.20 The Dutch court noted that in the proceedings the applicants had failed to advance significant and potentially decisive arguments based on EU law, which under Dutch procedural law could not now be taken into account. The competent court was obliged to give its ruling solely on the basis of the issues which were put before it. Even though this might prejudice the proper application of EU law, the Court of Justice had no quarrel with the Dutch rule. Its purpose was based on the assumption that in a civil suit it is for the parties to take the initiative, and that as a result the rights of the defence are safeguarded and the proper and expeditious conduct of proceedings is ensured.21 The Court explicitly noted that Mostaza Claro, decided a year earlier, was ‘justified by the need to ensure that consumers are given the effective protection’ which Directive 93/13 seeks to achieve.22 The applicants in Van der Weerd had by contrast missed their chance to raise a plea based on EU law: they were not a weaker party in need of rescue. Article 38 of the Charter asserts the high significance of consumer protection in EU law, and the Court has lately taken to citing it in case law dealing with the interpretation of Directive 93/13,23 but this is confirmation, not transformation, of long-standing judicial practice. The Court in this way uses its interpretative monopoly to endow EU consumer law with a distinct character and, via its application by national courts, accentuates the shape

17 

ibid, para 34. ibid, para 31. 19  ibid, paras 25, 36. 20  Joined Cases C-222/05 et al van der Weerd [2007] ECR I-4233. 21  ibid, paras 33, 35–36. 22  ibid, paras 39–40. 23  eg Case C-470/12 Pohotvost Pohotovosť s r o v Miroslav Vašuta judgment of 27 February 2014, para 52. 18 

264 

Stephen Weatherill, Stefan Vogenauer and Petra Weingerl

of consumer law as a distinct protective category within the legal orders of the Member States. Directive 93/13 has been a site for vigorous judicial amplification of legislative intent, but it is by no means alone in generating such energy. Directive 85/577 harmonises rules on the protection of the consumer in respect of contracts negotiated away from business premises.24 Commonly known as the ‘Doorstep Selling’ Directive, it assumes, as its Preamble states, a consumer who is ‘unprepared’ and vulnerable to a ‘surprise element’ when a trader visits his or her home. This, in short, is concern to protect the weaker party whose autonomy is fragile in the environment that falls to be regulated. In E Friz the Court was asked to consider whether the Directive precluded a particular German rule governing fair apportionment of losses.25 It did not, and in reaching this conclusion the Court noted that its purpose was ‘to ensure, in accordance with the general principles of civil law, a satisfactory balance and a fair division of the risks among the various interested parties’.26 Daniela Caruso’s chapter explores the importance of this ruling and in particular she emphasises the significance of the Court’s readiness to elevate ‘fairness’ to the rank of a general principle. In particular, as she notes, the Court did this without any explicit provocation by the referring court or by the Advocate General in the case. Caruso explains that the adoption of fairness as a principle in this way is to signal receptivity to distributive justice as a proper concern for contract law. As she shows, there is a vigorously expressed literature addressing the question whether this should be so. E Friz is the seed of a debate of this type that is particular to and emerging within the EU. Caruso makes a strong case in her chapter for the virtue of such a fairness principle as part of a wider agenda of promoting equality, and she praises the Court for its ambition. This cannot—and should not—be done by the Court alone. But the Court can—and does—impart momentum.

VII.  General Principles: Judicial Creativity Unleashed, Judicial Creativity Confined The Court is sometimes bolder still. Its Mangold judgment is the most vivid demonstration of its readiness to discover a general principle of law which is not to be found explicitly in the primary or secondary texts.27 Drawing on ‘various

24 

[1985] OJ L372/31. Directive 2011/83 [2011] OJ L304/64 now governs the field. Case C-215/06 [2010] ECR I-2957. 26  ibid, para 48. 27  Case C-144/04 [2005] ECR I-9981. 25 

Private Autonomy and Protection of the Weaker Party

 265

i­nternational instruments and … the constitutional traditions common to the Member States’,28 it identified a general principle of EU law which forbids discrimination on grounds of age. Moreover the Court added that such a general principle enjoys application not only in cases involving public authorities but also, as in Mangold itself, in cases involving private parties, in casu parties to a ­contract of employment. By siting this general principle at the level of (in effect) primary law the Court subverted the long-established restriction on unimplemented Directives’ capacity to operate ‘horizontally’. More broadly it demonstrated a surprisingly high level of disregard for private autonomy. It is of course not at all unknown in Europe that discrimination on defined grounds shall be forbidden in the private sphere: nor is it unknown that judges shall take the lead in initiating such patterns of development.29 Mangold is at one level simply the EU’s own episode in this evolving narrative. But this cannot quell the sense of disquiet about just why the Court chose to make this bold step, and how many more such steps may await in future. There are, however, indications that Mangold is and will remain an outlier. In Audiolux the Court held with restraint that to find a general principle of law demanded that it possess a ‘general, comprehensive character’.30 In Sonia Chacon Navas the Court refused to accept that non-discrimination on grounds of sickness counts as a general principle of EU law.31 In Fag og Arbejde FOA, acting on behalf of Karsten Kaltoft v The Municipality of Billund the Court rejected the contention that EU law contains any general principle forbidding discrimination based on obesity.32 The chapter written by Niilo Jääskinen recounts the story of this litigation, featuring the unfortunate Danish childminder Karsten Kaltoft represented by his union. The author could not be better informed: he was the Advocate General in the case itself. The Court concluded that EU law does not include a general principle of non-discrimination on grounds of obesity as such, though it did find that obesity may in certain circumstances fall within the notion of a ‘disability’, which for matters arising in the workplace would be within the scope of the prohibition foreseen by Directive 2000/78 establishing a general framework for equal treatment in employment and occupation. The relevant criteria for finding such unlawful treatment are carefully explained by Jääskinen.

28 

ibid, para 74. eg C Mak, Fundamental Rights in European Contract Law. A Comparison of the Impact of ­Fundamental Rights on Contractual Relationships in Germany, the Netherlands, Italy and England (Alphen aan den Rijn, Kluwer Law International, 2008); D Oliver and J Fedtke, Human Rights and the Private Sphere: A Comparative Study (London, Routledge, 2007); C Busch and H Schulte-Nölke (eds), Fundamental Rights and Private Law (Munich, Sellier, 2010). 30  Case C-101/08 Audiolux [2009] ECR I-9823, para 42. 31  Case C-13/05 Sonia Chacon Navas [2006] ECR I-6467. 32  Case C-354/13 judgment of 18 December 2014. 29  See

266 

Stephen Weatherill, Stefan Vogenauer and Petra Weingerl

VIII.  General Principles: Ring-Fenced by the Charter? The Court evidently has its own developing sense of restraint, but the Charter of Fundamental Rights might provide a more secure basis for ring-fencing the verdant growth of general principles. The Charter was granted binding force only in 2009 and consequently there is no mention of it in Mangold. However, in Seda Kücükdeveci v Swedex GmbH & Co KG, another employment dispute involving private parties in which age-based discrimination had occurred, the Court, citing Mangold and confirming that non-discrimination on grounds of age counts as a general principle of EU law, was able to rely also on the prohibition of discrimination based on age laid down by Article 21 of the Charter.33 It plainly remains to be seen how the Court will interpret the Charter and in particular how far it will go in finding its provisions capable of horizontal effect:34 this will require elucidation through the case law, much as the Court in the wake of Van Gend en Loos35 was invited over time to explain the vitality of particular Treaty provisions in proceedings before national courts. The Court is already not short of advice, and it is wildly conflicting in its visible normative preference for, on the one hand, the grant to private parties of autonomy from control in the light of the Charter and, on the other, its readiness to impose a degree of constitutionalised restraint on the exercise of private economic power.36 It is, however, plausible that the Court’s thirst for creative identification of new and unexpected general principles, of which Mangold stands as the loftiest landmark, has been slaked by the rise of the Charter. There are plausible institutional and constitutional reasons for shying away from the approach taken by the Court in Mangold, and they are visible in the subsequent more restrained rulings mentioned in the previous sub-section. In Audiolux the Court had been prodded by Advocate General Trstenjak, who in her Opinion warned that the Court ‘may not assume the role of the Community legislature if a gap in the law can be filled by the Community legislature’,37 and referred

33 

Case C-555/07 [2010] ECR I-365. cf in this vein Case C‑176/12 Association de médiation sociale judgment of 15 January 2014. 35  Case 26/62 [1963] ECR 1. 36  See for example, from a deep and deepening mine: O Cherednychenko and N Reich, ‘The Constitutionalization of European Private Law: Gateways, Constraints, and Challenges’ (2015) 23 European Review of Private Law 797; M De Mol, ‘Horizontal Direct Effect of Union Prohibitions of Discrimination’ in A Hartkamp et al (eds), The Influence of EU Law on National Private Law (Deventer, Kluwer, 2014) 219–50; M Fornasier, ‘The Impact of EU Fundamental Rights on Private Relationships: Direct or Indirect Effect?’ (2015) 23 European Review of Private Law 29; M Dougan, ‘The Impact of the General Principles of Union Law upon Private Relationships’ in D Leczykiewicz and S Weatherill (eds), The Images of the Consumer in EU Law (Oxford, IECL/Hart Publishing, 2016) ch 5; C Herresthal, ‘Grundrechtecharta und Privatrecht’ [2014] Zeitschrift für Europäisches Privatrecht 238. 37  Case C-101/08 Audiolux (n 30) para 107, citing ‘Louis, J-V, L’ordre juridique communautaire, 6th edition, Brussels/Luxembourg 1993, p 119, 120’. 34 

Private Autonomy and Protection of the Weaker Party

 267

to the superior democratic legitimation and expertise which the political institutions claim in matters that entail choice between different political and social interests. Niilo Jääskinen’s chapter examining the Kaltoft case takes care to situate the ruling in the context of sensitivity to the proper limits of EU competence. Both Article 6(1) TEU and the Charter itself preclude its use to extend the competences of the EU. This is not an all-embracing fundamental rights regime: it stretches as far as the EU Treaties stretch but no further. In short, general principles are now probably to be found within the explicit terms of the Charter, not beyond it. The Charter of Fundamental Rights, endowed with full legitimacy as a result of its location within the process of ratification of the Lisbon Treaty and moreover itself imprinted with the direction to respect the limits of EU competence, seems today to be the Court’s principal source of both inspiration and caution.

IX.  Conclusion: Selecting Priorities This, to conclude, is not to suggest that the Charter offers a clear and enduring settlement of the tension between private autonomy and the protection of the weaker party. Rather the reverse. What matters is precisely how and why judicial and legislative choices about competing priorities are made. The EU’s Charter of Fundamental Rights reflects the debate about the virtue and vice of contractual autonomy that has dominated thinking for (as MacQueen and Bogle remind us) centuries, transplants it to EU level, but chooses not to resolve it in favour of any particular set of values. Article 16 of the Charter states that: ‘The freedom to conduct a business in accordance with Union law and national laws and practices is recognised’; and Article 17, the right to property, declares in its first sentence that: ‘Everyone has the right to own, use, dispose of and bequeath his or her lawfully acquired possessions.’ This is, in short, a perspective that is favourable to contractual autonomy: it suggests a constitutionalised protection of private economic interests from public regulation. But Articles 16 and 17 coexist with other Charter rights and principles that are receptive to intervention in the public interest, inter alia in order to achieve protection of the weaker party. Provisions such as Article 31 on fair and just working conditions, Article 35 on public health, Article 37 on environmental protection and Article 38 on consumer protection are supportive of the virtue of limiting autonomy in the market in order to achieve broader goals in the collective interest. They provide a basis for asserting a constitutionalised respect for values that generate the adoption of restrictions on contractual autonomy. Moreover Article 17 of the Charter itself concedes that: ‘The use of property may be regulated by law in so far as is necessary for the general interest’, while, broader still, Article 52(1) admits of limitation on the exercise of the rights and freedoms recognised by this Charter as ‘provided for by law’, subject to the proviso that such limitation shall respect the essence of those rights and freedoms and shall be made only where necessary.

268 

Stephen Weatherill, Stefan Vogenauer and Petra Weingerl

All this is ‘Charter-speak’ for an approach which the Court pioneered long before the grant of binding force to the Charter in 2009. So, for example, the Court long ago recognised the right to property as a general principle of EU law, but chose to treat it not as an absolute right which defeats intervention in the market in the public interest, but instead as a right which is not absolute and which must be viewed in relation to its social function. The Court has never objected to legislative restriction on the exercise of the right to property ‘provided that those restrictions in fact correspond to objectives of general interest … and do not constitute a disproportionate and intolerable interference, impairing the very substance of the rights guaranteed’.38 This is the tangled web within which the tension between autonomy and protection of the weaker party falls to be developed. It is the gloriously evasive ‘social market economy’ to which the EU is committed by Article 3(3) TEU. The responsibility for the selection of priorities attaches to all the EU’s institutions, judicial and political, and, as the dynamic story of the growth of general principles told in this section and in this book more generally reveals, those institutions do not operate in isolation from each other.

38  eg Case C-491/01 R v Secretary of State ex parte BAT and Imperial Tobacco [2002] ECR I-11543; Case C-200/96 Metronome Musik GmbH [1998] ECR I-1953.

16 Private Autonomy and the Protection of the Weaker Party: Historical HECTOR L MACQUEEN AND STEPHEN BOGLE

I.  Intellectual History or a History of Ideas? Telling a history of autonomy and the weaker party raises problems of scope and methodology. Where to start? How to tell it? What kind of history? Two broad possible approaches (not necessarily mutually exclusive) are apparent in the exploration of ideas from an historical perspective.1 The first is that of intellectual history.2 This involves an exploration of the context and paradigms under which a term, such as autonomy, was used by historical actors or when it was used within a discourse between historical actors.3 The different questions asked by writers are explored with sensitivity to both context and the paradigms giving rise to an idea. Intellectual history does not, however, start from the presupposition that an idea such as autonomy can be discovered by someone, like a jewel in a mine. Nor does it assume that the idea in question has been perpetually considered or ignored by people until a particular moment in history. Rather the intellectual historian concentrates on when the term emerged in historical discourse and plots the particular linguistic, ideological and social contexts giving rise to its use.4 In contrast, a second approach is through the history of ideas.5 This provides an analysis of ideas from a logical, descriptive or normative perspective. The arguments of different writers are compared with close attention to their premises and conclusions then drawn. It demonstrates how, at different times or in different places, different or similar ideas have been offered to answer either recurring or similar questions. 1 

R Whatmore, What Is Intellectual History? (Chichester, Wiley, 2016). J Tully (ed), Meaning and Context: Quentin Skinner and his Critics (Princeton, Princeton ­University Press, 1988). 3  Q Skinner, Liberty before Liberalism (Cambridge, Cambridge University Press, 1998) 101–19. 4  Whatmore (n 1) 45–58. 5  An older form can be found in the classic work of AO Lovejoy: The Great Chain of Being: The Study of the History of an Idea [1934] (Cambridge, MA Harvard University Press, 1960). A more modern example is the work of K Haakonssen, Natural Law and Moral Philosophy: From Grotius to the Scottish Enlightenment (Cambridge, Cambridge University Press, 1996). 2 

270 

Hector L MacQueen and Stephen Bogle

After falling out of favour during the 1970s and 1980s,6 the history of ideas has re-emerged in a revised form in recent years, and continues to be a fruitful method of enquiry.7 Today, more than ever, the study of ideas from a historical perspective is sensitive to context, ideology and language; but it does not necessarily need to follow the approach adopted by the intellectual historian. Nor for that matter does an intellectual history need to avoid completely a higher-level analysis of the ideas used by a writer under scrutiny. Nevertheless, any history of an idea needs always to be aware of the dangers of projecting into the past ideas of the present, while being equally on guard for an explanation or understanding of an idea which is intricately linked to a specific context in a specific period. For present purposes, however, if something general is to be said about autonomy and the weaker party from the perspective of legal history, then it may be more profitable to speak of the idea of autonomy from the perspective of a history of ideas. Indeed, such an approach can help break down modern conceptions of how law has developed and ensure that present-day legal thought is informed and enriched by the ideas of the past.

II.  Negative and Positive Freedom (or Autonomy) Philosophers have differed strongly about autonomy over the last 250 years, distinguishing between personal autonomy (concerning simply an individual’s capacity to decide upon and follow a course of action and perhaps meaning that children, incapable adults and animals lack autonomy); moral autonomy (the concept usually attributed to Immanuel Kant (1724–1804) by which the individual reflects on a course of action as a moral choice which can be universalised as also a rule for others8); and political autonomy (under which the individual’s decisions about courses of action are taken into account within political communities). Each may be opposed by heteronomy (another term developed by Kant), in which a person’s actions are determined by the rule or rules of another.9

6  For a linguistic and methodological critique see Q Skinner, ‘Meaning and Understanding in the History of Ideas’ (1969) 8 History and Theory 3, whereas for a more politically potent criticism see M Foucault, The Archaeology of Knowledge and the Discourse of Language (New York, Pantheon Books 1972). 7  Haakonssen (n 5) 1–14; M Bevir, The Logic of a History of Ideas (Cambridge, Cambridge University Press, 1999); and DM McMahon and S Myon (eds), Rethinking Modern European Intellectual History (Oxford, Oxford University Press, 2014). 8 See J B Schneewind, The Invention of Autonomy: A History of Modern Moral Philosophy (Cambridge, Cambridge University Press, 1997). 9  See for overviews eg T Honderich (ed), The Oxford Companion to Philosophy (Oxford, Oxford University Press, 1995) svv ‘autonomy and heteronymy’, ‘autonomy in applied ethics’; Stanford Encyclopaedia of Philosophy, svv ‘Personal Autonomy’ (http://plato.stanford.edu/entries/personal-autonomy/); ‘Autonomy in Moral and Political Philosophy’ (http://plato.stanford.edu/entries/personal-autonomy/); Internet Encyclopaedia of Philosophy, sv ‘Autonomy’ (http://www.iep.utm.edu/autonomy/).

Private Autonomy and the Protection of the Weaker Party: Historical  271 We think it useful to open our discussion with Isaiah Berlin’s (1909–1997) famous essay on two concepts of liberty (or freedom, a term which he used ­interchangeably with liberty, and which for present purposes can be treated as synonymous with a conception of autonomy).10 Berlin’s investigation was within the context of political philosophy but nevertheless sheds a great deal of light upon how lawyers may use terms such as autonomy. Aware that freedom was a ‘protean word’ with more than ‘two hundred senses of it recorded by historians of ideas’,11 Berlin nonetheless identified both a negative and a positive sense. These two notions were ‘central’, with a ‘great deal of human history behind them and [he suggested] still to come’.12 To adopt a negative conception of freedom against a positive conception (autonomy) can also imply a particular conception of law’s authority and purpose; and vice versa. Negative freedom is most closely associated with classical English political philosophy and the work of philosophers such as Thomas Hobbes (1588–1679).13 But it can also be found to a lesser extent in the great works of Hugo Grotius (1583–1645)14 and Samuel von Pufendorf (1632–1694).15 Here freedom means freedom from constraint.16 It could feasibly be used to describe a state of nature where people were, in theory at least, completely unhindered in what they decided to do. It may be argued that this is an untenably high conception of freedom— in reality life will always limit the individual’s options. Whether through physical incapacity or lack of knowledge or understanding, people will always be constrained in some manner. But Berlin draws a distinction: ‘[Y]ou lack political liberty or freedom only if you are prevented from attaining a goal by human beings.’17 Although initially helpful in differentiating natural constraints upon people and their options, this distinction does not help settle disagreements about

10  I Berlin, ‘Two Concepts of Liberty’ in I Berlin, Four Essays on Liberty (Oxford, Oxford ­University Press, 2002) 167–218. Note that in his Liberty before Liberalism Quentin Skinner has argued that Berlin’s two conceptions misses out a third possibility, which he calls a ‘neo-roman’ sense of liberty. It is therefore worth clarifying that we understand Berlin’s two conceptions of freedom or liberty as analytically useful only to the extent of demonstrating the sense in which both can be used to justify coercion but we do not take this distinction to be an exclusive explanation of all the senses in which the terms liberty or freedom can be used. Indeed, below we consider Neil MacCormick’s conception of autonomy. See text accompanying nn 147–151. 11  Berlin (n 10) 169. 12  ibid, 207. 13  T Hobbes, Leviathan, The Matter, Form and Power of a Commonwealth Ecclesiastical and Civil [1651] (Cambridge, Cambridge University Press, 1996) ch 21. For discussion see R Tuck, Natural Rights Theories: Their Origin and Development (Cambridge, Cambridge University Press, 1979). 14 GL Williams and WH Zeydel (trans), H Grotius, De Iure Praedae Commentarius [1603] ­(Indianopolis, Liberty Fund, 2006) I, 18; H Grotius, The Rights of War and Peace from the Edition by Jean Barbeyrac [1625, 1738] (Indianopolis, Liberty Fund 2005), The Preliminary Discourse; R Tuck, ‘Introduction’ in H Grotius, The Rights of War and Peace (Indianopolis, Liberty Fund 2005). 15  S von Pufendorf, The Whole Duty of Man According to the Law of Nature [1673] (Indianopolis, Liberty Fund 2003) I, 6, 3. 16  Hobbes (n 13) 21.1–3. 17  Berlin (n 10) 170.

272 

Hector L MacQueen and Stephen Bogle

how far it can be said that the actions of one person are not restricting the freedom of another in either an immediate or even indirect sense. For example, it could be said that the structure of capitalism as practised in the Western world puts many in the developing world into poverty or, to use Berlin’s phrase, ‘economic slavery’.18 Or it could be said that our lack of knowledge or understanding is a consequence of economic policies pursued by government, and therefore it can be said that our freedom has been limited. No easy answers can be given to such questions. But that does not necessarily make the initial distinction un-useful. It merely means that one may disagree as to the ultimate cause of our poverty, lack of knowledge or understanding, with the consequence of disagreeing as to whether our lack of opportunity is a result of human interference. But of course, if one does accept the general sense of negative freedom as suggested by Berlin, then it is immediately obvious that this sort of freedom could lead to chaos and an exploitation of the weak. The weaker party is one who is not physically strong or wealthy or shrewd enough to protect himself or herself against the force, wealth or cunning of a stronger party. In the seventeenth c­ entury, this was of fundamental concern to political and philosophical discourse in Europe. Hobbes, for example, suggested that man through fear gives up his natural ­freedom to have security and protection from exploitation.19 Pufendorf came near to a similar conclusion, but ultimately supplemented Hobbes’ conclusion with the notion that humanity is also sociable.20 Grotius was more inclined to suggest that it was humanity’s reason and sociable nature that lead to the creation of society and the avoidance of chaos.21 Later Enlightenment thinkers such as Adam Smith (1723– 1790) and John Stuart Mill (1806–1873) proposed that human interests could be harmonised to the extent that there would not be conflict even in a H ­ obbesian state of nature.22 From the perspective of political philosophy, therefore, one might find disagreement as to the possibility of harmonising human interests and also about how much freedom one should be given and in what instances interference can be justified. Nonetheless, under this sort of definition of negative freedom, law is an instrument of control and coordination, and it is backed up by the threat of force. It may have a good reason for being implemented—social order and control—but inevitably, law will represent an instance of freedom being hindered or restrained by human beings. On this basis, laws can be explained and their constraints justified by the ultimate goal of peace and civil order; but freedom is nevertheless being constrained. 18 

ibid, 170. Hobbes (n 13) ch 17. 20  Pufendorf (n 15) II, 3, 5–13. 21 Grotius, The Rights of War, The Preliminary Discourse (n 14). 22  A Smith, The Theory of Moral Sentiments, 6th edn [1790], ed RL Meek, DD Raphael and P Stein (Oxford, Oxford University Press, 1976, reproduced by Liberty Fund, 1982) 1.1.1 (‘Of Sympathy’); JS Mill, On Liberty [1859] ch 3 in ML Morgan (ed), Classics of Moral and Political Philosophy, 3rd edn (Indianopolis, Hackett, 2001). 19 

Private Autonomy and the Protection of the Weaker Party: Historical  273 Positive freedom or autonomy, however, is ‘the wish on the part of the ­individual to be his own master’.23 It describes the situation whereby an individual has made a decision to act without any interference from an external force. Such a person will say: ‘I wish to be a subject, not an object; to be moved by reasons, by conscious purposes, which are my own, not by causes which affect me, as it were, from outside.’24 Quickly, however, the concern arises as to whether a person is actually a slave of the passions, to use the phrase of David Hume (1711–1776), or of the person’s nature and desires.25 A weaker party is therefore someone unable to control these desires and passions, prone to mistakes and errors or, indeed, akrasia (weakness of the will). Concerns of this type set freedom-seeking persons against themselves and undermine the idea that they are free. As Berlin notes, the dominant self, which is set apart from the passions and desires, is then ‘variously identified with reason, with my “higher nature”, with the self which calculates and aims at what will satisfy it in the long run, with my “real”, or “ideal”, or “autonomous” self.’26 At least two options emerge in such a situation. One can accept the inability to be truly free from nature, desire and passions, and therefore become stoic or aesthetic in one’s approach to life. Alternatively, one can try to restrain, control and resist one’s desires and the drive of nature. Individuals are free in this sense if they transcend their passions and desires in an attempt to react to reason alone— people who are rational and able to think for themselves about what course of action they wish to pursue on the basis of reasons offered to them, or to act in accordance with plans and self-imposed laws and rules of action. Such a notion of positive freedom can be easily equated to the original, Kantian, sense in which one might use the term autonomy, ie as self-governance or self-legislating. For a lawyer, this conception of freedom can lead to a far more extensive justification and explanation of law’s authority. It offers a far deeper and more philosophically elaborate explanation of law’s purpose and its role in restraining or compelling or enabling a person to action or inaction. Berlin notes that, although at first negative and positive notions of freedom may not seem at odds, they have nonetheless developed in ‘historically divergent directions’ and have come into ‘direct conflict with each other’.27 The conflict arises in that a positive conception of freedom can soon be used and exploited in such a manner that people are told that in order to gain ‘self-mastery’ they must follow a prescribed path—often something imposed on them by an authority. Prescription soon becomes laced with the rhetoric which says, on the one hand, you must be entirely free, but on the other hand, if you do not do as we request you will be determined un-free or incapable of acting freely. Accordingly, you are weak, in

23 

Berlin (n 10) 179. ibid, 179. 25 D Hume, A Treatise of Human Nature [1739–40], ed DF and MJ Norton (Oxford, Oxford ­University Press, 2001) II, 3, 3 (‘Of the influencing motives of the will’). 26  Berlin (n 10) 206. 27  ibid, 179–80. 24 

274 

Hector L MacQueen and Stephen Bogle

error and need to be compelled to realise your ‘autonomy’. Those who adopt a negative conception of freedom against an autonomy sense will therefore debate the areas and boundaries of a person’s freedom. Disagreement will arise amongst such proponents about when people should be left to do their own thing, and when they should be coerced or restrained. Those who adopt an autonomy-based conception of freedom will disagree as to who one should obey. Whose authority is legitimate over the self? Whose conception of autonomy should one adopt?

III.  Autonomy and Contract Law For a lawyer the concept of autonomy has a strong association with contract law.28 For example, the editors of the Draft Common Frame of Reference on E ­ uropean private law published in 2009 see the role of party autonomy as relatively ­limited in property law and non-contractual obligations, while in contract law, ‘[t]he assumption is that party autonomy should be respected unless there is a good cause to intervene’.29 Again, Sarah Worthington’s recent discussion of party autonomy in private law is almost entirely taken up with contract law: ‘[T]ort and unjust enrichment … exert considerable constraints on autonomy, but arguably none that are more than necessary for the good of society’, while ‘in the realm of contract law party autonomy reigns supreme’.30 Autonomy, as a concept, can be used in several ways when associated with ­contract law. For example, it can be used as part of an historical explanation of contract law: as a causal factor in its development generally or in particular rules.31 Alternatively, it can be used as part of a descriptive theory of contract law seeking to posit what the law of contract is.32 Hence, Thomas Gutmann argues that contracts ‘are tools for realizing individual self-determination by means of voluntarily entering legally binding agreements’.33 Charles Fried says that contract is

28  Of course, it does indeed also have a place within, for example, succession law (freedom of ­testation) and family law. 29  C von Bar, E Clive and H Schulte-Nölke (eds), Principles, Definitions and Model Rules of ­European Private Law: Draft Common Frame of Reference (Munich, Sellier 2009), Principles, paras 2–15 ­(quotation at para 2). 30 S Worthington, ‘Common Law Values: The Role of Party Autonomy in Private Law’ in A Robertson and M Tilbury (eds), The Common Law of Obligations: Divergence and Unity (Oxford, Hart Publishing, 2016) 301, 322. This article influenced the UK Supreme Court to restrict the scope of the rules against penalty clauses: see Makdessi v Cavendish Square Holdings BV; ParkingEye Ltd v Beavis [2015] UKSC 67, [2016] AC 1172. 31  J Gordley, The Philosophical Origins of Modern Contract Doctrine (Oxford, Oxford University Press, 1991) ch 8. 32  A descriptive will theory does not explicitly engage in ‘theory’ or expressly advocate one p ­ articular idea. It is implicit, however, in the organisations of texts; the definitions offered and the solutions ­proposed for resolving tensions or difficulties. 33  T Gutmann, ‘Some Preliminary Remarks on a Liberal Theory of Contract’ (2013) 76 Law and Contemporary Problems 39, 40.

Private Autonomy and the Protection of the Weaker Party: Historical  275 based on a morality of autonomy, respect for persons and trust. Promise is a kind of moral invention: its mode of causation is moral causation; it allows persons to create an obligation where there was none before, and thus it gives free individuals a facility for extending their reach by enlisting the reliable collaboration of other free persons.34

Fried and Gutmann exemplify how contract theory sometimes treads the line between negative and positive conceptions of freedom. Both recognise that contract law allows parties to realise their autonomy while simultaneously saying that this realisation is only possible for parties given the freedom to do so. On this basis, law is understood to provide a space within which one can exercise autonomy but no assumption is necessarily made about how that autonomy should be used.35 At the same time there is a close relationship between contract law and the realisation of autonomy.36 Often such a conception of contract is called a will theory of contract. The meaning and nature of such a will theory of contract is contested. Sometimes autonomy and the will theory are taken to represent a legal manifestation of nineteenth-century free-market economics and utilitarian philosophy.37 This therefore places autonomy and the will theory as a representation of the negative conception of liberty. At other times will theory is a juridical implementation of political or moral ideas (as per Fried or Gutmann, for example). In taking this approach, autonomy is used as a means of explaining why the will’s declaration of intention to contract should be respected by the law (and how other rules of contract should be framed), rather than on the basis of any alternative social or economic ground. Today, however, it is not uncommon to find a pluralist conception of contract which uses the concept of autonomy to describe, possibly, the creation of a contract, but not necessarily the content or the source of the obligation as being also based on autonomy.38 Such theories may have a strong sense of what should and

34 C Fried, Contract as Promise: A Theory of Contractual Obligation, 2nd edn (Oxford, Oxford ­University Press, 2015) 137. 35 However, such theories differ in how they justify the enforcement of a broken contract: J Raz, ‘Promises in Morality and Law’ (1982) 95 Harvard LR 916; Fried (n 34) ch 8; S Smith, Contract Theory (Oxford, Oxford University Press, 2004). 36  See LL Fuller, ‘Consideration and Form’ (1941) 41 Columbia LR 799; also D Kennedy, ‘From the Will Theory to the Principle of Autonomy: Lon Fuller’s “Consideration and Form”’ (2000) 100 ­Columbia LR 94. 37 eg W Friedmann, Law in a Changing Society (New York, Columbia University Press, 1959) 90–125; LM Friedman, Contract Law in America: A Social and Economic Case Study (Madison, ­University of Wisconsin Press 1965); MJ Horwitz, ‘The Historical Foundations of Modern Contract’ (1974) 87 Harvard LR 917–56; PS Atiyah, The Rise and Fall of Freedom of Contract (Oxford, Oxford University Press, 1979) 294. cf AWB Simpson, ‘The Horwitz Thesis and the History of Contracts’ (1979) 46 University of Chicago LR 533. 38  M Hogg, ‘Competing Theories of Contract: An Emerging Consensus?’ in LA DiMatteo, Q Zhou, S Saintier and K Rowley (eds), Commercial Contract Law: Transatlantic Perspectives (Cambridge, ­Cambridge University Press, 2013); B Bix, ‘Theories of Contract Law and Enforcing Promissory ­Morality: Comments on Charles Fried’ (2012) 45 Suffolk LR 719; R Kreitner, ‘On the New Pluralism in Contract Theory’ (2012) 45 Suffolk LR 915; H Dagan, ‘Autonomy, Pluralism, and Contract Law Theory’ (2013) 76 Law and Contemporary Problems 19.

276 

Hector L MacQueen and Stephen Bogle

should not be enforced within a contract, but they do not rest their justification for that upon any thick sense of autonomy or what a truly free contracting partner would have done. Autonomy in these sorts of theories is used as a justification as to why the will of a contracting party should be respected and how the will of a party may create an obligation.

IV.  Will Theory in the Natural Law Tradition Some histories of the Common law of contract acknowledge that the will theory seemingly so linked with the economic and political theories of the nineteenth century had much earlier antecedents. For example, David Ibbetson says: ‘There was nothing new in this; it was an idea that had been familiar to Common lawyers for at least half a millennium. What was new was the greater depth given to the idea, and the greater weight placed on it.’39 Several Anglo-American legal h ­ istorians, including John Baker,40 Philip Hamburger41 and Brian Simpson,42 allude to an earlier history of will-based ideas in contract law. There is, however, a strong case, very well explored in more recent times by such as the late Robert Feenstra, Gero Dolezalek, James Gordley and Wim Decock,43 for locating the emergence and juridification of will-based thinking in contract law in the work of Jesuit moral theologians writing in Spain between about 1500 and 1650 (the late, or Spanish, scholastics). Further, an especially important conduit for the transmission of their work to the rest of western Europe was the group of jurists whom Gordley has dubbed the ‘northern natural lawyers’. Grotius is pre-eminent in the group, which also includes Pufendorf and the slightly later figures of Jean Domat (1625–1696),

39  D Ibbetson, A Historical Introduction to the Law of Obligations (Oxford, Oxford University Press, 2001) 233. 40  JH Baker, ‘The Rise and Fall of Freedom of Contract’ (1980) 43 Modern Law Review 467. 41  P Hamburger, ‘The Development of the Nineteenth-Century Consensus Theory of Contract’ (1989) 7 Law and History Review 241. 42 AWB Simpson, ‘Innovation in Nineteenth Century Contract Law’ (1975) 91 Law Quarterly Review 247. 43  See eg (1) R Feenstra, ‘L’influence de la Scolastique espagnole sur Grotius en droit privé, Quelques experiences dans des questions de fond et de forme, concernant notamment les doctrines de l’erreur et de l’enrichissement sans cause’ in P Grossi (ed), La seconda scolastica nella formazione del diritto private modern (Milan, Giuffrè, 1973) 377; (2) G Dolezalek, ‘The Moral Theologians’ Doctrine of Restitution and its Juridification in the Sixteenth and Seventeenth Centuries’ [1992] Acta Juridica 104; (3) the ­following writings by J Gordley: ‘Natural Law Origins of the Common Law of Contract’ in J Barton (ed), Towards a General Law of Contract (Berlin, Duncker & Humblot, 1990); Philosophical Origins (n 31) 71–77; ‘Some Perennial Problems’ in idem (ed), The Enforceability of Promises in European Contract Law (Cambridge, Cambridge University Press, 2002) 2–10; Foundations of Private Law: Property, Tort, Contract, Unjust Enrichment (Oxford, Oxford University Press, 2006) chs 13, 16, 17; The Jurists: A Critical History (Oxford, Oxford University Press, 2013) ch V; and (4) W Decock, Theologians and Contract Law: The Moral Transformation of the Ius Commune (ca 1500–1650) (Leiden, Brill, 2013).

Private Autonomy and the Protection of the Weaker Party: Historical  277 Christian Thomasius (1655–1728) and Christian Wolff (1679–1754). Klaus-Peter Nanz has also highlighted the significance of Mattheus Wesenbecius (1531–1586), a Dutch Protestant humanist jurist who taught at Wittenberg, in the development of consent-based notions in contract law.44 The key to the understandings of the late scholastics was the medieval revival of Greek philosophy, and specifically Aristotle, and its folding into the bosom of Christian theology by Thomas Aquinas, in the thirteenth century. While both Aristotle and Aquinas were certainly seriously interested by law, neither was a ­lawyer, nor was either trying to write legal treatises. Their objective was rather to consider how a person might lead a virtuous life and serve the common good, and, in the case of Aquinas, thereby know and serve God. For both, what distinguished humanity from the rest of creation was the capacity to make reasoned decisions about how to act shaped, not only by consideration of the actor’s own interests, but also by a perception of how the action would impact upon others. Law itself was the product of such reasoning, but it remained open to ongoing analysis by human reason, so that gaps might emerge between the natural law and the positive law in force in any given place or time.45 The Spanish scholastics and, following them, the northern natural lawyers synthesised Aristotle and Aquinas in seeing promises in general as binding in nature on the persons who made them, reflecting the virtues of fidelity (keeping one’s word), liberality (the sensible giving away of resources to chosen others) and commutative justice (the equivalence of exchanges so that no one was enriched at the expense of another). This enabled a breakout from the Roman system of real, verbal, written and consensual contracts towards a general law of contract, and significant revision of the Roman position against the enforceability of so-called ‘naked pactions’. It also emphasised the moral basis for the exercise of the freedom of an individual’s will. For Gordley, however, this raises the question whether will theory as it developed in the nineteenth century, with its emphasis on individual freedom over more general considerations of justice, was founded on philosophical ideas of morality that had actually been largely discarded long before 1800.46 44 K-P Nanz, Enstehung des allgemeinen Vertragsbegriffs im 16 bis 18 Jahrhundert (Munich,­ J Schweitzer Verlag, 1985) 85–94. The book is reviewed helpfully for non-German speakers by WM Gordon (1987) 8 Journal of Legal History 373–6. Wesenbecius taught the Scottish jurist John Skene at Wittenberg c 1570: see JW Cairns, TD Fergus and HL MacQueen, ‘Legal Humanism and the History of Scots Law: John Skene and Thomas Craig’ in J MacQueen (ed), Humanism in Renaissance Scotland (Edinburgh, Edinburgh University Press, 1990), 52. H Berman, Law and Revolution, II: the Impact of the Protestant Reformation on the Western Legal Tradition (Cambridge, MA, Harvard University Press, 2003), 156–66, also briefly notes Wesenbecius’ contribution in arguing that the systematic work of Lutheran jurists anticipated the Spanish late scholastics. 45  The key texts here are Aristotle, Nichomachean Ethics, in J Barnes (ed), The Complete Works of ­Aristotle: The Revised Oxford Translation (Princeton, Princeton University Press, 1984) vol 2, 1729–867); Aquinas, Summa Theologiae, 1.75–102; 1/2.90–114, in AC Pegis, Basic Writings of Saint Thomas Aquinas (New York, Random House, 1944). 46 Gordley, Philosophical Origins (n 31) 9. In his Foundations of Private Law (n 43) 4–5, Gordley argues that the philosophical ideas should be reinstated: ‘[P]rinciples that commend themselves to our own common sense, that were once accepted almost universally long ago, that were discarded for the wrong reasons, still best explain private law.’

278 

Hector L MacQueen and Stephen Bogle

Martin Hogg has suggested that different will theories can be said broadly to represent different stages in the historical development of contract law.47 It is instructive for present purposes therefore to explore whether during the earlier stages in the development of contract law there existed a conception of autonomy, liberty or freedom which could be equated in any way with those two senses identified by Berlin. In particular, it is worth revisiting the early modern period when ideas of liberty, property and contract were being realigned with political and moral ideas.

V.  Autonomy and Contract Law in the Early Modern Period The seventeenth century was remarkably fertile for the generation of ideas about the state, authority, freedom and rights. Indeed, in recent years the study of this period has brought forth a wealth of new literature about the roots of modern political thought. Haakonssen and Seidler have shown that natural law theories of the seventeenth century are well worthy of study from the point of view of both philosophical and legal thought.48 Early modern debates were wide and varied, but from the point of view of legal history and particularly private law this was a key turning point.49 During this period, jurists and philosophers across Europe engaged in renewed effort to understand the roles of monarchs, religion and law, following the sixteenth-century Reformation, sparked by Luther’s crushing polemic on humanity, church and temporal authority.50 The period saw a significant restatement by Protestant theologians of humanity’s moral weakness.51 It also saw, however, a revitalisation of a more positive conception of humanity’s abilities through the Catholic Counter-Reformation.52 These movements had a profound 47  M Hogg, Promises and Contract Law: Comparative Perspectives (Cambridge, Cambridge U ­ niversity Press, 2011) 166–74. 48 K Haakonssen and MJ Seidler, ‘Natural Law: Law, Rights and Duties’ in R Whatmore and B Young, A Companion to Intellectual History (Chichester, Wiley, 2015). 49 Q Skinner, The Foundations of Modern Political Thought [1978] (Cambridge, Cambridge ­University Press, 1997); Tuck (n 13); B Tierney, The Idea of Natural Rights (Saarbrücken, Scholars Press, 1997); Haakonssen and Seidler (n 48). 50  F Wieacker (trans T Weir), A History of Private Law in Europe (Oxford, Oxford University Press, 1995); Berman, Law and Revolution, II (n 44); Gordley, Philosophical Origins (n 43); R Lesaffer (trans J Arriens), European Legal History: A Cultural and Political Perspective (Cambridge, Cambridge ­University Press, 2009); Decock (n 43). Schneewind (n 8) 4–9, describes the effect as a shift from morality as obedience to God to morality as self-governance, albeit that this emerged as a self-­conscious process only in the early eighteenth century. 51  F Wendel, Calvin, 8th edn (London, Collins, 1978) 185–284; A McGrath, The Intellectual Origins of the European Reformation, 2nd edn (Chichester, Wiley, 2003); D MacCulloch, A History of Christianity (London, Penguin, 2010) 716–55. 52  Skinner (n 49), vol 2, 135–73; Gordley, Philosophical Origins (n 31) 69–111; and Decock (n 43) 21–104.

Private Autonomy and the Protection of the Weaker Party: Historical  279 effect on both the political and legal apparatuses of Europe,53 leading Grotius, Hobbes and Pufendorf to seek secular explanations of society and its ­origins.54 This in turn had fundamental implications for how the law was formulated, and on how humanity’s abilities to act justly or unjustly were conceptualised. In very general terms, it might be said that while on one hand, Grotius, Hobbes and Pufendorf were inclined towards a negative conception of liberty and freedom,55 others, such as the late scholastics of the Counter-Reformation,56 or indeed the Calvinists, held a positive conception of freedom in the sense that they saw law as enabling people to realise their God-given freedom.57 In this context it is specifically worth giving attention to the work of the Scot, James Dalrymple Viscount Stair (1619–1695), whose work offers a mid-point between the late scholastics and the early modern natural rights theorists such as Grotius and Pufendorf, while also demonstrating the impact of Calvinism.

VI.  Northern Natural Law Will Theories In his Institutions of the Law of Scotland, written around 1659–1661, and first ­published in 1681, Stair emerges as a Protestant natural lawyer.58 In this regard, therefore, he is comparable to, inter alia, Pufendorf and Grotius. But Stair ­differs from his Protestant counterparts in important respects—differences particularly evident in Stair’s treatment of conventional obligations (although further ­examples can be found elsewhere in the Institutions). Stair asserted, in contrast to Pufendorf and Grotius, that a promise was enforceable without acceptance;59 he stated that, in making promises, people were, first and foremost, bound to God to perform their obligations60 (ie the social utility of performance was a by-product of a person’s performance, but not the reason why a person should perform61); he argued that no matter what equity required, positive law was unable to provide strict equality within bargains;62 and he made clear, here like Grotius and Pufendorf, that people had liberty to bind themselves by way of contracts and promises.63 But he then contended that people should use

53 

See above n 50. Haakonssen (n 5) 26–46; D Saunders, ‘Within the Orbit of this Life—Samuel Pufendorf and the Autonomy of Law’ (2002) 23 Cardozo Law Review 2178; Tuck (n 13) 156–62. 55  Haakonssen (n 5) 26–43. 56  Decock (n 43). 57  See further below, text accompanying nn 73–76. 58  James Dalrymple, Viscount Stair, The Institutions of the Law of Scotland [1681, 1693] (Edinburgh, Edinburgh University Press, 1981). 59  ibid, 1.10.4. 60  ibid, 1.10.1. 61  ibid, 1.1.18. 62  ibid, 1.10.14. 63  ibid, 1.10.1. 54 

280 

Hector L MacQueen and Stephen Bogle

that liberty to bring glory to God.64 Not all of these points can be explored in detail here. But we must explain why Stair chose a path generally distinct from Grotius and Pufendorf, resembling rather the late scholastics. It is also possible to detect both a positive and negative conception of liberty within Stair’s work.

VII.  Stair and the Late Scholastics At first blush, Stair’s Protestantism put him at odds with Catholics such as the late scholastics, particularly when it comes to the ability of humanity to establish civil society and maintain an institution such as contract law. However, there are similarities between them. First, the scholastics, in contrast to Grotius and Pufendorf, maintained a theological conception of law’s foundation. This had implications for their approach to notions of liberty and the purpose of contract law in general. Second, parallels emerge between Stair and the late scholastics through their use of similar sources, namely scholastic commentaries on Aquinas or Aristotle. ­Calvinism provides the vital link on both points. It has long been known that Stair, whose career began as a regent in philosophy at Glasgow University before he switched to legal practice, was influenced by Aristotle.65 Only more recently has the influence of Aquinas upon him also been identified and elaborated.66 Stair’s knowledge of Aristotle and Aquinas, and of the Spanish scholastics, may have been mediated through Grotius, whose influence upon him was also strong, but by no means total.67 While Grotius and ­Pufendorf began to reject to a greater or lesser extent an Aristotelian world-view, and particularly the scholastic methods of explicating Aristotle,68 Stair, like the late scholastics, continued to work within the Aristotelian-scholastic tradition. In fact, in the seventeenth century Protestants throughout Europe went on drawing from

64 

ibid, 1.1.20. See eg AH Campbell, The Structure of Stair’s Institutions (Jackson, Glasgow, 1954) 8–9. 66  D Reid, ‘Thomas Aquinas and Viscount Stair: The Influence of Scholastic Moral Theology on Stair’s Account of Restitution and Recompense’ (2008) 29 Journal of Legal History 189. 67  See JJ Gow, ‘The Introduction of the Theory of Justice in Scots Law’ (unpublished PhD thesis, University of Aberdeen, 1952); PG Stein, ‘Stair’s General Concepts: The Theory of Law’ in DM Walker (ed), Stair Tercentenary Studies (Stair Society vol 33, 1981); WM Gordon, ‘Stair, Grotius, and the Sources of Stair’s Institutions’ in JA Ankum, JE Spruit and FBJ Wubbe (eds), Satura Roberto Feenstra (Éditions Universitaires, 1985) (also in Gordon, Roman Law, Scots Law and Legal History: Selected Essays (Edinburgh, Edinburgh University Press, 2007)); DL Carey Miller, ‘Systems of Property: Grotius and Stair’ in idem and DW Meyers (eds), Comparative and Historical Essays in Scots Law: A Tribute to Professor Sir Thomas Smith QC (London, Butterworths, 1992); T Richter, ‘Molina, Grotius, Stair and the jus quaesitum tertio’ [2001] Juridical Review 219; idem, ‘Did Stair Know Pufendorf?’ (2003) 7 Edinburgh LR 367; ALM Wilson, ‘Stair and the Inleydinge of Grotius’, (2010) 14 Edinburgh LR 259. 68  Gordley argues that Grotius is more Aristotelian than Pufendorf; he further argues that neither really abandoned Aristotle (Gordley, Philosophical Origins (n 31) ch 5). But there is no doubt that both Grotius and Pufendorf disliked scholasticism: Grotius, Rights of War, The Preliminary Discourse (n 14) 11.3; Pufendorf (n 15) V, 9. 65 

Private Autonomy and the Protection of the Weaker Party: Historical  281 Aristotle, including his works on physics, ontology and logic as well as his moral philosophy. Against this background, one can explain some of the terminology and concepts shared by Stair with the late scholastics. For example, Stair’s description of human action in many ways reflects the ­philosophical theology bridging the seventeenth-century religious divide in the form of voluminous commentaries upon Aristotle.69 He spoke of resolution, deliberation and the will in much the same way as Luis Molina (1535–1600), Leonard Lessius (1554–1623) or Pedro Onate (d 1646).70 Such language was commonplace within the parasitic commentaries on Aristotle; and was, of course, most evident in the most extensive treatment of Aristotle in the seventeenth century (still), Aquinas’ Summa Theologica.71 Stair is far from alone in offering such a description of human action. Many other seventeenth-century Scots deployed a similar description of human action in their theological treatises. It re-emerges in Stair’s own theological treatise, the Divine Perfections, published posthumously in 1695.72 And, importantly, it is a description of human action found in the ­Institutes of John Calvin (1509–1564).73 The reason why Protestants such as Stair continued to use Aristotle was markedly different from the Spanish scholastics, however. Calvinists used Aristotle’s moral philosophy as a means of teaching students about virtue, and to demonstrate what humanity was capable of in a pre-lapsarian state. But Calvinists were in no way advocating that lapsarian people could be virtuous; at least, not without the aid of God. As Calvin himself remarked in his Institutes: ‘[I]n reading profane authors, the admirable light of truth displayed in them should remind us, that the human mind, however much fallen and perverted from its original integrity, is still adorned and invested with admirable gifts from its Creator.’74 Thus, ­Aristotle, and other profane philosophers, helped Calvinists study moral excellence and understand Adam’s moral virtue before the fall. But, in stark contrast to the late scholastics, Scots such as Stair were not learning about moral excellence and virtue as the means to discern the path to salvation or justification. That was only achievable through the grace of God. If students of Aristotle were ever in doubt, Calvin made matters clear in his Institutes. He corrects the philosophers ‘who set it down as beyond dispute, that virtue and vice are in our own power’.75 In chapter II of book II of the Institutes, he is adamant that: ‘[M]an cannot arrogate any thing, however

69 

Stair (n 58) I, 10, 1–2. Decock (n 43) 162–86. See further below, text accompanying nn 120–24. 71  J Kraye, ‘Moral Philosophy’ in CB Schmitt et al (eds), The Cambridge History of Renaissance ­Philosophy (Cambridge, Cambridge University Press, 1988). 72  J Dalrymple, Viscount Stair, A Vindication of the Divine Perfections illustrating the Glory of God in them, by Reason and Revelation: Methodically Digested into Several Meditations. By a Person of Honour (London, 1695). 73 J Calvin (trans H Beveridge), Institutes of Christian Religion [1559] (Grand Rapids, MA, ­Eerdmans, 1989) I, 15,18. 74  ibid, II. 2. 15. 75  ibid, II, 2, 3. 70 

282 

Hector L MacQueen and Stephen Bogle

minute, to himself, without robbing God of his honour, and through rash confidence subjecting himself to a fall’.76 This, it would appear, had a profound effect on how Stair formulated the law of Scotland with regard to promising. He saw promise-keeping as virtuous, as did the late scholastics, Grotius and Pufendorf. But in contrast to Grotius and the Jesuit theologians, he could not give the glory for promise-keeping—for undertaking and performing promises as a virtuous act—to a person alone. For the Calvinist, a person only attained virtue through God’s grace. In this light, Stair’s statement that a person gives up liberty through a promise or contract, ‘whereby God obliges us to performance, by mediation of our own will’ is wholly in keeping with the orthodoxy of Calvinist theology and the reason why a promise did not need acceptance. The obligation was owed, first and foremost, to God. As with the late scholastics, Stair thus maintained a theistic explanation of ­liberty, law and contract law. He lived through the turbulent 1640s, when theological discourse animated politics in Scotland and indeed all Britain. He also taught theology at the University of Glasgow at this time. In 1649 he was appointed to a law commission by the radical Presbyterian Scottish Parliament, which, being ­‘zealously desirous for the glory of God’, ‘the good of his people’ and ‘the administration of justice’ in Scotland, ‘to have as far as possible may be by the blessing of God a perfect rule for administration of justice in all causes before all judicatories’, gave the commission full power to consider the laws of Scotland.77 Nothing came directly of this, but the question of godly rule is answered by Stair in the opening book of his Institutions, written just over a decade later. His treatise on theology, the Divine Perfections, offers further insight into his legal thought. These writings manifest an overall concern to establish the relationship between positive law, including contract law, and godly rule. Further, Stair could not merely follow Grotius or indeed Pufendorf in how he analysed the law, because they took theological and political stances opposed to his. Pufendorf rang the curtain down upon the theological stage.78 Any attempt to justify political authority or positive law should no longer make an appeal to theology. Grotius had earlier suggested that natural law was discoverable by reason, as well as through God; but Pufendorf focused on what was discoverable by reason and observation alone, not theology. In his preface to The Whole Duty of Man, According to the Law of Nature, Pufendorf made a clear division between three modes of enquiry into the duty of man which by the law of nature man must follow. People ‘derive Knowledge of their Duty’ from ‘three Springs, or Fountain-Heads; to wit, From the Light of Nature; From the Laws and Constitutions of Countries; And from the special Revelation of Almighty God’.79 Under

76 

ibid, II, 2, 3. KM Brown et al (eds), The Records of the Parliaments of Scotland to 1707 (St Andrews, 2007–16, www.rps.ac.uk) 1649/1/306. 78  Saunders (n 54). 79  Pufendorf (n 15) Preface. 77 

Private Autonomy and the Protection of the Weaker Party: Historical  283 the ­influence of Lutheran theology, Pufendorf thought that ‘by our Reason we are not able to comprehend’ moral divinity.80 Humanity’s duty to God is revealed only in Holy Scripture. Pufendorf is far from abandoning natural law at this stage, however. As Haakonssen notes, Lutheran theology normally resulted in a ‘suspicion and neglect of natural law’, but, inspired by ‘Cartesianism and by Hobbes’, ­Pufendorf ‘saw the possibility of developing [natural law] … as a complete “science of morals”’.81 This was an explanation of natural law which differed from that of both Grotius and Stair. Pufendorf ’s innovation should not detract from another important point: he subscribed, as Lutherans did, to the absolute sovereignty of kings. In Scotland, however, during the mid-seventeenth century, theories of absolute sovereignty were unappealing for many Calvinists, including Stair. As is well known, in his Institutes Calvin says that in matters of civil government, there are two extremes. On the one hand, ‘the flatterers of princes, extolling their power without measure, hesitate not to oppose it to the government of God’.82 But on the other hand, there was the extreme of civil disobedience.83 In Scotland, during the mid-seventeenth century, the question of civil disobedience arose acutely, and particularly the extent of the sovereign’s authority. In that context, Scots Calvinists, such as Stair, were extremely cautious about endorsing theories of absolute sovereignty because of the tensions which existed between the Church of Scotland and King Charles I. However due to their theological position they could not develop a natural law theory akin to Grotius and P ­ ufendorf, as such a theory downplayed the role of God within the institution of society and civil order. As JD Ford has argued, Stair’s discussion of liberty and servitude within the Institutions suggests he was influenced by the Scots theologian, Samuel R ­ utherford (1600–1661).84 Rutherford had what to modern minds seems a rather convoluted explanation of civil government: God instituted kings but citizens needed to endorse their God-given king in order for him to have legitimacy. Rutherford was therefore somewhere between Locke and theories of divine right. Additional evidence of Rutherford’s influence on Stair can be found in within the Divine ­Perfections, where he says: God exerciseth his Dominion not only by the Laws of Nature, but by positive Laws and Institutions, which are not known by the Light of Nature, but by Revelation; such is the Government of the Husband over the Wife, by the Divine Institution of Marriage, and the Government of Common Wealths, by God’s Institution of a publick Judgment, to order and determine all Controversies, concerning such as should unite themselves in Civil Societies. … The Power of the Husband, and the Indurance and Dissolution of

80 

Haakonssen (n 5) 37–38. ibid, 37–38. 82  Calvin (n 73) I, 20, 1. 83 ibid. 84 JD Ford, ‘Stair’s Title “Of Liberty and Servitude”’ in ADE Lewis and DJ Ibbetson (eds), The Roman Law Tradition (Cambridge, Cambridge University Press, 2007). 81 

284 

Hector L MacQueen and Stephen Bogle

the ­Marriage are instituted by God, and unalterable by Man; but in other things their ­Pactions are effectual, as in other Cases: So also in Civil Government, the Form of Government, the Persons governing, the Extensions of their Power, more than what is requsite for Order and Determination of Controversies, are by the express and tacit Pactions of the Parties, except what God did immediately determine in the Government of Israel.85

Stair did not view government or positive law as developing from a transfer of rights from an individual to a sovereign. Unlike Pufendorf or Grotius, he could not merely adopt a theory of absolute sovereignty, establish natural law without an appeal to God, and thereafter offer a natural law account of law based on reason or Hobbesian explanations of the state of nature. Stair’s account of law, including contract law, had to strike a balance between Calvinist theology on the one hand and the political realities of Scotland on the other.86 This meant that he needed a natural law to be self-evident and to transcend the powers of a monarch. This, as has been noted before, means that Stair’s account of law and contracting retains, unlike Grotius and Pufendorf, a need for God, theological dogma, and the idea that people can understand something of God’s plan for order in the temporal world, including how persons contract with one another.

VIII.  Seventeenth-Century (Calvinist) Will Theory Stair thus emerges as a Calvinist natural lawyer. The positive law of Scotland, including the law relating to contract, flowed from equity, but nonetheless fell short of achieving anything like the natural law which God intended humanity to follow. But regardless, there is a need for law, according to Stair, and positive law does rest upon natural law, even if imperfectly. For him, ‘Equity is the body of the law, and the statutes of men are but as the ornaments and vestiture thereof.’87 ‘The formal and proper object of law’, Stair continued, ‘are the rights of men.’88 Stair built his legal work around this central idea of rights, beginning with the constitution and nature of rights; then their transfer from one person to another; and finally their enforcement. Rights were constituted by either divine or human law. Divine law was the law of nature, the first principles of which were known to men ‘without reasoning or experience’.89 Here the Pauline conception of humanity’s weakness is clear along with the idea that God has written on the hearts of people what they should do. This covered only certain basics, however—‘that God is to

85 

Stair (n 72) 266. an analysis of Stair’s political and religious thought, see: JD Ford, ‘The Rational Discipline of Law: A Historical Study of Stair’s Institutions of the Law of Scotland’ (unpublished PhD thesis, ­University of Cambridge, 1988) chs 6 and 7. 87  Stair (n 58) I, 1, 17. 88  ibid, I, 1, 22. 89  ibid, I, 1, 3. 86  For

Private Autonomy and the Protection of the Weaker Party: Historical  285 be obeyed, parents honoured, ourselves defended, violence repulsed, children to be loved, educated and provided for’. But the application of this law in more ­particular cases might be deduced by human reason. Stair explained: This law of nature is also called Equity, from that equality it keeps amongst all persons, from that general moral principle, Quod tibi fieri non vis, alteri ne feceris. … But equity is also taken for the law of rational nature, whereby nothing is to be done, which is not congruous to human nature, and becoming the same; and whereby that is followed, which if it were generally observed in the same circumstances, mankind would be happy. … This law of the rational nature of man is not framed or fitted for the interest of any, as many laws of men’s choice be: from the rigour whereof recourse ought to be had to this natural equity. … And though equity be taken sometimes for the moderation of the extremity of human laws, yet it doth truly comprehend the whole law of the rational nature; otherwise it could not possibly give remeid [remedy] to the rigour and extremity of positive law in all cases.90

Law was ‘a rational discipline, having principles from whence its conclusions may be deduced’.91 Accordingly, humanity was able to reflect in the positive law the natural law which God had written on his heart, albeit he would always do so imperfectly. The principles of equity, which were the efficient cause of rights and laws, were people’s obedience to God; the freedom of persons otherwise; which, however, being in their power, they might constrain by voluntary engagement with others. The three principles of positive law, which were the final causes or ends for which laws were made and rights established, were society, property and commerce. The principle of obedience produced those positive law obligations not resulting from voluntary engagement, ie what we would now call the law of delict (tort) and the law of unjustified enrichment (restitution), as well as much of family law. Where inequality arose so that one lost while another gained, restitution—ie restoration of the former equality—was required. This underpinned both the law of delict (or tort, or wrongs) and what we would call unjustified or unjust enrichment. Liberty led to personal freedom of action outside these obediential obligations, and also to the right of what Stair called ‘dominion’ over other things and creatures, ie property; while voluntary engagement produced the law of promises and contracts, which Stair termed ‘conventional obligations’.

IX.  Stair’s Conception of Liberty Stair began his analysis of the specific rules of law with the proposition that individual liberty—‘a natural faculty to do that which every man pleaseth, unless he be hindered by law or force’—was also ‘the most native and delightful right of man,

90  91 

ibid, I, 1, 6. ibid, I, 1, 17.

286 

Hector L MacQueen and Stephen Bogle

without which he is capable of no other right’.92 It would appear, however, that in offering a legal description of human liberty, Stair adopted a negative conception of the subject. He explained that liberty, although ‘the most precious right of man’, was not absolute. It was limited by, amongst other things, a person’s consent to the exercise of power over him by others.93 That included the initial consent to the exercise of power by a political sovereign.94 But more important for Stair were the ‘conventional obligations’ between private persons, which also ‘do arise from our will and consent’.95 By engaging with another in the appropriate fashion, a person could choose to give up liberty and become subject to that other’s power of exaction, ‘whereby he may restrain, or constrain us to the doing or performing of that whereof we have given him power of exaction; as in the debtor, it is the debtor’s duty or necessity to perform’.96 Stair’s understanding of liberty should not be mistaken for the same conception one finds in Grotius, Pufendorf or Hobbes. On the one hand, in his legal treatise Stair appears to offer a negative conception of liberty; on the other hand, when one reads his theological treatise a positive conception of liberty surfaces. It is only when both of these treatises are taken together that one gains a deeper understanding of Stair’s legal thought and particularly his contractual thought. In turn, this helps in relating Stair to the natural law schools of the early modern period and their fusion of liberty and contract law. In the Institutions Stair says: Liberty is that natural power which man hath of his own person, whence a free man is said to be suæ potestatis, in his own power; and it is defined in the law to be a natural faculty to do that which every man pleaseth, unless he be hindered by law or force.97

Grotius,98 Hobbes,99 and Pufendorf100 equated liberty with freedom from constraint or freedom in a state of nature. However, in the Institutions Stair equates liberty with the space in which people are free from the coercion of positive law. Stair in this sense uses the term liberty to describe an island of freedom within a sea of restraints and constraints.101 Of course, ‘where obedience ends’ a person has freedom to choose what to do with liberty,102 and importantly, conventional

92  ibid, I, 2, 1–2. On the various readings of this text, see ADL Wilson, ‘The Textual Tradition of Stair’s Institutions, with Reference to the Title “Of Liberty and Servitude”’ in HL MacQueen (ed), ­Miscellany VII (Stair Society vol 62, 2015). 93  Stair (n 58) I, 2, 5 (whence the quotation), 6–16. 94  ibid, I, 1, 16 (second para). Stair makes no comment here to suggest that a person’s consent to the exercise of sovereign power can be withdrawn. 95  ibid, I, 10, proemium. 96  ibid, I, 10, proemium. 97  ibid, I, 2, 1. 98  Grotius, The Rights of War (n 14), I.I.X. 99  Haakonssen (n 5) 31. 100  Pufendorf (n 15) I, 6, 3. 101  Stair (n 58) I, 2, 5. 102  ibid, I, 2, 5.

Private Autonomy and the Protection of the Weaker Party: Historical  287 obligations come from a space where people are otherwise free from obediential obligations. He says: It is a great mercy to man, that God hath obliged him only in a few necessary moral duties, and has left him free in much more, … with a liberty ad contradictoria, that he may do or not do, and ad contraria, that he may do this or the contrary providing that whatsoever he do, (even where he is free) be ordered and directed to the glory of God.103

There are, however, restrictions on what people can do with this liberty. For example, Stair records in the Institutions a variety of contracts deemed unenforceable: contracts made by married women;104 contracts made by minors to their lesion;105 exorbitant usurious contracts;106 immoral contracts;107 hoarding;108 unreasonable prices in contracts for food and drink in common hostelries;109 and contracts made to defraud creditors (collusion).110 Moreover Stair’s idea about what people should do with their liberty is unashamedly direct: So man being sent into the World to behold the Works of God & Glorifie him, for doing whereof, he hath some Rules written in his Heart by the Law of Nature, and in the World of God, and for the rest, is allowed to do as he conceiveth most conducible thereto, that whether he eat or drink, or whatsoever else, he do, he do all to the glory of God, it were a sad rack on the consciences of men, if their errors and mistakes in the matter of expediency were to lie as a guilt upon their consciences: but that bona fides or conscientia illaesa, so much spoken of in the law, is that which cleareth and acquitteth men in such mistakes. From this freedom doth arise, our personal freedom and liberty, whereby men are sui juris, but also their power of disposal of other things within their reach, or that dominion which God hath given them over the creatures.111

Hence, in his legal treatise Stair makes explicit strong views as to what is a godly life. This is reaffirmed in his Divine Perfections: ‘There are three sorts of Men, the Godly, the Honest, and the Prophane; the Godly are governed by the Love of God, which is the Principle of true Piety and Religion, reaching not only to Divine ­Worship, but even to the Acts of Honesty and their own Happiness.’112 What is notable in the Divine Perfections, however, is that Stair makes plain that civil authority and its coercive power is distinct from spiritual authority and its instructive power. He says: [T]he Civil Government is for the outward State of their Society, and the Means of their Government is outward, by extrinsick Rewards and forcible Punishments. The Ecclesiastick Government is about the inward State of those of their Society, in so far as Man’s 103 

ibid, I, 1, 19. ibid, I, 4, 16. 105  ibid, I, 6, 44. 106  ibid, I, 15, 7. 107  ibid, I, 10, 13. 108  ibid, I, 10, 14. 109  ibid, II, 3, 72. 110  ibid, I, 3, 14. 111  ibid, I, 1, 20. 112  Stair (n 72) 264. 104 

288 

Hector L MacQueen and Stephen Bogle

Knowledg can reach, to promote Holiness, and internal and eternal Happiness, and about their outward Acts only, as they signify their inward Condition; and their Rewards and Punishments are only by application of the Divine Ordinances, in exciting Joy or Grief, Fear or Hope, as is conducible for the inward State, but without temporal Rewards, and forcible Punishments.113

God is the ultimate source of control and authority for Stair in both the Divine Perfections and the Institutions. However, in the former Stair says that civil ­government is instituted only to punish and reward the ‘outward State of their Society’, whereas ‘ecclesiastick government’ is about the ‘inward State’ and ‘their only Rewards and Punishments are …, in exciting Joy or Grief, Fear or Hope’. This cannot be realised through temporal rewards or forcible punishments. The purpose of humanity is to obey and bring glory to God, but at best civil authority has a very limited role in ensuring that citizens lead a godly life. Hence, humanity is free under the terms of positive law to contract. People are expected, nonetheless, to use that liberty to contract in such a manner that it brings glory to God. But there is no role for a civil authority to ensure that contracts adhere strictly to man’s spiritual obligations. For Stair, man does not need positive law to tell him what is required by God to lead a spiritual life. Stair argues instead that by revelation and reason God makes clear to people how they should lead a godly life. But he also sees people’s freedom as an integral part of God’s plan for humanity. For example, he says in the Divine Perfections: [B]y his Wisdom [God] saw it congruous to his own Nature, and for his Glory, to ­create his Rational Creatures with Reason and free Choice, that they should not act by meer Instinct as Brutes, but be governed by Rewards and Punishments; which necessarily importeth that sometimes they should be permitted to make wrong use of that Freedom.114

Acting according to reason is in accordance with the natural law prescribed by God for humanity. What is more for Stair, acting according to reason is exercising one’s liberty: I conceive Liberty may be better defined the Hability of Self-determination upon a rational Motive. This will agree both to the Liberty of God and of Rational Creatures, and to the plenary Power both of the outward and of the inward Acts, or to the Liberty of the inward Acts only, wherein Liberty doth chiefly reside, and which are exempted from Force and Necessity. I call it an Hability, which imports more than a Power, which sometimes is not in a present Capacity to act, and in the Schools is called a Power in actu signato, or in actu primo, as a sick Man hath a Power to walk: The Rule by which the Power is determined, is the last practical Act of the Judgment about that which is in Consideration, whether it be the doing, attempting, desiring, intending, or choosing of something that is in the natural Power, whether it be in the lawful Power or not; which

113  114 

ibid, 267. ibid, 71.

Private Autonomy and the Protection of the Weaker Party: Historical  289 therefore I understand to be a Judgment, not of what is judged to be just, but what is judged to be fit in the present Circumstances, or to be either good, as just, profitable, or pleasant. Liberty consisteth not by one single Act, but by different Acts, upon different Occasions or Objects. There is also a Freedom preferring one thing to another in choosing it, and rejecting the other, or the rest, if more than two be in the Thought; and this the Learned call a Liberty ad contraria, as the former is called a Liberty ad contradictoria, the acting as is judged fit comprehends both. And it is always even in the most inconsiderate Acts; for tho’ there be no Deliberation or Hesitation, yet there must be a Judgment, that the thing is fit to be chosen or done, which differenceth these Acts from the Acts of Brutes, whose Appetites follow immediately upon their Perceptions, without any ­Deliberation or Judgment.115

Stair can then be said to have a positive conception of liberty as defined by Berlin. To act freely is to act according to reason. Liberty is not about external constraint but is rather ‘self-determination upon a rational Motive’. This contrasts, of course, not only with the type of liberty that Hobbes envisages in the Leviathan,116 but also with that set out by Grotius or Pufendorf. Stair shares far more with the late scholastics Lessius and Francisco Suarez (1548–1617). The purpose of the juristic Counter-Reformation was to ensure that the laity gained salvation through the guidance of theologians who knew the positive law, and could frame it within the language of the Church.117 As Decock has shown, Lessius, Suarez and the other Jesuit theologians cemented the will at the heart of their account of contract law with a positive conception of liberty in mind.118 Where Stair differs from the late scholastics, and where he shares an affinity with Hobbes, Grotius and Pufendorf, is that he does not seek to establish right action, ie a godly life, through the concrete positive laws of the state.119 This is of crucial importance to how one understands Stair and his Institutions. The late scholastics, although not seeking to change the ius commune, did try to establish a theological hold on its interpretation and show its relationship to a godly life and salvation. They wished to show how positive law was part of one’s journey to salvation. Stair too was giving the sources of law in Scotland and the ius commune a religious gloss. But he did not believe that positive law could aid humanity on the road to salvation; he remained thoroughly Calvinistic in that sense. When speaking of positive law, Stair is far more similar to Hobbes, ie he manifests a negative conception of liberty in terms of the authority and role of positive law. This leaves Stair to describe contract and promise in a manner differing from how many other lawyers in Scotland and Europe would have described the law relating to ­contracting behaviour during the early modern period.

115 

ibid, 114-15. Tuck (n 14) 119-55. W Decock, ‘Jesuit Freedom of Contract’ (2009) 77 Tijdschrift voor Rechtgeschiendenis 423. 118  ibid, 445. 119  Stair (n 72) 267; Stair (n 58) I, 1, 20. 116  117 

290 

Hector L MacQueen and Stephen Bogle

X.  The Will as a Central Concept In a significant passage in the Institutions, Stair says that: ‘[I]t is not every act of the will that raiseth an obligation, or power of exaction. … We must distinguish between three acts in the will, desire, resolution and engagement.’120 Neither desire (a tendency or an inclination of the will towards its object), nor resolution (a determinate purpose to do that which is desired), were enough to create a right in another. To achieve a right, there had to be engagement (the conferral or ­statement of a power of exaction in another). Such engagement was possible by one person alone by way of a promise, with no requirement for any action by the promisee: ‘the obligatory act of the will is sometimes absolute and pure’.121 A contract (or paction), on the other hand, is ‘the consent of two or more parties to some things to be performed by either of them’, ‘not a consent in their opinions, but a consent in their wills, to oblige any of them’.122 For Stair, in determining whether or not any voluntary obligation exists, ‘it is much to be considered, whether the consent be given animo obligandi, to oblige or not’,123 ie the exercise of an individual’s private autonomy to create a legal relationship with another. The same words may or may not be interpreted as obligatory, depending not only upon the speaker’s intention, but also on the circumstances in which they are uttered: [I]f it be jestingly or merrily expressed, whatsoever the words be, there is no obligation; because thereby it appears there is no mind to oblige; [but] if the words be in affairs or negotiations, they are interpreted obligatory, though they express no obligation but a futurition, which otherwise would import no more than a resolution; as Titius is to give Mevius an hundred crowns, in any matter of negotiation, this would be obligatory, but otherwise it would be no more but an expression of Titius’ purpose so to do; yet because it is inward and unknown, it must be taken by the words or other signs, so if the words be clearly obligatory and serious, no pretence that there was no purpose to oblige will take place.124

The emphasis on ‘affairs or negotiations’, ie business dealings, as a context in which the necessary obligatory intention can be taken as a given, is worth noting here. Stair’s perception of engagement as a way in which persons exercised their fundamental right of liberty further meant that for him freedom and sanctity of contract were also fundamental. ‘There is nothing more natural than to stand to the faith of our pactions’, he wrote.125 In a famous aphorism he rejected the

120 

Stair (n 58) I, 10, 2. ibid, I, 10, 4. 122  ibid, I, 10, 6. 123  ibid, I, 10, 6. 124  ibid, I, 10, 6. 125  ibid, I, 1, 21. 121 

Private Autonomy and the Protection of the Weaker Party: Historical  291 ­ nenforceability of ‘naked pactions’; ‘every paction produceth action’,126 ‘with u even pactum corvinum de haereditate viventis … binding with us’.127 The major exception to the general rule of enforceability (apart from the examples cited ­earlier) was the pactum de quota litis, ‘to prevent the stirring up, and too much eagerness in pleas’.128 Decock has shown that for the late scholastics and moral theologians respect for freedom of contract sprang from respect for the moral value of acts of Godgiven free will rather than the Enlightenment and later perceptions of the economic and social value of persons pursuing their self-interest.129 While for Stair the starting point was certainly the moral position with regard to the exercise of free will, which being God-given, was subject to the obediential obligations also flowing ultimately from the will of God, a further clear perception was that one of positive law’s primary aims is ‘freedom of commerce’.130 This led him to take a narrower approach to equality of exchange in onerous transactions than many others, including Grotius.131 Equality was a subjective matter, with ‘no determinate or certain rule but [the parties’] own opinions’.132 Scottish custom recognised no equivalent to the laesio enormis except in the protection of minors.133 In the following concluding passage on the question, he turned to Roman rather than theological or canon law sources in recognising the reality of the marketplace: And therefore it is safest to conclude with the law, l si voluntate, C de rescin Vend which saith, This is the substance of buying and selling, that the buyer having a purpose to buy cheap, and the seller to sell dear, they come to this contract, and after many debates, the seller by little and little diminishing what he sought, and the buyer adding to what he offered, at last they agree to a certain price, or as Seneca says, lib 6, de beneficiis, cap 15. ‘It is no matter what the rate be, seeing it is agreed between the buyer and the seller; for he that buys well, owes nothing to the seller’. Therefore the equality required in these contracts, cannot be in any other rate than the parties agree on.134

126 

ibid, I, 10, 7. I, 10, 8. A pactum corvinum de haereditate viventis is a ‘crow-like’ bargain about the ­inheritance of a still-living person. 128  ibid, I, 10, 8; and see above text accompanying nn 104–10. A pactum de quota litis is an agreement for a share of the subject of a law suit. 129  See Decock (n 43) chs 3.3–3.5. 130  Stair (n 58) I, 1, 18. See further DN MacCormick, ‘Stair and the Natural Law Tradition: Still Relevant?’ in HL MacQueen (ed), Miscellany VI (Stair Society vol 54, 2009) 5. 131  cf Decock (n 43) ch 7. 132  Stair (n 58) I, 10, 14. 133  ibid, I, 10, 14. 134  ibid, I, 10, 14. Professor Walker as editor supplies the reference to the Code: C.4, 44, 8. The other reference is to Book 6 Chapter XV of a dialogue by the Stoic philosopher and dramatist Seneca the Younger (c 4 BCE–65 CE) entitled De Beneficiis (On Benefits). 127 ibid,

292 

Hector L MacQueen and Stephen Bogle

This would certainly have chimed well enough with the thought of Adam Smith and David Hume in the eighteenth century, and indeed most of the lawyers and economists of the nineteenth century.135 But Stair also considered and offered protection to the weaker party. Since contracts are ‘deeds of the rational will’, they could not be entered by those lacking the power of reason, such as ‘infants, idiots [and] furious persons’ as well as those affected by ‘fear’, ‘drunkenness’, ‘disease’ or ‘err[or] in the substantials of what is done’. As already noted, minors were protected against lesion in their contracts.136 The ‘act of contracting … of purpose to oblige’ had to be in relation to things within the parties’ powers; ‘contracts of impossibilities’ and ‘in things unlawful’ were void.137 Some examples of the latter have already been mentioned, such as usury and the charging of unreasonable prices for food and drink in common ­hostelries.138 There might further be abatement of price for the latent insufficiency of goods sold, while penalty or similar clauses ‘ought to be and are reduced to the just interest, whatever the parties’ agreement be’.139 Fraud (which might be presumed in certain circumstances and was generally a much wider concept for Stair than that found in modern law) and extortion were wrongs giving rise to the obediential obligation of reparation, which could be set off against the obligations arising under any resultant contract (rather than simply striking it down).140

XI.  Freedom of Contract in the Eighteenth and Nineteenth Centuries The Enlightenment knocked down the theological presuppositions which had supported the thinking of the Spanish scholastics and the northern natural lawyers. But it did not remove the standing of private autonomy as a central concern for the law. David Hume challenged the idea of promises (and also property) as morally binding, arguing that their recognition was a purely human construct reflective of people’s need to live with each other in relatively peaceful and stable societies. Adam Smith emphasised the social benefits which flowed from each individual pursuing his or her self-interest but having to reconcile that with o ­ thers’ self-interest through the legal mechanisms provided by contract and p ­ roperty. 135  Hume (n 25) III, 2, 5 (‘Of the Obligation of Promises’); Smith, The Wealth of Nations, I, 7 (‘Of the Natural and Market Price of Commodities’), ed RH Campbell, AS Skinner and WB Todd, (Oxford, Oxford University Press, 1976); PS Atiyah, The Rise and Fall of Freedom of Contract (Oxford, Oxford University Press, 1979); W Swain, The Law of Contract 1670–1870 (Cambridge, Cambridge University Press, 2015). 136  See above, text accompanying nn 105, 133. 137  For the foregoing, including all quotations, see Stair (n 58) I, 10, 12–16. 138  See above, text accompanying nn 104–10. 139  Stair (n 58) I, 10, 14–15. 140  ibid, I, 9, 8–14.

Private Autonomy and the Protection of the Weaker Party: Historical  293 The exercise of private autonomy, or positive freedom, was the best way to achieve socially desirable results. Legal positivism, however, increasingly rejected the notion of any link, let alone a close one, between law and morality; and the thinking of lawyers came to be more under the influence of economists than of the moral philosophers. As Franz Wieacker put it (in Tony Weir’s translation): ‘[F]reedom of contract and association were becoming means of achieving social and economic power in the advancing Industrial Revolution.’141 A powerful effect of these trends in the nineteenth century was a lessening in the protection of the weaker party which had previously been afforded by natural law thought. Equality of exchange or just price notions were derided. At least in Scotland and England, the concepts of both fraud and invalidating error were progressively narrowed down by judicial decisions. Further, the growth of heavy industry, mass employment, mechanised mass transport systems for people as well as goods, and consumer markets was not to be unduly impeded by the imposition of liabilities upon entrepreneurs to those hurt or damaged in the process. In any event entrepreneurs were increasingly protected by the corporate forms under which their businesses were transacted from any personal liability for harm caused. The un-negotiated standard form was accepted as nonetheless a contract to the terms of which the non-proferens was taken to have agreed to make them binding even if directly contrary to that party’s interests. Any autonomy for weaker or socially disadvantaged parties lay primarily in the decision to contract or not (which was often not a very real choice), rather than in relation to the substance of the contract once formed.142

XII.  Twentieth-Century and Social Law The moves during the twentieth century towards tenant, employee and consumer protection countered the nineteenth-century pattern of development, but not so much through the mechanism of general contract law as by way of special provision, much of it regulatory or even criminal law in nature. The welfare state postWorld War II seemed to accelerate a move away from private to public law. Back in the 1960s Wieacker analysed what he saw as a movement from the bourgeoisliberal state to the soziale Rechtsstaat (‘by which we mean a state in which society takes responsibility for the social, economic and moral well-being of its members while respecting the independence of law and of individual rights’): ‘[T]he three crucial features of this trend are the modification of private rights by reference to their social function, the restriction of legal powers by social ethics, and the retreat

141  F Wieacker (trans T Weir), A History of Private Law in Europe (Oxford, Oxford University Press, 1995) 380. 142  ibid, 380–83 (Germany).

294 

Hector L MacQueen and Stephen Bogle

from the formalism of the classical private law system of the nineteenth century.’143 Social law was taking over from private law: ‘[S]ocial and economic processes should be controlled by co-operation between groups in society or through a settlement mediated by the public power rather than by either direct state regulation or the free play of individual autonomy operating independently.’144 At the same time, the social state’s acceptance of the rule of law ‘must at least leave private law with the role of identifying and delimiting areas of freedom of individual action’. Wieacker continued: This is still fully accepted wherever these rights can be exercised without involving social power, as in the areas of the right of personality, personal and property rights within the family, property for consumption, and associations which do not impinge on vital social functions: our courts are therefore right to protect personality rights as against the public’s interest in being entertained or even informed and … leave questions of direct social utility out of account. The same is, or ought to be, true as regards freedom of contract and association in the ‘ideal’ sphere, that is, for private, cultural or intellectual purposes not impinging on the economy or the rest of society.145

It followed from the final sentence quoted, however, that the exercise of autonomy in areas affecting economic or social power was no longer to be left free. Instead, it was appropriate to move much more to ensure ‘that the distribution of resources, the creation of assets, and the maintenance of the necessities of life are effected in a just manner’.146

XIII.  Twenty-First Century: Whose Autonomy? Whose Weakness? The concept of autonomy is an elusive one. But it remains possible to say ­something about it, even if in very general terms. As has been shown, in the wake of World War II Berlin argued that to adopt an autonomy-based conception of freedom or liberty requires one to determine questions about whose autonomy, who is weak, and ultimately, who is the right authority to determine such questions. He argued that history has shown autonomy to be an extremely malleable idea, prone to exploitation as a means to justify coercion and force. As far back as the seventeenth century, one can see in the writings of Stair two things which are of note for present purposes: firstly, a strong conception of a godly life (or a good life); but, secondly, a hesitancy about using the coercive force of positive law and civil government to implement that godly life. The resultant tension is evident in

143 

ibid, 427–28. ibid, 434. 145  ibid, 486. 146  ibid, 486. 144 

Private Autonomy and the Protection of the Weaker Party: Historical  295 his contractual thought in that he said one’s freedom to contract should be used to bring glory to God, but he nevertheless characterised the liberty granted to man under the positive law in negative terms. When put into context, Stair’s conclusion is understandable. He lived through a period of tremendous political and religious upheaval and experienced first hand the bloodshed that can follow when the ends of life are bitterly and bloodily disputed. This may, in part, explain why he conceived of positive law in a limited sense and demonstrated an apprehension about trying to implement a thicker conception of freedom through positive law, including the regulation of contract law and human liberty. Stair did not therefore use the ideal of people’s liberty to justify spiritual regulation or restriction of their contracting freedom by positive law or civil government. Much has changed since then, but viewing Stair’s work in this light helps one properly contextualise contemporary debates about contract law, its purpose and the use of positive conceptions of liberty. The ideal of autonomy may be used today to justify selfish and ultimately antisocial individualism. This leads to its rejection as a moral or political force by those who argue instead that a condition of heteronomy is preferable. A person’s sense of individuality in this view is self-deception and merely a social construct which can only be exercised meaningfully against those less powerful than that person. Between these poles of opinion there exists a range of views. Yet it may not be an untenable concept.147 The writings of the late Neil MacCormick provide a useful reformulation. He said autonomy is a quality of persons, that is of conscious acting subjects, who shape their will according to reasonable judgments based on the information derived through external and internal senses and gained in discussion with others or from reading and reflecting. By that rational will they construct and act upon conclusions concerning duties and rights that are inherently universalizable.148

But, he continued, Most of the moral position of each of us has emerged through a taking of individual responsibility for a body of moral opinion and tradition initially acquired heteronomously, and continuing reflection on the quality of our principles therefore engages critical reflection upon inherited tradition. This again gives reason for co-reflection with others from other traditions. Above all, the insight that one is oneself autonomous entails recognition of the like autonomy of every other, hence the equality of all moral agents as such. Any moral co-ordination of moral opinions can then only come about through the reaching of common but independently endorsed conclusions, and this implies a ­readiness to engage in fully open and non-coercive co-deliberation with others.149

In other writings, MacCormick proposed ‘the Smithian categorical imperative’, in which Adam Smith’s figure of the ‘impartial spectator’ whose moral views were 147 

See above, text accompanying n 9. N MacCormick, Institutions of Law: An Essay in Legal Theory (Oxford, Oxford University Press, 2007) 250. 149  ibid, 251–52. 148 

296 

Hector L MacQueen and Stephen Bogle

based upon feelings of sympathy with others perceived to have been wronged was brought together with Kant’s understanding of what constituted moral autonomy, briefly stated at the outset of this piece.150 MacCormick saw the resultant blend as ‘judgement-oriented’ rather than ‘legislation-oriented’, yet as still fully compatible with the value of autonomy, because the actor has to make its own decision about what to do. Even if regard is had to precedents, rules or principles, that is a matter of the decision-maker’s own integrity in viewing the decision as compatible with a whole way of life which has a social context. The kind of autonomy that is possessed by the person who seeks to exercise self-­ command, and to judge, and decided in the way mandated by the Smithian categorical imperative, is the very autonomy that seems worth asserting. This in fact seems to capture well our capability to act and to judge each for herself/himself, while maintaining, at least in aspiration, shared maxims of judgement as common standards of (dis-) approval, not a set of solitudes or some grand but self-justified solipsism.151

This may provide a modern conception of autonomy avoiding not only the ­dangers Berlin described but also the selfish or antisocial use of autonomy. The question, however, remains as to whether autonomy is an a­ppropriate ­principle with which to underpin modern contract law and, if so, whether it needs to be revitalised. Even in the half-century since Wieacker wrote, one might ask whether the trends he saw are still apparent in a world where communism has almost fallen, socialism is in retreat, and instead we see privatisation, the ­‘contracting-out’ of previously public functions, the ever-increasing power of multinational corporations and the globalisation of markets both on and off line. How far, if at all, have either private autonomy or protection of the weaker party continued to be legally meaningful concepts for the individual human being in those contexts? Human rights law may have bettered the individual’s position against the state; but how far has the horizontal as opposed to the vertical effect of human rights actually enhanced the position between individuals and other ­non-public actors themselves?

150  N MacCormick, Practical Reason in Law and Morality (Oxford, Oxford University Press, 2008) 63–66. 151  ibid, 67.

17 Personal Freedom and the Protection of the Weak through the Lens of Contract: Jurisprudential Overview DORI KIMEL

I. Introduction The relationship between personal freedom and the normative basis, the scope, and the appropriate means for the protection of the weak is one of the most contentious, as well as most pressing and most universal, questions of political and legal theory of our time. And contract law offers a uniquely suitable lens through which to examine this relationship, because both of its constituent parts, as well as the relationship itself, are implicated in contract law doctrine, and in the philosophical foundations of contract law, in what is probably a more central and more paradigmatic way than in any other legal domain. A rough sketch of contract as a legal domain that is characterised primarily by voluntariness and bilaterality should suffice to explain why this is so. Contract law has been seen traditionally as the one legal domain that is tasked primarily with the enhancement of personal freedom, and indeed, for many, the distinction between contract and ‘neighbouring’ legal institutions is to be understood in precisely such terms. The distinction between contract and tort, for example, has focused traditionally on the voluntary nature of contractual obligations, showing those to be, unlike tort obligations, freely assumed as well as self-authored. Contract law, so the standard story goes, confers the power on parties to shape their own obligations, whereas tort law imposes obligations; contract law provides an extension of our freedom, whereas tort law is in the business of restricting it—or, at most, of protecting it against encroachment by others. Tort law, if you like, makes us less susceptible to suffering loss—including loss of valuable freedoms—at the hands of others (in much the same way that the criminal law does); but contract law is unique in giving us net freedom, making us freer overall. Meanwhile, when it comes to the protection of the weak, another distinctive characteristic of contract comes to the fore, and that is the bilateral nature

298 

Dori Kimel

of contracting. This characteristic brings issues such as prevalent inequalities of bargaining power, sharp commercial practices, scope for exploitation and advantage-taking (and so on) into a particularly sharp relief, because it makes it all but inevitable for contractual activity to bring parties of unequal strengths into contact. Such contact can sometimes have devastating implications for so-called weak parties—devastating, inter alia, precisely in terms of loss of personal freedom, a prospect that sharply contrasts with that grand ethos of the institution of contract as a facilitator and promoter of personal freedom. Moreover, it has long been observed that an unfettered freedom of contract does not merely create an arena in which pre-existing inequalities of pertinent strengths are reflected, but that by doing that it has the tendency to exacerbate rather than ameliorate such inequalities: the marketplace it facilitates can be predatory by nature, so that strong predators are likely to thrive in it and get stronger still, whereas those who enter it weak are more likely than not to become weaker still. Already in this sketch of contract law as a legal domain that is centrally informed by a certain ideal of personal freedom but which at the same time places particularly dangerous pitfalls in the path of weak parties (and, indeed, exacerbates and perhaps even generates particular forms of weakness), we can see not only that the two components of this equation—personal freedom and the protection of the weak—are implicated particularly centrally in contract law, but also that the relationship between them comes into sharp focus in this domain. We can see the potential to view this relationship as, predominantly, one of tension: the extension of personal freedom and the need to protect the weak can be thought of as pulling contract law in opposite directions. I will say a little more about this idea.

II.  The Tension Thesis The view to which I will allude as ‘the tension thesis’ equates the ethos of contract as a facilitator of personal freedom with laissez-faire politics, and hence with general antipathy towards restricting or regulating contractual activity with the aim of protecting weak parties. In its extreme form, it sees all restrictions on the freedom of contract which are thus motivated—with the possible exception of those doctrines concerned with the most direct attacks on the bare notion of ‘voluntariness’ in contractual undertakings, such as duress or (to a limited extent) undue influence—as departures from this ethos. Those who subscribe to the tension thesis need not, however, view all such departures as illegitimate, or even undesirable. Indeed, the tension thesis has been offered with contrasting motivations and ideological underpinnings. Whereas versions of it have been deployed in defence of (usually right-wing, libertarian) objections to the legitimacy or desirability of protecting weak parties in the contractual arena (or, more broadly, the desirability of attempting to curb the outand-out predatory nature of the market), other versions have been offered by those

Jurisprudential Overview

 299

who, by contrast, think that it is in fact incumbent on legal orders to protect weak parties, at least against the harshest implications of the market. For those who believe that some such protection is a legitimate, desirable or even mandatory goal for the legal order, the alleged tension between that goal and the ‘personal freedom’ ethos has been used as a means for discrediting the latter, rather than the former. The canonical iteration of this sort of theoretical stance has been offered by Patrick Atiyah. In his seminal book The Rise and Fall of Freedom of Contract,1 as well as in subsequent writing, Atiyah identified an extreme, all-but-unlimited vision of the freedom of contract—one that leaves no scope for even the mildest forms of protection even of those most palpably in need of it—as the direct concomitant of seeing contract as based on personal freedom or moral autonomy. He powerfully attacked that vision, as well as the theoretical approach by which it is underpinned, as not only out-of-touch (and to an increasing degree) with legal reality, but also as morally bankrupt. ‘Whatever its normative power as a matter of philosophical or political argument’, he wrote, ‘liberal theory seems unlikely to carry the day in a democratic society.’2 Considerations related to respect for moral autonomy ‘can ring very hollow when used to defend a grossly unfair contract secured at the expense of a person of little understanding or bargaining skill’.3 Atiyah was right, of course, to maintain that talk of respecting moral autonomy rings hollow when used to defend exploitation and gross unfairness, but wrong— and quite spectacularly so—in his understanding of such respect as liable to produce this sort of talk in the first place. In the rest of this overview, I will argue that the view by which the promotion of personal autonomy and the protection of weak parties are predominantly contrasting or mutually excluding enterprises could hardly be further from the truth. Modern thought about the moral and political significance of freedom has done two things in the context of contract theory: it has provided ever more reason, and a deeper foundation, for seeing contract law as grounded in freedom, but at the same time it has also made the traditional perception of tension between that ethos of contract and the need to protect weak parties look, philosophically speaking, increasingly incoherent. It shows that the richest, most meaningful and morally defensible conception of personal freedom is not only compatible with various restrictions on the freedom of contract—including many which are aimed specifically at the protection of weak parties—but demands them. It is a moralpolitical ideal and a conception of personal freedom which not only leaves scope for the protection of weak parties, but in many ways creates that scope and tends to broaden it. It identifies what constitutes weakness in the first place, it mandates the protection of the weak, and it provides important guidelines on the means by which to do it. 1 

2 

P Atiyah, The Rise and Fall of Freedom of Contract (Oxford, Clarendon Press, 1979). P Atiyah, ‘The Liberal Theory of Contract’ in Essays on Contract (Oxford, Clarendon Press, 1986)

128. 3  ibid, 148.

300 

Dori Kimel

III.  From Negative Freedom to Personal Autonomy The background for the demise of the tension thesis is a shift of emphasis, in moral and political thought more broadly, from focus on so-called ‘negative’ freedom—that is, the absence of coercion or other obstacles—to focus on positive freedom—that is, freedom as a capacity, as empowerment. Personal autonomy has emerged from this shift of emphasis as perhaps the master ideal, the very paradigm of positive freedom. For many, it provides the key to people’s expectations of life in general, the key to people’s expectations of the various political and legal institutions that help shape their lives, and perhaps the key to understanding what other freedoms (in both the negative and the positive sense) matter to us, and what freedoms we would be willing to eschew or sacrifice on a variety of worthy altars. At its core, personal autonomy is an ideal of self-authorship. It concerns people exerting control, through the exercise of independent choice, over a significant proportion of fundamental aspects of their lives. It captures the distinct value to be found in a life that is spent in the pursuit of freely chosen commitments, projects, careers, relationships and so on—a value that is absent in an otherwise identical life in which such pursuits are not freely chosen but are somehow dictated, or chosen in ignorance of alternatives, or embraced through lack of real choice.4 Because personal autonomy is essentially an ideal of self-authorship, it fits quite naturally with the view of contract as rooted in freedom, and its growing dominance as a moral-political ideal has given this view of contract a new lease of life: what better manifestation of self-authorship than a person acting, inter alia, as the author of her own legal obligations? As the legal domain which facilitates precisely this, contract has come to be seen as a paradigmatic extension of personal autonomy, and hence as a freedom-maximising institution in a rather profound sense.5 What is more instructive for present purposes, however, is to appreciate the implications of the theoretical shift from negative freedom to personal autonomy in terms of the scope for protecting weak parties in the contractual arena. The key to this, I believe, lies in understanding the relationship between personal autonomy and the availability of options. This particular conception of freedom does not require, not even prima facie, that people simply have as many options to choose from in as many scenarios in which choice is called for, and as little interference in the choosing. What it does require is a sufficient range of options pertaining to central dimensions of people’s

4  Here and elsewhere in this comment I am relying on Joseph Raz’s account of personal autonomy. See J Raz, The Morality of Freedom (Oxford, Oxford University Press, 1986) esp chs 14, 15. 5  I have expanded on this theme in From Promise to Contract: Towards a Liberal Theory of Contract (Oxford, Hart Publishing, 2003), and subjected it to fresh scrutiny and some refinement, in ‘Personal Autonomy and Change of Mind in Promise and in Contract’ in G Klass, G Letsas and P Saprai (eds), Philosophical Foundations of Contract Law (Oxford, Oxford University Press, 2014) 96.

Jurisprudential Overview

 301

lives, as well as the capacity and the opportunity to make such choices independently, and make them well. Moreover—and this is where the true chasm between negative freedom and personal autonomy, as rival conceptions in the philosophical foundations of contract, becomes most apparent—it is not focused on options per se, but on quality options: options that have some positive meaning and value, and at least the potential to enhance rather than detract from the moral quality of the lives of those who choose them. Since personal autonomy does not attach significance to the absence of obstacles and interference for their own sake, but rather to people’s ability to be meaningfully in charge of their lives, prizing it does not produce a prima facie reason to want people to have open to them the kind of options which would only render those who choose them less autonomous overall, or less in control of their own destinies in the long run. Prizing personal autonomy, in other words, does not entail commitment to a self-defeating vision of it. Furthermore—and this pertains particularly to would-be strong parties, rather than weak—personal autonomy does not recommend and does not require the general availability of options which, when chosen freely, turn the freedom enjoyed by the chooser to be a component of a moral blemish on their lives, rather than a hallmark of the value of their lives as autonomous agents. As Joseph Raz put it, personal autonomy depends for its value on the worthiness of its exercise.6 The fact that I have chosen autonomously to become an academic can give my life as an academic a certain added value which would have been absent had I had no real choice in the matter, or had I made that choice under hypnosis, severe manipulation or coercion. If I join the British National Party, by contrast, hypnosis or manipulation would be mitigating features, whereas autonomous choice, morally speaking, would only make matters worse. Prizing personal autonomy means seeing special value in the independent embrace of that which is worthwhile, or at least acceptable; it does not entail seeing value in deplorable pursuits when these are embarked on in the least excusable circumstances. We can see that, by stark contrast to the tension view, commitment to the ideal of personal autonomy, once properly understood, produces, if anything, a prima facie reason to remove worthless options, or at any rate to prevent conditions arising in the first place in which people are particularly likely to choose them—because they find themselves under pressure to do so, or under severe temptation, or through manipulation or exploitation or ignorance or sheer lack of choice. As an inspiration for any sort of political action, this sort of conclusion is far broader in its implications than our current focus. The protection of weak parties, as something that is systematically required by commitment to personal autonomy, is merely a private case. It is an important private case, however, and one that is brought into sharp relief in the contractual context, because here it is particularly clear

6 

Raz (n 4) 378–81.

302 

Dori Kimel

to see how the same logic applies equally to both sides of the strength–weakness equation: by curtailing both the opportunity to be the predator and the risk of falling prey, the legal order can be seen as increasing, rather than undermining, the prospects of both sides to partake in the value of personal autonomy.

IV.  Familiar Objections: Paternalism, State Neutrality, Self-Defeating Means A prima facie reason to remove worthless options is just that—a prima facie reason. The ultimate strength of the reasons supporting any measure designed to protect weak parties, as well as the strength of all counter-considerations, would depend not only on the merits of the individual case (what kind of weakness is at stake, what are its causes, protect against what exactly, etc), but also on matters such as the legal domain in question, the kind of involvement the state has in creating the conditions in which weaknesses exist in the first place or are liable to render those who display them vulnerable, the state’s involvement in furnishing actors with particular options in this domain, and what exactly is involved in removing them. In the particular context of the law of contract, the process of reaching concrete policy conclusions is not immune, of course, from the usual need to descend to the coalface of practical reasoning and engage with the merits of individual cases. Nevertheless, it is a context in which certain time-honoured, general objections to the idea of the state playing an active role in removing autonomy-endangering options with the aim of protecting weak parties are particularly straightforward to dismiss. One familiar argument in favour of the laissez-faire approach, and specifically against the idea of the state actively removing (or limiting the availability or the attractiveness of) autonomy-endangering options, is that the overall quest for personal autonomy would be served better still by a general policy of leaving such options open, so that people can avoid them independently, as something which is an exercise of personal autonomy in and of itself. Dubious as it is in most contexts, when this sort of objection is applied to the need to regulate, with the aim of protecting weak parties, bilateral commercial activity that is habitually conducted under conditions of severe inequalities of bargaining powers and other relevant strengths (knowledge, access to justice, etc) and is often subject to market conditions that further exacerbate the vulnerability of the weak, it looks particularly vacuous, if not downright cynical. In this context it usually amounts, in practice, to endorsing a state of affairs whereby members of potentially vulnerable groups (consumers, tenants, employees, the very poor, etc) have nothing other than bad options open to them. Leaving them in that situation supposedly with the aim of allowing them to avoid bad options freely can only be described as adding insult to injury.

Jurisprudential Overview

 303

This sort of objection is sometimes linked to a broader opposition to state paternalism. Whatever its merits in general, however, anti-paternalism has no meaningful perch in the present context. The bilateral nature of contracting, alongside the largely symmetrical effect of choice-curtailing measures designed to preserve autonomy—personal autonomy, as we have seen, suffers on both sides of the predator–prey equation—means that paternalism is hardly ever truly at issue in this context. Measures that have the effect of preventing weak parties from agreeing to be exploited at the same time prevent strong parties from engaging in exploitation. Such measures can be, and usually are, defended on the latter ground, rather than (or at least alongside) the former. Sweeping aversion to pure paternalism, inasmuch as it has a basis at all, has little application in this arena. Another distinct characteristic of the law of contract—this time, its facilitative nature—makes it easy to dismiss an alternative, perhaps broader ground for laissez-faire ideology in this domain, and that is the supposed virtue of state neutrality.7 Contract law confers the power upon parties to make legally binding agreements, and avails them, in the pursuit of such agreements, of the adjudicating and enforcing agencies of the state. That means that the legal order is implicated in the moral quality of the agreements it recognises and enforces in a particularly direct way. When the law fails to criminalise a particular type of exploitation, or advantage-taking or undue influence or manipulation, it merely fails to do what (arguably) it ought to do. But when the law recognises and enforces agreements which are morally tainted in similar ways, it actively partakes in enabling them as well as in securing their ends. When the law of contract allows strong parties to capitalise on the vulnerability of weak parties, the law does not merely fail to help the weak parties, but plays a central role in creating the conditions for their downfall. By recognising the validity of such agreements, it does not merely fail to curb predatory conduct, but actively supports it: first by creating the normative framework in which the predator prevails, and then by helping secure the rewards. This is the main reason to reject the notion that the legal order should—or that it even can—remain neutral with regard to contests between weak and strong parties to contract. In the contractual arena, where the law is a facilitator—and, at that, a facilitator of bilateral activity—it is usually the case that the law has no option but to take sides, and the reasonable way to take sides—reasonable, at least, so long as commitment to the protection and promotion of personal autonomy is at stake—often can only take the shape of curbing the freedom of contract so as to protect weak parties. Be that as it may, a general policy of non-intervention in the freedom of contract cannot genuinely be defended on grounds of neutrality, because it is not neutral: it amounts to coming down on the side of strong parties in their quest to capitalise on the weaknesses of their counterparts.

7  For an example of the tendency to attribute commitment to state neutrality to autonomy-based contract theory see Atiyah (n 2) 133. For my own discussion and references see From Promise to Contract (n 5) 121–25.

304 

Dori Kimel

The analogy to the criminalisation of conduct brings to mind another source of circumspection, at least, with regard to the idea of the state playing an active role in removing bad options (be it on paternalistic grounds or otherwise) in the name of protecting personal autonomy. That concern is that the means for removing such options can themselves be inimical to personal autonomy, or to the political and legal culture that commitment to personal autonomy requires a society to cultivate. The means themselves, that is, or the general willingness to deploy them, can have practical ramifications or symbolic dimensions such that their deployment is liable to do more harm to the conditions of personal autonomy than the good there is to achieve through the removal of particular autonomy-endangering options. This anxiety is well founded, and has an undoubted role to play in delineating the moral limits of the law in the service of promoting of personal autonomy. Particularly when it comes to paternalistic legislation, it militates against the use of certain means for the removal of autonomy-endangering options in all but cases of extreme necessity, and as anything other than last resort. The anxiety only pertains, however, to very specific means—primarily the use of coercion, which is indeed inimical to the very ethos of self-authorship,8 and, due to its tendency to be a blunt tool, is liable to have an excessive impact on its actual scope.9 The protection of weak parties in the contractual arena, however, hardly ever calls for the use of such means. Shaping the scope of the freedom of contract so as to maximise its autonomy-enhancing potential and prevent contractual activity and the institutions that support it from serving as instruments of gross unfairness or exploitation need not, and for the most part does not, rely on coercion, but on the systematic setting of conditions for the validity or enforceability of contract terms, the imposition of limitations on the type of goods or services for the provision of which contracts can be made, the setting of standards of acceptable conduct between parties, and the like. The outright criminalisation of certain types of contract is not a central part of the apparatus with which most jurisdiction set out to protect weak parties; it tends to be rather rare, and is usually reserved for the suppression of the most egregious types of exploitation of the most vulnerable, or for agreements to carry out that which is a criminal enterprise on its own.

V. Conclusion I will conclude by approaching the equation from its other side. The power to make legally binding agreements can make a significant contribution to personal

8  See M Dan-Cohen, ‘In Defence of Defiance’, in Harmful Thoughts (Princeton, Princeton University Press, 2002) 94. 9 The ‘blunt tool’ argument is Joseph Raz’s—see Raz (n 4) 418–20. For a discussion see John Stanton-Ife, ‘The Limits of Law’ (2006) Stanford Encyclopaedia of Philosophy, at: http://plato.stanford. edu/entries/law-limits/.

Jurisprudential Overview

 305

autonomy. It can be seen as a paradigmatic exercise of personal autonomy, and perhaps as a necessary condition for it. At the same time, this power also represents a particularly potent threat to autonomy: it is a power that can backlash, and can end up doing a great deal of harm to precisely the sort of values that support the idea that people should have it in the first place. It can render those who exercise it less in control in their lives, more vulnerable to exploitation by others.10 The likelihood of such a backlash is greater when it comes to weak parties, and this increased likelihood is in fact a central part of what defines ‘weakness’ in this context. Weak parties are precisely those for whom the institution, absent special protection, poses more of a threat to personal autonomy than serves as a platform for its expansion and successful exercise. They are those, in other words, whose position, either in general or vis-à-vis particular counterparts, is such that contracting, absent protection, would more likely lead them to relinquish valuable control over their lives rather than successfully gain or exercise such control. So the challenge for contract law, if it is to be able to justify itself as a freedommaximising institution, is not to give people simply as much power to make legally binding agreements, but rather to give them the right kind of power: the kind that would, overall, maximise the autonomy-enhancing potential of contracting, and minimise its autonomy-curtailing risk. The protection of weak parties thus emerges as not only compatible with the enhancement of personal autonomy in this domain, and not only as required by it, but as part of its very meaning. And this relationship is mutual: understating contract as tasked primarily with the promotion of freedom is a significant part of the enterprise of identifying relevant weaknesses, recognising the need to protect those who exhibit them, and devising the means by which to do so. Meanwhile, such protection is a significant part of what it means for contract to be rooted in freedom in the first place, and for the state to be able to justify its investment in this institution by reference to its general duty to promote the personal autonomy of its citizens. The way I see it, much of what has happened in the common law of contract and in the European Union law of contract over the past decades—the very same trajectory that Atiyah has analysed as a consistent move away from the entire ethos of freedom, and which manifests itself in ever-expanding restrictions on the freedom of contract which are primarily designed to make contractual activity safer for weak parties–is to be understood not as a move away from the ethos of private autonomy, but the exact opposite. It is the way in which autonomy-prizing political and legal cultures face up to the challenge of tilting the overall balance in the law of contract away from the endangerment of personal freedom, and in favour of its true enhancement. This, I believe, has proved to be the most enduring challenge in this domain, and the endeavour to meet it is set to remain the most powerful engine of development and reform.

10 

I have given an account of specific dimensions of this threat in ‘Personal Autonomy’ (n 5).

306 

18 Private Autonomy, Weak Parties and Private Law: Views from Law and Economics FERNANDO GÓMEZ AND MIREIA ARTIGOT

I. Introduction When analysing the law and legal institutions, or the legal dimensions of given social and economic matters, lawyers, including academic lawyers, tend to focus on the details and the specific features of the issues at stake. The more attentive ones also consider the circumstances that frame the legal side of things. But it is less common, we believe, to pay attention to large societal dimensions, and to ultimate social goals of the institutional aspects under analysis. Surely, values and principles loom large, sometimes in a dominant way, even, but this is not the same as keeping the broad picture in mind. When dealing with inequalities in interactions governed by private law rules, the above statements are also apt to describe prevalent approaches from legal academia, let alone by the operators of the legal system. We think it is important to remember what the fundamental goals of individuals and organisations who interact are, and to what extent such interactions are a source of social—and legal—concern. The economic view of things is, we believe, helpful to apprehend those dimensions of the analysis. In that light, social interactions are important because they allow people to improve their lot through cooperation and trade. In fact, from an economic viewpoint, getting agents to engage in, and succeed, in trade is crucial for social and economic prosperity. And trade cannot be understood narrowly as merely the exchange of goods or services for money, but should be conceived broadly as the set of interactions through which people obtain from others—individuals, firms, governments and other organisations—what they need or prefer to pursue their own goals and ends, given their capacities and endowments. This crucial process of trade is shaped by the composition and internal complexity of societies. When a society is characterised by homogeneous preferences,

308 

Fernando Gómez and Mireia Artigot

identical or very similar technologies and productivities, and scarce specialisation and division of labour, trade has a limited role to play. When, however, societies are defined, as the present ones are, by a variety of technologies, a diversity of preferences across individuals and groups, and by a high degree of specialisation in knowledge and skills, and a far-reaching division of labour, trade becomes an essential mechanism in the functioning of societies. Heterogeneous preferences, different skills and division of labour have accompanied human societies for centuries, if not even longer, and modern legal systems have always been designed and operated in societies essentially defined by those properties. Complexity, heterogeneity and specialisation also affect the organisation of trade and the resulting need for cooperation. In present-day societies, there seems to be substantial theoretical and empirical evidence indicating that trade is expected to expand and be more efficient under decentralised decisions and actions determined by the agents themselves, be they channelled through more or less organised markets, or outside of markets. Centralised decision-making is very unlikely to work as a general mechanism to organise cooperation and trade in large and complex social settings. This is not to say that all trade and cooperation decisions are decentralised in modern societies. In very different contexts, many interactions and cooperative endeavours are organised through centralised fiat, sometimes, or even often, after deliberation, bargaining and/or voting. However, it seems hardly imaginable that the myriad of transactions and interactions through which trade takes place in a complex society would not be essentially delegated to the individual parties involved and affected. Even if centralised allocation were imaginable on such a scale, the information and coordination costs of a social arrangement of that kind would be of enormous proportions, and the end results for society would be far from desirable. This decentralised ordering of trade and cooperation links the economic dimensions with the notion of private autonomy that is important (many would indeed argue it is a cornerstone) to private law in most legal systems. We do not claim that private autonomy in private law is a consequence of the decentralised allocation of decisions over trade as an economic or social phenomenon.1 Private autonomy may have evolved as a theoretical and legal notion independently of economic views of trade and social cooperation, but that does not eliminate the relevance of the link. This link also shows how focusing on the ‘individual’ transaction does not fully allow the proper understanding of both trade and private autonomy. A broader perspective is, we believe, necessary.

1 For instance, W Decock, Theologians and Contract Law. The Moral Transformation of the Ius Commune (ca 1500–1650) (Leiden and Boston, Martinus Nijhoff, 2013) 162ff underscores the ­anthropological and religious underpinnings of private autonomy among the European legal theorists of the 16th and 17th centuries, where private autonomy seems to have been crafted as a relevant notion in private law.

Private Autonomy, Weak Parties and Private Law

 309

II.  Gains from Trade at the Macro Level To understand the macro and micro advantages of transactions and trade—and subsequently the challenges involved in their efficient functioning—it is not superfluous to summarise the basic theory of international trade where these advantages appear distinctively. Individuals, as well as countries, do not have uniform endowments, technologies, production capacities, tastes and preferences, and natural resources. Such diverse conditions determine the productivity as well as the production levels different agents and economies can achieve. Because of this diversity, some agents and economies are relatively better placed than others for the production of certain goods and services. Trade, at the domestic and at the international level, allows parties to exchange goods and services without the need to produce everything individuals are interested in consuming. If agents and economies do not interact with each other, and hence do not exchange goods and services, their consumption capacities are conditioned by their production constraints. Trading with each other enables them to expand their consumption constraint beyond their production constraint. International trade is the exchange of capital, goods and services among different agents and economies, and in the global or supranational sphere, often involves exchanges across international borders. Whenever the different endowments and technologies available result in diverse productivity levels in different goods, it is well established2 that agents and economies should specialise in the production of what they are relatively better at doing, and through trading with others, obtain other goods they are relatively worse at producing. At the level, whenever economies have d ­ ifferent productivity levels in the production of different goods, i.e. have a comparative advantage in the production of one good, international trade ­models3 show that economies should specialise in those goods in which they have a comparative advantage in and engage in trade with other economies for other goods in which they do not have a comparative advantage. Comparative advantage refers to the idea that each economy should produce goods for which their domestic opportunity costs are lower than the opportunity costs of other economies when producing the same good. If productivity and hence, opportunity costs, are different, comparative advantage in the production of one good will allow an economy to produce the good in a cheaper way than its trading partners, and to sell it at a relative price higher than its opportunity

2 See a general and introductory overview on the Ricardian and Hecksher–Ohlin models in PR Krugman, M Obsfeld and M J Melitz, International Economics: Theory and Policy, 10th edn (London, Pearson, 2015). 3  The different contexts and assumptions that determine the degree of specialisation and the limits of the different productivity levels in the creation of surpluses have been broadly discussed in the different models of international trade and are beyond the scope of this paper.

310 

Fernando Gómez and Mireia Artigot

cost, but lower than the opportunity cost of the trading partners who would not buy the product otherwise. The difference between an economy’s opportunity cost of the product in which it will specialise and the relative price of the good paid in international markets will be the surplus generated by trade. This surplus, obtained by selling in international markets the products over which an economy has a comparative advantage, while buying elsewhere the products in which such economy has no comparative advantage in producing, allows economies to reach consumption levels well beyond their production constraint. Given these different productivity levels, trade among economies allows for lower prices in competitive sectors, benefiting both buyers and sellers. It creates productive advantages for countries and makes larger markets possible, which in turn allow for economies of scale—producing at larger scale offers higher efficiency potentials than producing at a lower scale. Overall, trade generates surplus for all the economies involved in it. It should be noted that gains from trade are generated for all the economies involved, gains that emerge from ther different productivity levels, and regardless of their respective levels in endowments, productivities and size.4 Despite the overall benefits generated by trade, this does not imply it benefits every member of the economies involved, or that it does not cause harm to those engaged in the industries in which the economy does not specialise. When an economy specialises in the production of goods and services in which it enjoys a comparative advantage, this represents a threat to non-competitive sectors—those where other economies are relatively better and hence are able to produce at lower prices than the ones they can offer domestically and internationally. The non-competitive sectors will be exposed to foreign competition that is more efficient, creating a threat to their continuing viability. Individuals employed in those industries, even if the economy overall benefits (perhaps even largely) from trade, are likely to be harmed, at least in the short term, and will resist the exchange with foreign markets and hence the involvement of the economy in trade.5 As happens all the time with phenomena at the macro or societal level, there will be winners and losers, since people are differently positioned with respect to the intervening factor or shock. Thus, it is not only important whether trade generates surplus (it does; otherwise it would not take place) but also how this surplus is distributed in order fully to grasp the implications at the macro level.6

4 

Krugman et al (n 2). B Ohlin, Interregional and International Trade (Cambridg, MA, Harvard University Press, 1933). Ohlin’s doctoral thesis was supervised by Heckscher, hence the so-called Heckscher–Ohlin trade model. 6  See S Galle, A Rodríguez-Clare and M Yi, ‘Slicing the Pie: Quantifying the Aggregate and Distributional Consequences of Trade’, mimeo, University of California Berkeley (2015). The paper is available at https://dl.dropboxusercontent.com/content_link/OPPYL1CdXXWXxrYhqNVajgBZKCnBmTUcCOEc CHUT9aQ7JUMunN2shB9AaBCf3y36/file. See also W Brooks and P Pujolas, ‘Gains from Trade: The Role of Composition’, University of Notre Dame Economics Working Paper (2015) available at http:// www3.nd.edu/~wbrooks/Composition.pdf 5  See

Private Autonomy, Weak Parties and Private Law

 311

Hence, one should differentiate between, on the one side, the aggregate gains from trade and the expansion of consumption possibilities beyond an economy’s production constraint, and on the other, the distribution of such gains across different economic agents or groups. Some may be harmed by international trade because of the more competitive foreign competition while others will benefit because of the economies of scale generated and the improved technology trade will result in. This leads unavoidably to problems of social choice. This basic insight from international trade can be applied at the micro level. That is, differences in productivity levels for different goods and services—in our previous discussion on international trade, the productive position of the different economies in the global world—coupled with the differences in preferences, make room for specialisation and for the creation of gains from trade for the different parties involved in it. This is not only true at the country level, but also, and perhaps even more so, for individuals and for organisations when considering their single interactions.

III.  Gains from Specialisation and from Trade at the Micro Level International trade provides valuable insights into interactions between individuals at the micro level. In parallel to what happens at the macro level, centralised decision systems of allocation of resources are not likely to maximise social welfare. Trade and exchange are pillars of economic growth. Contracts are the main instrument to allow for economic exchange. Through contracts, parties generate surplus beneficial for their own welfare. Individuals and organisations use contracts as instruments to arrange cooperation in exchanges and trade. Contracts allow agents to coordinate their actions to achieve the desired cooperative outcomes, and to implement and sustain cooperation outside immediate and simultaneous interactions.7 Party autonomy for the contracting players provides them with an essential welfare-enhancing instrument for deciding, shaping and regulating their transactions. Economic theory would suggest that when maximising welfare, parties would enter into contracts as long as their welfare levels under the terms of the contract were higher than their welfare levels outside the transaction. Even though often much neglected by economic theory, legal contracts and the supporting rules and institutions that accompany them are an essential tool to allow for welfare-enhancing transactions and thus for economic growth.

7  See B Hermalin, AW Katz and R Craswell, ‘Contract Law’ in A Mitchell Polinsky and S Shavell (eds), Handbook of Law and Economics, vol 1 (Amsterdam, North-Holland, 2007) 8ff.

312 

Fernando Gómez and Mireia Artigot

The contractual instrument, though, is not immune to caveats and limitations of different kind and magnitude. Basic economic theory models typically assume that parties have perfect information about all relevant issues affecting their actions and payoffs from the interaction (previous history, internal and external factors, etc). If that were true, contracts as voluntary transactions would arise as an efficient system whereby parties allocate resources maximising their utilities or, more generally, their welfare. At the same time, basic economic theory models often assume a symmetric position of parties entering into a contract for an economic exchange.8 Relaxing the perfect information assumption allows for a much more realistic setting. Also, a more accurate and illuminating view on the contracting context provides for endless examples where the position of the contracting parties is asymmetric. Such asymmetries may be generated by the different information the parties possess about the subject matter and other intervening factors, about the other contracting party and about background parameters affecting the transaction. Whenever information is not perfect, the parties’ position is not symmetric as to a relevant dimension and there are costs involved in contracting, parties may not be able to allocate resources according to their preferences so that they maximise their welfare. This does not mean that, subject to constraints determined by the information available, the situation of the parties and the costs of contracting, the parties will not try to optimise their contracting decisions, even if the end result may depart, perhaps widely, from the efficient allocation. But removing those constraints, or at least altering them to make them less binding, would improve the surplus generated by the transaction and thus overall efficiency. Regardless of the presence of contracting costs, imperfect information and asymmetries, private autonomy still emerges as a superior general mechanism for creating value than a centralised system of allocation of resources. Thus, even in an imperfect world, private autonomy understood as the framework of decentralised action by agents in diverse social settings, despite its shortcomings, still seems more appealing as an instrument for generating social and economic value than a centralised system, given that it allows for parties to adopt their decisions and adjust the contract terms to their preferences and maximise their welfare, given the constraints, even if the latter are important, as we have observed before. Informational and contracting costs will not disappear under a generalised central allocation, and it seems that centralised resources may be put to better use by positively affecting the constraints than by attempting to control allocations directly. The privileged position of private parties in contracting should not imply that there is no role for regulators and for governments to organise markets and other institutions in which these decentralised transactions take place. Governments, and more generally regulators, have a crucial role in sustaining systems through

8 See for the basic introductory theory on perfectly competitive markets, N Gregory Mankiw, Principles of Economics, 7th edn (Stamford, CT, Cengage Learning, 2015).

Private Autonomy, Weak Parties and Private Law

 313

which agents interact, for example, by creating a system of protection of property rights, by originating and running all institutions necessary to provide a reliable monetary system, and perhaps by providing a certain degree of social insurance through a welfare system.

IV.  Allocating Resources Efficiently: Challenges Arising from Private Autonomy as an Allocation Mechanism As was observed already in the previous section, the fact that private autonomy emerges as a superior general mechanism for generating private as well as social surpluses through individual transactions compared to a centralised allocation system does not imply that a decentralised mechanism is exempt from inherent limitations. Private decentralised action faces important challenges when aiming at allocating resources efficiently. Such challenges are present at two different levels. On the one side, they operate at the individual level, i.e. at the micro level, using the prevailing terminology in economics. On the other, challenges are also present at the aggregate level, in economic terminology, at the macro level. Exploring the two levels of analysis separately is useful in order to shed light on how legal policies should address the asymmetries present in transactions between agents. Such asymmetries arise in the context of, for example, inequalities between contracting parties or in cases that legal scholars may characterise as subordination of weaker parties. Asymmetries, though, reach well beyond these situations. At the micro level, parties exercising their private autonomy when involved in transactions face costs along several dimensions.9 These dimensions are: first, transaction costs; second, the creation of incentive and commitment problems; third, the informational level; and fourth, the cognitive and behavioural one. With respect to transaction costs, whenever these are positive, they may impact on the expected value of the transaction and as a result, on the parties’ negotiations and the terms of the transaction. At the same time, the existence of positive transaction costs may also produce barriers or obstacles in achieving successful cooperation between the parties involved in the transaction. The lack of full commitment to stick to the cooperative path of behaviour in a transaction generates incentives for opportunistic behaviour, and hence for not behaving in compliance with the terms of the agreement adopted. One, or both, of the parties in the transaction may be tempted to renege on the actions required for achieving and sustaining cooperation and reaching a welfare-maximising

9  It should be noted that transactions may also generate externalities, ie third-party effects. For the purpose of the analysis in this section, such effects are not considered.

314 

Fernando Gómez and Mireia Artigot

equilibrium. In this sense, problems of hold-up and opportunism may arise when one of the parties has incentives to maximise private surplus instead of the parties’ joint surplus, and hence not cooperate to reach the jointly maximised surplus.10 Additionally, information available to parties may not be complete and perfect regarding all the dimensions and parameters affecting the transaction. The level of information as well as the dimensions over which parties have information may strongly condition the surplus potentially created by the transaction, the parties’ position when deciding whether to cooperate or not, as well as the actions available to them. When one or more parties are afflicted by imperfect information, they have to make decisions without knowing where they stand in the game, so that they are influenced by beliefs and by commitments they may have and adopt in such contexts.11 Also, information may not be symmetric for all parties to the transaction. One or more of them may have information to which others are not privy. Getting parties to cooperate and to maximise gains from trade in such a scenario becomes much harder.12 For instance, consumers acquire only limited information about certain ­relevant characteristics and variables affecting the transactions on goods and services through various means, eg direct observation, past purchases or third parties. At the same time, market forces can, under certain conditions, induce producers to disclose information.13 However, market forces alone may not generate enough or adequate information for consumers to adopt their decisions. The legal system could be of use in generating the incentives to provide the amount of information to consumers that would be closer to the optimal in order to take their decisions. In order to generate such incentives the legal system could have mandatory disclosure requirements and standardised informative messages that could reduce the consumers’ search costs for information, while increasing the amount of relevant information for consumers before entering into the transaction. However, one of the remaining questions when introducing mandatory disclosure requirements

10 See VP Goldberg, ‘Discretion in Long-Term Open Quantity Contracts: Reining in Good Faith’, Columbia Law & Economic Studies Working Paper 176 (2000) http://papers.ssrn.com/paper. taf?abstract_id=234705, arguing that one of the factors influencing the risk of hold up depends on the access to market alternatives: the more difficult its be for contracting parties to have access to alternative markets, the more significant the risk of opportunistic behaviours. If, instead, parties had access to market alternatives, they would not be that vulnerable to the other parties’ strategic behaviour and this could minimise the price divergence between the contract price and the opportunity costs of the parties 11  For a theoretical yet accessible introduction to games with imperfect and asymmetric information, JE Harrington, Games, Strategies and Decision Making (Worth, New York, 2009) ch 9–10. 12  See B Hermalin, AW Katz and R Craswell, ‘Contract Law’ in A Mitchell Polinsky and S Shavell (eds), Handbook of Law and Economics, vol 1 (Amsterdam, North-Holland, 2007) 34ff. 13 On unravelling of verifiable information, see S Grossman, ‘The Informational role of Warranties and Private Disclosure of Product Quality’ (1981) 24 Journal of Law and Economics 461 and P Milgrom, ‘Good News and Bad News; Representation Theorems and Applications’ (1981) 12 Bell Journal of Economics 380.

Private Autonomy, Weak Parties and Private Law

 315

is the design of the kind and amount of information that should be disclosed, and this is no easy task, as the literature dealing with cognitive and behavioural limitations—more on this below—and information processing suggests that promoting disclosure might not only be difficult but unfeasible and, in some cases, even counterproductive.14 In addition to the above-mentioned structural challenges faced by parties when allocating resources through private autonomy, recent literature on behavioural psychology and behavioural economics has spotted other important issues based on the cognitive and behavioural shortcomings of agents in decision-making.15 When taking decisions with economic significance, agents may often decide, choose and act in very different ways to those predicted by traditional economic models based on the agents’ traditional (in economics, at least) concept of rationality.16 One of the areas where individuals are likely not to behave as rational choice models would predict is consumer behaviour. Consumers decide, in many cases, in ways remarkably different from those predicted by rational choice models, and thus sellers may take advantage of the consumer’s cognitive biases by aiming at shaping consumers’ expectations about a product advancing the seller’s interest— making it endogenous through marketing or commercial practices, for example— instead of correcting the biases.17 It is not a surprise, thus, that in light of the ample opportunities for firms’ practices to affect consumer information levels negatively, or to take advantage of consumers’ bounded rationality, the legal system intends to deter or discourage them through legal rules on misleading advertisement18 or on unfair commercial practices in a business-to-consumer relationship.19

14  O Ben-Shahar and CE Schneider, More Than You Wanted to Know: The Failure of Mandated Disclosure (Princeton, Princeton University Press, 2014). 15  For a technical treatment of behavioural matters, see DR Just, Introduction to Behavioral Economics (London, Wiley, 2014). For a presentation and discussion of general issues concerning the limitations of rational choice, and the impact of cognitive biases in legal analysis, see CR Sunstein, Behavioral Law and Economics (Cambridge, Cambridge University Press, (2000). 16  See MD McCubbins, M Turner and N Weller, ‘Testing the Foundations of Quantal Response Equilibrium’ in AM Greenberg, WG Kennedy and ND Bos (eds), Social Computing, BehavioralCultural Modeling and Prediction, Lecture Notes in Computer Science, vol 7812 (Berlin, Springer ­Verlag) 144–53, suggesting that individuals do not take decisions as predicted by rational theories that result in a Nash equilibrium but have cognitive limitations and biases and face strategic uncertainties that condition the equilibrium they may reach—in the author’s words a quantum response equilibrium. See also G Lucas, MD McCubbins, and M Turner, ‘Against Game Theory’ in R Scott and S Kosslyn (eds), Emerging Trends in the Social and Behavioral Sciences (Hoboken, NJ, John Wiley & Sons, 2015). 17 See R Spiegler, Bounded Rationality and Industrial Organization (Oxford, Oxford University Press, 2011) for an overview of the industrial organisation models when consumers show cognitive and behavioural biases that can be exploited by firms. 18  Directive 84/450/EEC of 10 September 1984 relating to the approximation of the laws, ­regulations and administrative provisions of the Member States concerning misleading advertising, [1984] OJ L250. 19  Directive 2005/29/EC of the European Parliament and of the Council of 11 May 2005 concerning unfair business-to-consumer commercial practices in the internal market, [2005] OJ L149, 22–39.

316 

Fernando Gómez and Mireia Artigot

A.  Challenges and Asymmetries at the Micro Level The inherent challenges to parties exercising their private autonomy through contracts do not always—perhaps even rarely—have symmetric impact upon both sides to the transaction. The challenges we have characterised result in asymmetries between the parties, inequalities between the agents transacting and differences between categories of agents. The information available to the parties involved in the transaction may be different regarding the amount each one may access and process, and may also vary as to the parameters and factors such information refers to. At the same time, some parties may be more prone to misperceive information provided by the other party. Some agents may also be placed at a disadvantageous position with respect to information regarding relevant factors of the transaction due to cognitive limitations or biases that may condition their expectations regarding the expected value of the transaction and their position as well as the other parties’ position in the transaction. Such information asymmetries and the misperceptions regarding relevant factors may create incentives for parties to behave opportunistically, maximising their private surplus instead of the joint surplus of the parties. Further, certain parties may systematically have larger benefits from opportunism or may adopt opportunistic decisions—lowering the quality of products or services, shrouding product attributes—that are hidden to the other party. At the same time, the costs of reacting against opportunism of the other party may not be the same for both. Consequently, one party may have higher incentives to behave opportunistically knowing that the other party would face high costs of reacting against such behaviour. The sum of all these factors conditions the position of the parties in the transaction as well as the surplus generated by the transaction and may generate important structural differences.

B.  Challenges at the Macro Level Parties face limitations at the micro level resulting from challenges when maximising the surplus generated by private transactions. Such limitations are also present at the aggregate level, the macro level, but additional complexities linked to the macro dimension also arise and exacerbate the difficulties of reaching efficient allocations. On one side, at the societal level, much more than at the level of the individual transaction, conflicts are pervasive. Changes in factors such as technology, economy, demography and legal frameworks benefit some parties while damaging others. At the aggregate level there are always changes going on, and their impact is always heterogeneous on different groups, territories and industries.

Private Autonomy, Weak Parties and Private Law

 317

At the same time, at the macro level, defining the rules and terms of transacting is much more complex than at the individual level. The aggregation of individual preferences does not result in a clear-cut definition of social preferences. Concerning some issues where there is general consensus, given that individual preferences are homogeneous or at least reasonably so across individuals, aggregation is reasonably feasible. Individual preferences, though, are generally quite diverse and hard to adjust in the face of an individual transaction. This is much more so when diverse individual preferences need to be translated into social preferences ­necessary for making collective decisions and choices. The lack of a fair description of social preferences regarding a specific question presents an important ­challenge when designing laws and policies that could be welfare maximising at the aggregate level. Another factor that is relevant for the parties’ welfare is the definition of social policies and how individuals and groups can influence them. The closer social policies are to the individuals’ preferences, the better their welfare will be. However, the possibility of agents influencing makers of social policy so that they design policies according to their preferences is very different. The parties’ access to political power and to decision-making in order to influence policy-makers in designing policies reflecting their preferences is not uniform. Such access is in itself mediated, determined and conditioned by economic, cultural, historic and legal factors. Such asymmetry represents an additional complexity for defining welfare-maximising policies. It is our view that private law should be self-restrained concerning the macro challenges. Private law should not dare to confront inequalities at the macro level because it is generally ill-placed and ill-equipped to tackle them. Private law should take into account and be aware of the macro challenges but with no intent to fix them, since—perhaps unforeseen—winners and losers will result. For example, a legal rule banning adjusted-rate mortgages may benefit certain low-income families who were forced to take out adjusted-rate mortgages when they had preferred fixed-rate mortgages, but it may hurt parties with low salaries and liquidity constraints, who would prefer adjusted-rate mortgages. Using a Kaldor–Hicks approach,20 if the welfare improvement of the parties benefited by the legal ban was higher than the welfare loss of the parties hurt by the ban, the new equilibrium of the legal ban would be efficient. However, if the welfare improvement was lower than the welfare of the parties hurt by the ban, the legal ban would not be efficient in the Kaldor–Hicks sense. Private law should aim at adopting measures that could eventually—if possible—benefit everyone, but should be aware that the impact of any measure at the macro level will be different depending on the individuals affected by it and a balance of the effects would be necessary in order to conclude whether the measure should be adopted or not. 20 See JR Hicks,’ The Foundations of Welfare Economics’ (1936) 49 Economic Journal 696 and N Kaldor, ‘Welfare Propositions in Economics and Interpersonal Comparisons of Utility’ (1939) 49 Economic Journal 549.

318 

Fernando Gómez and Mireia Artigot

V.  The Economic Relevance of Parties’ Inequality21 and the Role of Private Law to Redress it The inequality (in information, in behavioural and cognitive dimensions, in incentives and in the ability to obtain relief) of parties involved in a transaction matters for efficiency at the micro level.22 Consequently, it seems reasonable to consider corrective measures that could improve the parties’ position while achieving higher efficiency levels in the transaction. Different measures could be adopted in order to achieve more efficient equilibria. A first type of measure potentially would aim at redressing the parties’ unequal positions by curtailing the incentives for opportunism to the party most likely to engage in it. Creating measures to discourage or make more expensive such opportunistic behaviour may deter such behaviour and hence allow for more cooperative outcomes. For example, the quality of goods is often difficult to assess for consumers. It is the manufacturer or the seller who is more likely to have superior information regarding product quality. However, the manufacturer or the seller may have incentives to behave opportunistically by lowering the product quality while maintaining its price. If the consumer was not able to observe and assess the different product quality or was only able to assess certain quality parameters (that were thus maintained) but not the quality parameters that were lowered, its purchasing decision as well as the terms of the transaction—including the product price—would be distorted. In order to avoid incentives generated by the difficulties faced by consumers in obtaining information about product quality, legal rules, eg the European Sales Directive,23 include the consumer’s expectations on product quality as one of the features defining conformity with the contract.24 In this way, the expectations on product quality of the consumer, regardless of their accuracy, are part of the features defining conformity and hence whether the product conforms with the contract or not. By doing that, the European lawmaker diluted the (undesirable) incentives potentially generated by the asymmetry of information on product quality between the consumer and the product manufacturer or the seller. Other types of measures are related to the ones just described. As explained above, one of the parties may be in a better position to behave opportunistically. 21  It should be noted that parties’ equality or inequality does not refer to their homogeneous or heterogeneous nature, size, capabilities or endowments, but refers to their position in negotiating the terms of the exchange with the other party, and in being able to adjust the terms of their negotiation to their preferences. 22  The more accurately informed, rational and symmetrically positioned in the negotiation, the more the parties will be able to adjust the terms of their exchange to their preferences and will be able to maximise the joint surplus created by the transaction. 23  See Directive 1999/44/EC of the European Parliament and of the Council of 25 May 1999 on certain aspects of the sale of consumer goods and associated guarantees, [1999] OJ L171, 12–16. 24  See Art 2(d) of Directive 1999/44/EC.

Private Autonomy, Weak Parties and Private Law

 319

However, there are cases where parties may be in a similar if not the same position to behave opportunistically but their reaction in order to mitigate the consequences, or to sanction the other parties’ opportunistic behaviour, may not be equally easy or equally costly. In such cases, adopting measures of relief by creating mechanisms to improve the ability to react to the other parties’ opportunism or to reduce the costs of reacting may improve the bargaining position between the parties, hence their equality and the equilibrium surplus reached in the transaction. A paradigmatic example of this unequal position of the parties when reacting to the opportunistic behaviour of counterparties and seeking relief is access to justice. It is much easier for a big, financially solvent corporation to access the justice system and seek and obtain legal relief than for a single individual consumer who may not be able to afford the costs of litigating against a producer or seller. For example, in the context of mass torts, even though redress would not only benefit the aggrieved party, the costs of litigating might be significantly higher than the remedy obtained. Hence, a collective action problem arises. Being aware of the financial difficulties individuals might face when intending to seek relief, parties involved in the harm-creating activity might have incentives to take advantage of their favourable position in terms of access to justice and under-invest in precaution. In order to eliminate the incentives for under-investment in precaution and align the parties’ position in litigating, the European legislator has adopted measures to facilitate access to justice for victims of mass torts so that multiple claims are possible.25 As noted above, one of the factors that affects the parties’ equality in a transaction is the information available to them, i.e. the amount and the kind of information disclosed to them. Improving the amount as well as the access to relevant information to the less-informed party becomes crucial for redressing the position of this party in the transaction. Sometimes information may not be easily accessible to parties, parties may exchange irrelevant or noisy information to create distorted expectations to the other party in the transaction, or the context may not provide parties with relevant information. Such limited or biased information may affect equally or unequally both parties in the transaction but could also affect only one of them. In the context of consumer contracts, though, parties are likely to be unequally situated regarding the information on the subject matter and terms of the transaction. The seller is generally in a better position to be well informed about the product and to decide whether, and to what extent, to disclose information to the consumer. In order to encourage information disclosure and to ensure as much as possible an informed decision by consumers, the European lawmaker has introduced information requirements for consumer contracts before

25  See the Commission Recommendation of 11 June 2013 on common principles for injunctive and compensatory collective redress mechanisms in the Member States concerning violations of rights granted under Union Law, [2013] OJ L201, 60–65.

320 

Fernando Gómez and Mireia Artigot

the consumer is bound by a contract.26 Such information refers to many different factors, eg the main characteristics of the goods and services, the identity of the trader or the total price of the goods and services, including taxes. Further, in order to ensure fruitful acquisition and processing of such information, it establishes that the information cannot be disclosed in a haphazard manner, but in a clear and comprehensive way considering the personal and psychological characteristics of the consumer.27 Further measures refer to the cognitive and behavioural biases parties may be afflicted with in the negotiation. Such biases and behavioural shortcomings may affect both parties unequally either because they have different experiences and hence may be able—or not—to learn from their past choices or because they may be better skilled at exploiting the biases or limitations of the other party. Redressing such differences may help improve the parties’ position when adopting decisions.28 Accomplishing this task through legal rules is by no means easy, especially when cognitive and behavioural biases differ widely, as they typically do across groups or populations of similarly situated contract parties. Finally, parties are diverse, having different endowments, experiences, reputations, learning capacities, etc. Adapting standards of conduct required by law will certainly improve the efficiency of the outcome achieved. The adoption and application of equal standards to very diverse contexts will not result in efficient outcomes. In addition to equity-based considerations—uniformity is neither equivalent to equality nor to equity—efficiency is also affected. Whenever possible, diverse parties require diverse standards so that efficiency is enhanced.29 Moreover, parties’ symmetry and equality matters for efficiency of the terms of their exchange. Evidence shows that trust is more likely and mutual retaliation is less likely the more equality there is in the outcome of interactions between agents.30 At the aggregate level, the cross-country correlation between social trust and national income equality is well documented.31 At the same time, decreasing 26  See Art 5 of Directive 2011/83/EU of the European Parliament and of the Council of 25 October 2011 on consumer rights. 27  See recital 34 of Directive 2011/83/EU. 28  C Jolls and Cass R Sunstein, ‘Debiasing through Law’ (2006) 35 Journal of Legal Studies 199, suggesting that the legal system on what they call ‘debiasing law’ should not focus on inducing rational decisions but on addressing the cognitive bias some parties in the transaction—or both—may experience. For example, it is well settled in the literature on product safety and consumers’ perception of risks that consumers tend to underestimate and be optimistic about product risks. In order to improve the consumers’ risk perceptions, Jolls and Sunstein suggest that debiasing through availability heuristic and framing may be more effective in enhancing risk perception than approaching it through trying to improve individual rationality. 29  J Ganuza and F Gómez, ‘Realistic Standards: Optimal Negligence with Limited Liability’ (2008) 37 Journal of Legal Studies 577 and F Gómez and J Ganuza, ‘Law Enforcement, Infringers Heterogeneity and Optimal Legal Standards’, Working Paper, Universitat Pompeu Fabra (2015) 1–26. 30 See E Fehr and Klaus M Schmidt, ‘The Economics of Fairness, Reciprocity and Altruism— Experimental Evidence and New Theories’ in S-C Kolm and JM Ythier (eds), Handbook of the Economics of Giving, Altruism and Reciprocity, vol 1 (Amsterdam, Elsevier, 2006) 639ff. 31  See A Bergh and C Bjornskov, Trust, Welfare States and Income Equality: What Causes What?, Working Paper, IFN Aarhus University (2013).

Private Autonomy, Weak Parties and Private Law

 321

the marginal utility of wealth is the best representation of preferences in most situations, so a redistribution favouring more equal ex post allocations of wealth may also be considered desirable for purely economic reasons (independent of fairness concerns in support of more equal distributions of wealth between parties with systematically different wealth endowments).

VI.  Private Law as an Instrument for Dealing with the Inequality of Parties in Transactions The symmetry or equality of parties contributes to maximising the desirable role of contracts in economic exchanges. The more parties can negotiate so that contracts can reflect and accommodate their preferences, the more contracts will contribute to maximising the parties’ welfare. As explained above, private law can create incentives as well as discourage conducts in ways that may help desirable symmetry between the contracting parties. Private law emerges in theory as a valuable instrument to enhance equality between the parties. However, this task is remarkably challenging given that in order to be effective, it will be necessary to anticipate the parties’ position, their incentives and cognitive biases, among others. Further, other parameters might be more informative about the parties’ position in the transaction. Ex ante information regarding parameters relevant for the parties’ conduct and decision is necessary so that private law can enhance the parties’ position; and often such information is neither available nor accurate. There are many relevant dimensions affecting the parties’ equality and choosing the right focus on remedial measures as well as finding the relevant parameters to define such measures is remarkably important and not obvious at all. In the context of contracting, it is well settled in the literature32 that when parties are in an asymmetric position that affects their bargaining power, the term of their relationship and content of their contract may be adversely affected. Private law may be conceived as an instrument to redress such imbalance in power and strengthen the position of the economically weaker party in the negotiation. However, private law provides for horizontal measures thought necessary ex ante for balancing the parties’ position that have to be applied to the micro context with specificities often not considered ex ante. Further, characterising and defining the different features of abstract situations where it could be possible to infer that there is an imbalance between the parties is not always an easy task. At the same time, in certain contexts, not only concerning private law, but also

32  A Choi and G Triantis, ‘Market Conditions and Contract Design: Variations in Debt Contracting’ (2013) 88 NYU Law Review 51; A Choi and G Triantis, ‘The Effect of Bargaining Power in Contract Design’ (2012) 98 Virginia Law Review 1665.

322 

Fernando Gómez and Mireia Artigot

other parameters, might be relevant for providing information about the parties’ position and their capacity for negotiating. For these reasons, private law might not always be the instrument best placed to address and eventually improve the outcome of the market.

A. The Parties’ Economic Context: Market Structure as a Background Factor Affecting Contract Terms Having a large firm contracting with a consumer or set of consumers should not always by itself raise concerns about the terms and outcome of the transaction. The market structure, ie whether any of the contracting parties has market power either because of monopoly, oligopoly or monopolistic competition, might be a better predictor of how parties will perform in the negotiation. From this perspective, market structure would not only have an impact on the price and quantity of product exchanged in the market but also on contract terms. If the market structure was close to resembling a perfectly competitive market, with a supply fragmented with lots of suppliers of perfectly—or close to being perfectly—substitute goods, it is reasonable to consider that consumers, ie the demand side of the market, will be able to obtain favourable contract terms in the negotiation—even if not individually negotiated. If contract terms of another supplier were better than the ones offered, the consumer could easily shift to another supplier. In contrast, in markets where suppliers had market power, it could be reasonable to expect that the parties’ asymmetry could under certain conditions, especially when there is also asymmetric information, generate unfavourable contract terms for the consumer. Private law should also take into account information relevant for the parties’ position—such as the market structure—resulting from parameters beyond the specific transaction.33 Ex ante design and horizontal application (which typically characterize private law) may, in certain contexts, not be the adequate features and a global view of the parties’ nature as well as of the economic context, might be the necessary approach to adopt. The macro context, as mentioned above, is relevant for the parties’ negotiating capacity. However, private law is not the best instrument for addressing inequalities arising at the macro level such as the market structure. When creating the right incentives, private law should take into account the parties’ context, its effects on their position and hence on their bargaining power. However, the micro level will not, by itself, be able to balance the asymmetric position of the parties at the macro level.

33 See A Stremitzer, ‘Opportunistic Termination’ (2012) 29 Journal of Law, Economics, and Organization 381.

Private Autonomy, Weak Parties and Private Law

 323

B. Inequality Per Se versus the Effects of Inequality for the Terms of the Transaction The parties’ inequality does not necessarily imply an imbalance or asymmetry in their capacity to negotiate and agree on contract terms. Their inequality, because of their size, economic position, market share, and knowledge and information about the market does not automatically and uniformly translate into asymmetries in their bargaining position and hence in their contract provisions. It is not the asymmetric nature of the parties per se what should matter for private law but the impact and effects of the different nature of the contracting parties on the contract terms and incentives, on the access to relief for the weaker party, on the information available to the parties or on the cognitive and behavioural limitations of the parties in the negotiation. In this regard, there are cases where the abstractly considered weaker parties because of their smaller size or lower income, for example, might be the ones in the best position to negotiate upon certain dimensions of the transaction. In cases where, for example, products are built and tailored following the instructions and specifications of the buyers, the party with the best information about the product object of the transaction—even if smaller compared to the other—might be the one in the best position to negotiate the contract terms. So when designing instruments that could enhance the parties’ equality in the transaction, the analysis should not be ex ante, horizontal and using parameters that could be misleading about the nature of the transaction and position of the parties. It is in this transaction-specific context where private law may truly operate as a valid and useful instrument to enhance the parties’ equality in the transaction. Private law, even though ex ante defined and horizontally applied, is able to provide guiding principles and standards that will have to be interpreted in the specific case at hand. Even though not individually designed for each specific case, its application will allow for some room for adaptation and interpretation that will be able to take into account specificities of the transaction that will be relevant and helpful in maximising surplus.

C. A Close look at Inequality: On the Need for a Multidimensional Analysis of Inequality in Transactions Inequalities or asymmetries do not always appear in one dimension and in the same direction in a given transaction. Rules and policies dealing with the parties’ inequality should be aware of this complexity and aim at avoiding being ­inconsistent and counterproductive. Inequalities may be unidimensional or multidimensional. Whenever inequality arises from a single factor, identifying it and addressing its mechanisms will be relatively simple in order to rebalance the parties’ position. But this is not always

324 

Fernando Gómez and Mireia Artigot

the case. Sometimes, different factors produce inequalities, but not running necessarily in the same direction. Thus, financially weaker parties with less access to relief, for example, might have better information than the other party. In a case like this, creating mechanisms for providing relief to the financially weaker party—albeit better informed—might help, but may also exacerbate the informational asymmetry. Another example could be cases where the more informed party may be more behaviourally biased because of inertia or because of structural rigidities that could introduce difficulties in taking certain decisions. Whenever inequalities arise with respect to different dimensions and eventually multiple directions, the analysis, diagnosis and mechanisms to address them are certainly more complicated, because addressing inequality in one dimension might cause inequality in another. A party may have better information on certain dimensions relevant for the contract—contract terms, for example—but not on others—such as the ex post range of actions by the other party. At the same time, the less-informed party might have larger room for opportunism due to hidden actions, wealth constraints or lack of reputational incentives. The optimal redress mechanism for inequality in multidimensional and multidirectional cases does not clearly arise. For example, when analysing the relationship between household risk management—using family income as the ­variable for assessing risk—and choice of mortgages—fixed-rate and adjustablerate mortgages—the consumer’s decision is not uniform across population groups defined by income or education, or across jurisdictions.34 For example, in Europe, a study of fixed-rate mortgages versus adjustable-rate mortgages shows that families with higher incomes, but who are also more heavily indebted, tend to prefer adjustable-rate mortgages over fixed-rate mortgages. In light of the comparison between these two types of mortgage such choice seems counterintuitive given that variable-rate mortgages are considered to be more harmful to consumers than fixed-rate mortgages. In the US, households with higher income and better credit ratings should be expected to show lower default rates. However, their mortgage choices challenge this result. US households with high incomes and good credit ratings choose complex mortgages that enable them to postpone repayment and show higher default rules than under traditional mortgage structures such as fixed-rate mortgages and adjustable-rate mortgages.35 Choosing the adequate level to assess inequalities and designing redress in the mortgage market is not self-evident and is likely to require weighing different factors. In addition to considering the relevant dimensions and directions of the factors affecting inequality, very often it will be important to decide whether to consider relevant features of the individual transaction or agent or just the ‘type’

34  M Ehrmann and MH Ziegelmeyer, ‘Household Risk Management and Actual Mortgage Choice in the Euro Area’, European Central Bank (2014), available at www.ecb.europa.eu/pub/pdf/scpwps/ ecbwp1631.pdf 35  G Amromin, J Huang, C Sialm, and E Zhong, ‘Complex Mortgages’, NBER Working Paper No 17315 (2011) www.nber.org/papers/w17315

Private Autonomy, Weak Parties and Private Law

 325

or ‘category’ of transaction/agent instead. There is a large heterogeneity within groups that are characterised by different incentives, and informational, financial or behavioural traits. Following this line, firms can be large and small, experienced and new, well-established and fly-by-night, opportunistic and disreputable, while consumers may be inexperienced and savvy, wealthy and poor, patient or impatient, cognitively biased or more rational, stable and volatile. This is an issue that we explore in detail in the following subsection.

D. The Importance for Private Law in Choosing an Adequate Level of Information When designing a remedy or policy in private law for addressing parties’ inequality there may be a lot of information available regarding the type of transaction, the parties’ characteristics, their position and the potential parameters that might create incentives for opportunistic behaviour. In principle, the more information the better, since tailoring inequality redress to the characteristics of the transaction will be easier with a larger amount of information. Also, updating the informational basis for making legal decisions seems to improve the process. However, not all information might be deemed relevant by the law. Some of the information could be discarded and, in fact, it is often thrown away for legal decision-making. It is important to ensure that irrelevant or innocuous information for the transaction is discarded while the informative and valuable is kept and used. There may be good reasons to discard certain pieces of information. The (legal and social) values of generality, consistency and predictability of outcomes might suggest not to use information specific to the transaction. Discarding information may also be useful for reducing information costs to decision-makers, as individual transaction information is costly to obtain. At the same time, discarding information might increase the psychological distance of the decision-maker from the individual case in order to preserve the objectivity in evaluations.36 The factors mentioned surely do not play in the same way in all scenarios so it will be necessary for private law, when addressing inequalities, to determine carefully the optimal ‘vertical’ scope of assessment or redress, i.e. whether the individual, small group, or large scale class or category is the right one to determine the outcome, and thus discard all information concerning lower levels of intervention. Thus, if the right level is that of a general class or category, more detailed information about groups or individuals in the category to which the case refers will not be taken into account, since it has been considered—ideally, because it is considered optimal to do so—that the relevant level of decision is that of the general category.

36  T Kogut and I Ritov, ‘The “Identified Victim” Effect: An Identified Group, or Just a Single Individual?’ (2005) 18 Journal of Behavioral Decision Making 157.

326 

Fernando Gómez and Mireia Artigot

Which level should prevail is not at all obvious, and there is no reason to think that the solution is the same for all problems affecting inequalities in trade. Sometimes the best decisions are probably to be taken at the individual level, with all detailed information at disposal. In other cases, it may be, on balance, more advisable to disregard individual information and restrict the decision-maker just to the information pertaining to the category of parties to which the individual contractor belongs, or to the kind of transaction at stake. Perhaps even this level of optimal decision-making happens at a very abstract level, where most information, except the basic essentials of the transaction, is discarded to determine the legal rule or outcome. Once the scope of the relevant scale of information is determined, the law will decide to collect, use and eventually update the information corresponding to the chosen level or scale. The body of information discarded, though, should ideally be the irrelevant one and not the one providing valuable insight into the nature of and the parties to the transaction. Choosing the relevant information, however, is not an easy task and discarding valuable information is not as infrequent as one would expect. The Court of Justice of the European Union provides examples where (potentially at least) valuable information about the case is not considered, while horizontal attributes of a category—eg a consumer as an ideal type—are applied. A good illustration of this kind of approach is Costea v SC Volksbank Romania SA,37 where, interpreting the concept of consumer under Directive 93/13 on unfair terms in consumer contracts, the role, knowledge and context of the specific transaction was not considered but a general horizontal concept of the consumer was applied. In this sense, the ECJ held that ‘the concept of “consumer” … is objective in nature and is distinct from the concrete knowledge the person in question may have, or from the information that person actually has’. Thus, a consumer is to be seen as a weaker contractual party, even if, as in the case at hand, he is a lawyer-consumer with high level of technical knowledge, and the loan contract was secured by a mortgage taken out by his law firm on the building. One could consider that the Court discarded valuable information in this case. Generally, having more information on the specifics at the micro level, such as the level of skills of the parties to the transaction, would allow more precise ‘inequality redress at the micro level’. This information, hence, should not be discarded, unless to do so would seriously undermine values of generality, consistency and predictability. Discarding information is something that private law, and European consumer law, should do with care.

37  Case C-110/14 Horaţiu Ovidiu Costea v SC Volksbank România SA, judgment of 3 September 2015, request for a preliminary ruling under Art 267 TFEU from the Judecătoria Oradea (Romania).

Private Autonomy, Weak Parties and Private Law

 327

VII. Conclusions Economic exchange is an essential element of economies and of the creation of welfare. It seems hard to dispute that, despite traditional legal or economic assumptions or preconceptions, parties involved in transactions are very often not on equal terms. And the more symmetric parties are, the more the transaction terms resemble their preferences and the more parties can generate a larger joint surplus from the transaction. Among the different mechanisms available to improve the parties’ symmetry in trade, private law arises as an instrument to generate better incentives, to improve consumers’ information through mandatory disclosure or other means, to encourage cooperation, and create mechanisms for facilitating redress for parties, enabling them to keep in check the opportunistic behaviour of the other party. But informational problems will remain pervasive despite private law efforts. Additionally, individuals experience cognitive limitations and biases that affect the way they take decisions and the way they can interact in economic transactions. Moreover, the macro dimensions (winners and losers, preference aggregation, political influences) of inequality complicate the matter in a dramatic way. Such dimensions should be taken into account when designing private law instruments. Not because, or so we think, because private law may adequately resolve them— it may not, we believe—but because the full picture needs to be kept in mind in order to grasp the challenges and limitations in the hands of private lawyers.

328 

19 Fairness at a Time of Perplexity: The Civil Law Principle of Fairness in the Court of Justice of the European Union DANIELA CARUSO*

I. Introduction The title of this chapter is loosely modelled on the work of Thomas M Franck, an influential international law scholar preoccupied with the notion of fairness.1 In Franck’s writings, fairness was portrayed as a relative and indeterminate concept.2 At times, however, the author seemed to revert to a foundationalist defence of fairness as a distinctive trait of advanced legal systems, setting them apart from chaotic and despotic regimes.3 While remote from present-day European troubles, Franck’s oscillation captures the spirit of the time in EU legal discourse. On one hand, disenchantment with the EU legal project as an instrument of justice is on the rise.4 Trenchant critiques have been aimed at the very design of ‘integration through law’, which has proven structurally unable to cure problems in the socioeconomic peripheries of the Union,5 has failed at creating true *  Professor of Law and Jean Monnet Chair, Boston University School of Law. Thanks to Marija Bartl, James Fleming, Bianca Gardella-Tedeschi, Steven Garza, Duncan Kennedy, Fernanda Nicola, and Mark Pettit for comments on earlier drafts. Errors are mine. 1  TM Franck, Fairness in International Law and Institutions (Oxford, Oxford University Press, 1998); TM Franck, ‘Epistemology at a Time of Perplexity’ (2002) 13 European Journal of International Law 1025 (defending fairness against cultural relativism). 2 Franck, Fairness (n 1) 14 (‘[F]airness is relative and subjective … a human, subjective, contingent quality’). 3  See E MacDonald, International Law and Ethics after the Critical Challenge: Framing the Legal within the Post-Foundational (Leiden, Martinus Nijhoff, 2011) 220–21 (documenting this shift in Franck’s writings); I Scobbie, ‘Tom Franck’s Fairness’ (2002) 13 European Journal of International Law 909. 4  D Kochenov, G de Búrca and A Williams (eds), Europe’s Justice Deficit? (Oxford, Hart Publishing, 2014). 5  D Kukovec, ‘Law and the Periphery’ (2014) 21 European Law Journal 406.

330 

Daniela Caruso

­post-national cohesion,6 and seems founded on a false or framed mode of democratic deliberation.7 In the particular context of private law, a growing body of literature is casting doubts on the ability of EU-led consumer protection to correct the ­inequities of Europe’s markets, especially in the aftermath of the global financial crisis and in light of the structural dynamics of legal integration.8 On the other hand, jurists seem unable to relinquish faith in fairness, or justice, as a plausible aspiration of the EU legal system—a guiding principle, procedural and discursive if not substantive, that EU law should, and can, uphold because the world would certainly be worse without it.9 The depth of interdependence between Member States, reached through free trade and centralised policies over the past six decades, has disabled a number of domestic justice mechanisms,10 with the result that if no justice can be found on the EU plane, all hope is lost.11 The chasm between faith and scepticism runs deep, not only between various scholarly camps, but also among scholars of similar persuasion, and even within the body of single authors’ contributions.12 The general principle of fairness, recently articulated by the Court of Justice of the European Union (CJEU), is bound to prompt exactly this sort of ambivalence among scholars. Fairness in private law could be dismissed as hopelessly indeterminate: yet another venue of judicial balancing, a technique already seen ad nauseam in Luxembourg, whereby lip service is paid to conflicting considerations, but no real solace can be found against regressive outcomes of law and policy choices.13 At the same time, the judicial articulation of a general principle of fairness in private law could be seen as a prompt for domestic courts to entertain context-sensitive considerations, in such a way as to redress, within the boundaries of judicial discretion, the predicaments of situationally disadvantaged parties. A Rawlsian reading of fairness would point precisely in this direction.14 Particularised, policy-oriented inquiries into the distributive stakes of each dispute, such as the relative impact upon low-income subjects of (in)validating certain contract

6  JHH Weiler, ‘Van Gend en Loos: The Individual as Subject and Object and the Dilemma of European Legitimacy’ (2014) 12 International Journal of Constitutional Law 94, 103. 7  M Bartl, ‘Internal Market Rationality, Private Law and the Direction of the Union: Resuscitating the Market as the Object of the Political’ (2015) 21 European Law Journal 572. 8  See D Caruso, ‘Qu’ils mangent des contrats: Rethinking Justice in EU Contract Law’ in Kochenov et al (n 4) 367 (surveying this literature). 9  See O Gerstenberg, ‘Constitutional Reasoning in Private Law: The Role of the CJEU in Adjudicating Unfair Terms in Consumer Contracts’ (2015) 21 European Law Journal 599 (providing a theoretical framework and a comprehensive analysis of the CJEU’s case law on fairness in B2C relations). 10  A Somek, The Cosmopolitan Constitution (Oxford, Oxford University Press, 2014). 11  J Habermas, The Crisis of the European Union: A Response (Cambridge, Polity, 2013). 12  See Section IX. 13 M Lasser, ‘Fundamentally Flawed: The CJEU’s Jurisprudence on Fundamental Rights and ­Fundamental Freedoms’ (2014) 15 Theoretical Inquiries in Law 229. 14  J Rawls, A Theory of Justice (Cambridge, MA, Harvard University Press, 1971) 76.

Fairness at a Time of Perplexity

 331

terms or (dis)allowing certain remedies, could ultimately raise the standards of substantive justice in the EU.15 These pages, based on the premise that distributive analysis in private law adjudication is desirable, set forth the argument that the CJEU has recently proven capable of identifying distributive trade-offs in the name of fairness, that such uses of judicial discretion do not exceed the scope of the Court’s institutional competence, and that the Court should not retreat from this laudable path. To be sure, at times the Court has also misused the idea of fairness, reducing it to an empty representational device through which facile slogans could unduly displace serious distributive analysis.16 Nevertheless, the Court’s occasional embrace of the principle in substantive terms deserves attention and praise. In the midst of sobering reflections on law’s complicity in perpetuating Europe’s inequalities, it may be important to take stock of narrow, but clear, progress through law.17 As a result of fairness-driven holdings in Luxembourg, state legislators have already faced considerable pressure to control aggressive debt collection practices in austerity-ridden countries.18 Progress also results from the inter-­ institutional dialogue increasingly triggered by the supranational regulation of private autonomy.19 The transmission belt that connects the Troika’s recipes for growth to the Member State governments’ austerity reforms is notoriously rigid and leaves little room for textured accounts of its social consequences. It is in the context of contract law adjudication that, by contrast, such accounts find a stage. The post-crisis anecdotes of private misery that are found in the litigation of contract disputes acquire not just visibility, but also legal, justiciable form. Shoring up such hard-won expressive outlets in the face of countervailing messages is essential. The argument proceeds as follows. Section II outlines the CJEU’s ‘discovery’ of fairness as a general principle of civil law in the ground-breaking case E Friz.20 Section III equates the general principle of fairness with distribution-sensitive adjudication, and illustrates how the CJEU’s balancing of conflicting interests in its preliminary rulings could be regularly informed by lucid forms of d ­ istributive analysis. Section IV posits that, given the conceptual and systemic autonomy of civil law principles, judicial fairness can radiate meaning beyond the confines of business-to-consumer (B2C) disputes and onto broader inequality debates without any undue blurring of canonical partitions (private/constitutional, legal/political).

15 LA Fennell and RH McAdams, ‘The Distributive Deficit in Law and Economics’ (2016) 100 Minnesota Law Review 1051 (providing examples of distribution-sensitive judicial reasoning in contracts and other areas of law). 16  See Section VII. 17  See U Mattei and F Nicola, ‘A “Social Dimension” in European Private Law? The Call for Setting a Progressive Agenda’ (2006) 41 New England Law Review 1. 18  See Section IV. 19 ibid. 20 Case-215/08 E Friz GmbH v von der Heyden [2010] ECR I-2947, para 48.

332 

Daniela Caruso

By way of comparison, Section V identifies several functional equivalents of the general principle of fairness in US contracts adjudication and recounts their rise and fall. Section VI shows how the decline of judicial fairness has led US progressive jurists to pursue the goal of fairer transactions outside of court, and points to the downsides of excising fairness from contracts adjudication. On the basis of this assessment, Section VII affirms the usefulness of a judicial principle of fairness in supranational private law, but it cautions the CJEU against resorting to false or facile dichotomies in the balancing exercise that fairness requires. The chapter concludes by joining an emerging strand in current legal scholarship. This strand remains critical of the shortcomings of the EU legal architecture from the perspective of distributive justice, and yet clings to law—as both process and substance—whenever law can effectively fence off two ongoing phenomena: the deployment of economic dogma as a conversation-stopper in policy-making circles, discussed in Section VIII, and the endless proliferation of ‘unweighted’ narratives in EU discourse, which are too often balanced against one another as if they all had equal importance and legitimacy (Section IX).

II.  Discovering Fairness In recent years, the CJEU has worked on the assumption that a general principle of fairness must guide the adjudication of disputes between private parties. In E Friz, a ground-breaking case decided at the dawn of this decade, the Court held that it is ‘in accordance with the general principles of civil law [to ensure] a satisfactory balance and a fair division of the risks among the various interested parties’.21 The CJEU did so while interpreting a text that, contrary to the directives on Unfair Contract Terms (UCT) and on Unfair Commercial Practices (UCP), did not adopt fairness as a central concept and only referred to it obliquely.22 While arguably in line with the ample set of CJEU cases that embrace a thick notion of consumer protection,23 this was a judicial move of a different order of magnitude. At stake in E Friz was not just the normal practice of providing authoritative readings of secondary legislation when the statutory text calls for fairness in particular contractual contexts. Rather, identifying fairness as a result of general principles of civil law meant adopting it as a plausible interpretive aid whenever applicable laws do not lend the judge precise guidance. 21 

ibid (emphasis added). Directive 85/577/EEC of 20 December 1985 of the European Economic Community to protect the consumer in respect of contracts negotiated away from business premises (referring only once in its preamble to a 1974 preliminary programme of the European Economic Community for a consumer protection and information policy, which asked ‘that appropriate measures be taken to protect ­consumers against unfair commercial practices in respect of doorstep selling’). 23 V Trstenjak and E Beysen, ‘European Consumer Protection Law: Curia Semper Dabit Remedium?’ (2011) 48 Common Market Law Review 95. 22 

Fairness at a Time of Perplexity

 333

The facts and the law of the E Friz case have been thoroughly analysed by other scholars, making a full summary here redundant.24 In brief, in 1991 Mr von der Heyden had received an unsolicited visit by a tax consultant who had convinced him to invest, together with other partners, in the modernisation of decrepit real estate in Berlin. For contracts concluded in this haphazard fashion, the EU doorstep selling directive grants consumers a right to repent, ie to cancel the deal in a period of no less than seven days from due notice of this right.25 If the consumer receives no such notice—as was the case here—the possibility to cancel lasts longer.26 Accordingly, Mr von der Heyden withdrew from the partnership in 2002, after a period of over ten years, hoping to recoup the full value of his investment on restitutionary grounds. The partnership, however, refused to refund Mr von der Heyden, and asked instead that he pay them over €13,000—the negative difference between his original investment and his share of the steep losses suffered by the partnership over the years. This result would be in compliance with German law, and in particular with the judge-made principle of ‘defective partnership’, duly highlighted for the CJEU by Advocate General Trstenjak in her opinion.27 The language used by the Court to endorse the latter result is worth quoting in full: 48 As the Bundesgerichtshof observed in its decision for reference, [the German] rule is intended to ensure, in accordance with the general principles of civil law, a satisfactory balance and a fair division of the risks among the various interested parties. 49 Specifically, first, such a rule offers the consumer cancelling his membership … the opportunity to recover his holding, while taking on a proportion of the risks inherent to any capital investment of the type at issue in the main proceedings. Secondly, it also enables the other partners or third party creditors … not to have to bear the financial consequences of the cancellation of that membership, which moreover occurred following the signature of a contract to which they were not party.28

The Court’s promotion of fairness to the rank of general principle of civil law in the context of E Friz is highly meaningful to the development of EU law. The Court decided this case in 2010—two full years into the financial crisis, at a time in which consumers all over Europe and beyond shared Mr von der Heyden’s desire to walk back from improvident investments.29 The jurists who wrote the E Friz 24  See MW Hesselink, ‘The General Principles of Civil Law: Their Nature, Roles and Legitimacy’ in D Leczykiewicz and S Weatherill (eds), The Involvement of EU Law in Private Law Relationships 131 (Oxford, Hart Publishing, 2013). 25  Directive 85/577/EEC (n 22). 26 Case C-481/99 Heininger v Bayerische Hypo- und Vereinsbank AG [2001] ECR I-09945; Case-412/06 Annelore Hamilton v Volksbank Filder eG [2008] ECR I-02383 (discussing the issue of limitation on the time for repentance). 27  E Friz (n 20). 28 ibid. 29 AG Trstenjak pointed out that ‘Investment in [junk] property, which Germans opted for primarily on account of the resultant tax benefits, has often failed to deliver the expected results, and

334 

Daniela Caruso

judgment were surrounded by news of pervasive financial disasters, and were constantly reminded that the consequences of poor financial market regulation would hit some pockets of the EU population much more heavily than others. This was fertile terrain for the judicial discovery of a principle of fairness30—a principle that had not been named either by the referring court or by the ­Advocate General, and that was invoked sua sponte by the CJEU. Apportioning losses fairly, ie protecting those who were hopelessly stuck with the partnership from the sudden flight of those who could withdraw their membership, seemed to be what justice required. The Court decided accordingly. To be sure, a narrow reading of the case is also possible: the decision could have been merely a pluralist embrace of Germany’s own partnership law, developed by German judges on the basis of German principles. Besides, according to the doorstep selling directive, the restitutionary consequences of withdrawal must be determined by national law, and the CJEU had little choice but to defer to the German definition of such remedies. But the language of the Court goes beyond mere judicial necessity. It is a language of solidarity between those who could flee and those left behind in the quagmire of financial disaster. It is at least plausible that the judges could see the analogy between the uneven distribution of losses inside the E Friz partnership on one hand, and Europe’s larger inequities on the other: the widening of the spreads, the predicament of the PIIGS (Portugal, Italy, Ireland, Greece and Spain) and the plight of the unemployed.

III.  Fairness as Distributive Justice Another reason why the judgment is momentous is that it embraces a substantive— as opposed to procedural or discursive—notion of fairness. The interests of all the parties to the E Friz dispute were not just formally represented in the opinion, but tangibly recognised in the outcome and in the way the loss from poor investments was apportioned between affected investors. Adopting a fairness principle of this sort signals openness to distributive justice in contract law and departs from philosophical or utilitarian approaches that reduce contract exclusively to ‘promise’ or ‘efficiency’.31

investors have therefore looked for ways of terminating those investments by relying inter alia on the Community directives concerning consumer protection. E Friz (n 20) para 3. 30 K Lenaerts and JA Gutiérrez-Fons, ‘The Constitutional Allocation of Powers and General Principles of EU Law’ (2010) 47 Common Market Law Review 1629, 1635 (referring to judicial ‘discovery’ of general principles, which are assumed to pre-exist in the legal system of reference). 31 See A Bagchi, ‘Distributive Justice and Contract’ in G Klass and others (eds), Philosophical ­Foundations of Contract Law 193 (Oxford University Press, 2015) (providing a persuasive account of reasons to adjudicate contract disputes in light of distributive justice, and a full survey of relevant literature).

Fairness at a Time of Perplexity

 335

The question whether distributive considerations really belong in private law adjudication continues to animate legal scholarship.32 Many object to distributive justice as a goal of contract law, either because of adherence to the view— based on legal philosophy—that contracts enforcement should cling to corrective justice only,33 or because of the conviction that distributive justice is more efficiently achieved through taxation and other forms of transfers, as opposed to adjudication of private disputes.34 A complete refutation of these arguments exceeds the scope of this paper, but abundant counter-arguments exist in theoretical literature.35 Foremost among such arguments is the finding that, as a matter of positive law, redistributive motives are already pervasive in the law of contracts, and that many rules and doctrines cannot be understood without resorting to substantive fairness.36 Fairness competes with other values for primacy, but it is obviously in the race. Fairness is regularly balanced against other values in the system, while also nesting within itself an evaluation of relative costs and benefits. In a Rawlsian sense, fairness requires aiding ‘the least advantaged’,37 which in contract law means minimising the negative impact of rules on those who would lose the most due to background unfairness. Whenever an interpretive gap leaves the judge room for manoeuvre, a lucid assessment of circumstances and distributive considerations must precede and guide the adjudicatory process. Flattening the fairness mandate onto a context-indifferent plane, as if all consumers and traders deserved invariant amounts of autonomy and protection, would not promote substantive justice.38 A contextualised and distribution-sensitive reading of the fairness principle in court is a necessary (though not sufficient) antidote to the ongoing reformalisation of private law—a process in which, for the sake of the internal market, the EU legal system creates new average groups and reinterprets justice as market access.39 A paradigmatic example of judicial balancing, based on thoughtful distributive considerations, is to be found in Justice Brennan’s opinion in Penn Central v New York.40 Following denial of a permit to build above the existing station’s 32 

Fennell and McAdams (n 15). EJ Weinrib, The Idea of Private Law (Cambridge, MA, Harvard University Press, 1995). 34  L Kaplow and S Shavell, ‘Why the Legal System Is Less Efficient than the Income Tax in Redistributing Income’ (1994) 23 Journal Legal Studies 667. cf Fennell and McAdams (n 15) (providing a thorough critique of the assumption that taxation is always a preferable strategy for wealth redistribution). 35  Bagchi (n 31). 36 D Kennedy, ‘Distributive and Paternalist Motives in Contract and Tort Law, with Special Reference to Compulsory Terms and Unequal Bargaining Power’ (1982) Maryland Law Review 563. 37 J Rawls, ‘Justice as Fairness: Political not Metaphysical’ (1985) 14 Philosophy and Public Affairs 223. 38  Kukovec (n 5). 39  Bartl (n 7) 586 (explaining that EU private law, in the name of the Internal Market, has ‘creat[ed] new average groups and [stressed] procedural rights rather than substantive protection’, thereby eclipsing substantive justice). 40  Penn Cent Transp Co v City of New York, 438 US 104 (1978). 33 

336 

Daniela Caruso

structure, appellants sought monetary compensation based on the idea that zoning regulation (in particular, the preservation of historic landmarks) had resulted in the taking of their jus aedificandi. Famously, Brennan noted that the City’s preservation of historic landmarks over time was the very reason for the enormous value of Penn Central’s property. In the balancing of property rights against public interest, it became therefore clear that appellants derived great advantage from the very regulatory practice that they deemed harmful, and deserved, as a consequence, no compensation.41 The mode of judicial reasoning adopted in Penn Central may strike some European observers as excessively open-ended, but it is hard to see how the dispute could be seriously resolved without resort to Brennan’s distributive logic. A lucid consideration of benefits and harms is not beyond the realm of judicial functions, but is rather essential to an intelligent adjudication of each case. A distributive analysis of the type performed in E Friz in application of the doorstep selling directive confirms the plausibility of a distribution-sensitive application of fairness in judgment. The consideration of the situational predicaments of all creditors affected by Mr von der Heyden’s withdrawal added to the clarity of the judgment, allowed for context-tailored application of pro-consumer legislation, and avoided one-sided results that might go against the intent of the EU legislator.

IV.  The Expressive Function of Judicial Fairness As noted, the elevation of fairness to the status of substantive principle in the case law of the CJEU has possibly far-reaching implications for the EU legal discourse. In the realm of EU-level consumer disputes, fairness parlance has naturally blended with issues of social justice and economic recovery and with the constitutional protection of fundamental rights. Thematically and structurally, such connections are crucial.42 The regulation of private autonomy at the EU level— the very podium on which fairness has been promoted to the rank of general principle—is an important part of serious reflections on equality, and on the lack of it.43 Recent contributions to the literature on inequality in the Western world 41  ibid 147 (‘Even where the government prohibits a noninjurious use, the Court has ruled that a taking does not take place if the prohibition applies over a broad cross section of land and thereby “secure[s] an average reciprocity of advantage.” Pennsylvania Coal Co v Mahon, 260 US, at 415, 43 SCt at 160. … It is for this reason that zoning does not constitute a “taking.” While zoning at times reduces individual property values, the burden is shared relatively evenly and it is reasonable to conclude that on the whole an individual who is harmed by one aspect of the zoning will be benefited by another.’). 42  See, eg, HW Micklitz and I Domurath (eds), Consumer Debt and Social Exclusion in Europe (Farnham, Ashgate, 2015). 43 JE Stiglitz, The Price of Inequality: How Today’s Divided Society Endangers Our Future (WW ­Norton & Company, 2013); T Piketty, Capital in the 21th Century (Cambridge, MA, Belknap Press, 2014).

Fairness at a Time of Perplexity

 337

have shed light on the fact that returns on capital (Piketty) or on real property (Stiglitz) have been much higher in modern times than economic growth. Such spectacular returns have created an ever-deeper chasm between rich and poor, with dire consequences for republican ideals and for democracy itself. A necessary complement to such macro-economic vehicles of inequality is to be found in contract law, which to this day rewards higher bargaining power, enables largerthan-growth returns for ‘the haves’,44 and exerts a constant upward pressure on the Gini coefficient. ‘Fairness’ is code for all the ways in which such dynamics might be controlled at law, and is therefore an inherent part of contemporary equality discourse in law, politics and beyond. The emergence of general principles of law in the jurisprudence of the CJEU has raised questions of federalism and institutional competence.45 Scholars have debated whether contract law in particular is sufficiently harmonised at the EU level as to express truly common principles and, if so, whether it is proper for the CJEU—as opposed to legislatures or other epistemic communities46—to name and define them.47 Others have ably tackled this larger debate.48 It may therefore suffice here to note that, at least with regard to the principle of fairness in private law, such concerns seem unfounded. In line with acquired interpretive principles of civil law, general principles are judicial tools but need not be judicial creations, and have historically coexisted with utterly formalist conceptions of ­adjudication.49 As elements of closure, they must be derived by induction from positive legal norms. Fairness is by now fully reflected in many corners of the consumer legislative acquis and is definitely ready for judicial consumption. Invoking fairness as a general principle of (EU) civil law is no sign of i­mpermissible activism.50 44  M Galanter, ‘Why the “Haves” Come Out Ahead: Speculations on the Limits of Legal Change’ (1974) 9 Law and Society Review 1. 45  See, eg, S Weatherill, ‘The “Principles of Civil Law” as a Basis for Interpreting the Legislative Acquis’ (2010) 6 European Review of Continental Law 74; AS Hartkamp, ‘The General Principles of EU Law and Private Law’, (2011) 75 RabelsZ 241; Hesselink (n 24); N Reich, General Principles of EU Civil Law (Cambridge, Intersentia, 2014). 46  Such as the bodies responsible for the development of the PECL or the DCFR. See Hesselink, ‘The General Principles of Civil Law: Their Nature, Roles and Legitimacy’ 167 and 175. 47  See M Claes, ‘The European Union, its Member States and their Citizens’ in Leczykiewicz and Weatherill (n 24) 29. 48 See J Basedow, ‘The Court of Justice and Private Law: Vacillations, General Principles and the Architecture of the European Judiciary’ (2010) 18 European Review of Private Law 443, 456; G Alpa, ‘General Principles of Law’ (1994) Annual Survey of International and Comparative Law. 49  The Italian experience with general principles during and after fascism attests to the compatibility between general principles in private law adjudication and formalist understandings of legal interpretation. See Alpa (n 48); S Vogenauer, ‘“General Principles” of Contract Law’ in L Gullifer and S Vogenauer (eds), English and European Perspectives on Contract and Commercial Law: Essays in Honour of Hugh Beale (Oxford, Hart Publishing, 2014) 291, 293–99; E Hondius, ‘Principles and the Law’ (2012) 20 European Review of Private Law 289, 291 (‘The Italian Codice Civile of 1942 is the only ­modern Civil Code to use principles explicitly’). 50  H-W Micklitz, ‘Mohamed Aziz—Sympathetic and Activist, but did the Court Get it Wrong?’ in A Sodersten and JHH Weiler (eds), Where the Court Gets it Wrong (2013) European Constitutional Law Network,www.ecln.net/florence-2013.html.

338 

Daniela Caruso

E Friz attests to the possibility of discursive osmosis across the private–public divide in a way that does not disrupt the relative autonomy of private law. The type of fairness embraced by the E Friz Court speaks to broader, systemic issues of the EU legal order, but does not ignore the distinction between principles of civil law and higher-order principles of European constitutional law.51 It only allows the practice of private law adjudication to radiate meaning and to perform a constitutionally appropriate expressive function. Put simply, the embrace of a particular rhetorical mode in the adjudication of contract disputes may amplify the relevance of certain concerns and enhance their weight beyond the courts, well into the political arena. In better words: The politics of contract technicality … is an ideological and rhetorical, rather than a directly distributive politics. In discussing technical issues, legal scholars make arguments, and these arguments ‘resonate’ with, or are homologous with, or are mutually re-enforcing vis-à-vis arguments in domains conventionally thought to be political rather than merely technical. … If contract law requires businessmen of equal bargaining power to look out for one another, then it is more plausible that public law should require strong groups to look out for weak ones.52

Recent literature has portrayed the dialogue between Spanish courts and the CJEU as a serious challenge to the austerity politics embraced in Madrid, both by means of preliminary references and in the context of contract disputes.53 This portrait, if correct, illustrates the power of principle in private law adjudication. Courts engaged in the discussion of substantive fairness are on one hand only solving atomistic contract disputes; on the other hand, by striving to promote fairness between private parties, their judgments become an important element in their natural epistemic landscape. The CJEU can perform this important expressive role without invoking constitutional precepts or rehashing legislative debates, but simply building on the solid jurisprudential tradition of culling principles from within private law rules. Fully framed by the legislator through rules of private interaction, and then developed organically from within, the principles of private law may hold superior legitimacy exactly because of their merely indirect reference to the rest of the legal and

51  See M Dougan, ‘The Impact of the General Principles of Union Law upon Private Relationships’ in Leczykiewicz and Weatherill (n 24) 71 (firmly restating this distinction). 52 D Kennedy, ‘The Political Stakes in “Merely Technical” Issues of Contract Law’ (2002) 10 European Review of Private Law 7. 53  F Gomez-Pomar and K Lyczkowska, ‘Spanish Courts, the Court of Justice of the European Union, and Consumer Law’ (2014) 4 InDret (noting an increase in preliminary references from Spanish courts concerning the compatibility of Spanish rules with Directive 93/13, and finding that such references ‘[change] the strategic interaction between the Spanish Government and the courts’ with regard to the protection of mortgage debtors); FE de la Rosa, ‘The Treatment of Unfair Terms in the Process of Foreclosures in Spain: Mortgage Enforcement Proceedings in the Aftermath of the ECJ’s “Ruling of the Evicted”’ (2015) 2 ZEuP 366.

Fairness at a Time of Perplexity

 339

political order.54 Rather than being influenced by ideology, they can safely radiate from the inside out. A comparison with the US experience, where fairness as a principle or policy carries diminishing weight in adjudication and is no longer central to judicial culture, seems to recommend exactly this course of judicial action.

V.  Fairness in US Contract Law: Rise and Fall In order to facilitate transatlantic comparison, it is essential to outline and debunk canonical distinctions. The idea that courts should be guided, when adjudicating private disputes, by meta-principles of justice has a particular pedigree in continental Europe—one that is notably lacking in US law. Judge-made law is anathema to the civilistes because the system cannot have holes and must ­contain all answers.55 When judges encounter a gap or ambiguity in positive law, they need to find guidance in principles that are inherent to their legal order, and that manifest themselves as obvious, pervasive themes in all expressions of legitimate authority. General principles aptly ensure the closure and completeness of the system. This assumption of closure is not typical of the common law and is definitely not essential in US contract law. Organic growth and adaptability—staples of the common law—would be impaired if judges were to find solutions exclusively inside the box of written rules and precedents. As a matter of course, US contract law does not use the label of general principles in the way EU judges do.56 So much for clichés. In practice, the need to convey objectivity and restraint in court is as much a necessity in the United States as it is elsewhere.57 The fact that many judicial posts are filled by virtue of election or executive appointment makes it especially important for the bench to dispel any impression of ideological bent. Decisions informed by ‘public policy’—an acceptable criterion of last resort

54 The doctrine of mistake, for instance, is an old device by which certain jurisdictions have since time immemorial provided shelter for the unwary. H Collins, The Law of Contract, 4th edn (Cambridge, Cambridge University Press, 2003) 276. 55  Art 4 of the Code Civil Napoleon (1804) provides that a judge who refuses to give judgment on the pretext that legislation is silent, obscure or insufficient may be prosecuted for denial of justice. The various historical vicissitudes of the several European states have added nuance to this basic concept, but have not eliminated it. The reference to general principles as mechanisms of closure and gap-filling also derives from the philosophy of liberal state codifications in the 19th century—one of complete break with natural law, custom, and residual norms of jus commune—with the result that no solution of a dispute between private parties can find its source beyond the confines of positive codification. 56  But see A Scalia, ‘The Rule of Law as a Law of Rules’ (1989) University of Chicago Law Review 1175 (lamenting the absence of general principles in US constitutional adjudication); E Patterson, ‘Cardozo’s Philosophy of Law’ (1939) 88 University of Pennsylvania Law Review 71 (‘[Benjamin Cardozo’s] search for principles of value behind precedents resembles the agelong search for natural law’). 57  D Kennedy, A Critique of Adjudication [fin de Siècle] (Cambridge, MA, Harvard University Press, 1998) 1–2.

340 

Daniela Caruso

in US contract adjudication58—demonstrate the extent of this judicial angst: in their opinions, judges take pains to list a series of legislative, judicial and executive practices that substantiate the particular policy they are embracing. Only if the whole legal system seems to be pointing in a given direction can the judge adopt a corresponding policy as guiding criterion.59 Findings of public policy, just like general principles in Europe, are cast in terms of judicial restraint.60 Hennigsen v Bloomfield—a famous case decided at the dawn of the 1960s—provides a clear instance of this practice. In its opinion, the Supreme Court of New Jersey noted that the contract signed by the purchaser of a defective automobile might be sufficiently conscionable in a technical, doctrinal sense, and yet it was to be partly voided—ie purged of exculpatory clauses—in light of a general trend towards heightened consumer protection and responsible manufacturing.61 This trend found ample ‘objective’ documentation in the draft Uniform Commercial Code and New Jersey’s Uniform Sale of Goods Act.62 The public policy that enabled judicial voidance of a contract clause was ‘discovered’ in the legal system at large, and avowedly not concocted in or by the court. In such judgments, it is apparent that public policy in US courts is a functional equivalent of the general principle of fairness invoked by the CJEU in the E Friz judgment.63 Further equivalents emerged around the same time. In the late 1960s, following the ground-breaking case of Williams v Walker-Thomas64 and against the background of rising pro-consumer activism,65 the idea of substantive fairness in contract law adjudication came to be embedded in unconscionability analyses. The old equitable remedy grew into a versatile tool for policing unfair contract

58  ‘In weighing a public policy against enforcement of a term, account is taken of … the strength of that policy as manifested by legislation or judicial decisions.’ Restatement (Second) of Contracts § 178(3) (1981). See DA Friedman, ‘Bringing Order to Contracts against Public Policy’ (2012) 39 Florida State University Law Review (noting that ‘public policy defenses that specify a violation of a statute or regulation tend to be twice as successful than those that appeal broadly to public policy’). 59  See, eg, Matter of Baby M, 109 NJ 396, 537 A2d 1227 (1988); Henningsen v Bloomfield Motors, Inc, 32 NJ 358, 161 A2d 69 (1960). 60  See J Lipshaw, ‘Formalism, Scientism, and the M-Word: A Comment on Professor Movsesian’s Under-Theorization Thesis’ (2006–07) 35 Hofstra Law Review 23, 30 (posing the question, ‘[w]hy, if we all acknowledge we are now legal realists, and the law is the product of social, personal and political influences, do we, as lawyers and judges, continue to speak of the law as though it were there to be discovered, à la Langdell or Williston, preexisting but untapped, the work of some unnamed Author?’). 61  See PA Alces, A Theory of Contract Law: Empirical Insights and Moral Psychology (Oxford, Oxford University Press, 2011) 129–33. 62  Henningsen (n 59) 386 (speaking of the Uniform Sales Act); ibid 404 (speaking of the Uniform Commercial Code). 63  I use the concept of functionalism in the loosest possible sense, as an enabler of comparison. See R Michaels, ‘The Functional Method of Comparative Law’ in The Oxford Handbook of Comparative Law (Oxford, Oxford University Press, 2006) 339 (offering an insightful dissection of the concept and positing its lasting usefulness). 64  Williams v Walker-Thomas Furniture Co, 350 F2d 445 (DC Cir 1965). 65  President John F Kennedy, Special Message to the Congress on Protecting Consumer Interest (15 March 1962).

Fairness at a Time of Perplexity

 341

clauses in court, and for a while it allowed judges to redress business overreach in consumer contracts.66 In order to achieve substantive fairness through contract adjudication, courts would also refer to economic and political phenomena that upset the original balance of interests between the parties to a contract. Still into the 1990s, a time of welfare contraction in the US, it was not uncommon for courts to name growing unemployment, and reduced choice for would-be employees, as additional reasons to invalidate restrictive terms in employment contracts; or to highlight the increased need for childcare services, resulting from mothers’ early return to work, as a reason to void exculpatory clauses drafted by childcare providers.67 In contrast to such past trends, the current landscape of contract adjudication is much less inclined to embrace any functional equivalent of a fairness principle in adjudication.68 Today US courts take little advantage of the adjudicatory tools in their arsenal—such as public policy, unconscionability or impracticability—to redress the uneven spread of losses in post-crisis markets. Residual efforts of this kind are visible in some states more than others.69 More often, judges cling to the formalist enforcement of contracts as written.70 Substantive justice can sometimes be achieved nonetheless, eg via rules of formation that simply leave out of the deal its most unfair clauses.71 But there is general agreement, among scholars,

66 A Fleming, ‘The Rise and Fall of Unconscionability as the “Law of the Poor”’ (2014) 102 Georgetown Law Journal 1383. 67  D Caruso, ‘Contract Law and Distribution in the Age of Welfare Reform’ (2007) 49 Arizona Law Review 666. 68 See S Landrum, ‘Much Ado About Nothing? What the Numbers Tell Us About How State Courts Apply the Unconscionability Doctrine to Arbitration Agreements’ (2014) 97 Marquette Law Review 751; P MacMahon, ‘Good Faith and Fair Dealing as an Underenforced Legal Norm’ (2015) 99 Minnesota Law Review 2051. 69  A quick count of the contract cases that address explicitly the issue of ‘unfair terms’ and contractual ‘fairness’ finds West Virginia at the top of the list. The West Virginia Supreme Court of Appeals has indeed actively policed unfair contract terms. See Brown ex rel Brown v Genesis Healthcare Corp, 228 WVa 646, 724 SE2d 250 (2011); Brown v Genesis Healthcare Corp, 229 WVa 382, 729 SE2d 217 (2012) (Brown II) (concerning the validity of pre-dispute arbitration clauses in nursing home contracts). This trend has met with reproach at the US Supreme Court. See Marmet Health Care Ctr, Inc v Brown, 132 SCt 1201, 1203–04 (2012) (‘West Virginia’s prohibition against predispute agreements to arbitrate personal-injury or wrongful-death claims against nursing homes is a categorical rule prohibiting arbitration of a particular type of claim, and that rule is contrary to the terms and coverage of the FAA’). A similar judicial trend is visible in California. See TH Riske, ‘No Exceptions: How the Legitimate Business Justification for Unconscionability Only Further Demonstrates California Courts’ Disdain for Arbitration Agreements’ (2008) Journal of Dispute Resolution 591. 70  See GM Cohen, ‘The Financial Crisis and the Forgotten Law of Contracts’ (2012–13) 87 Tulane Law Review 1 (critically stating that ‘[m]ost discussions to date of possible responses to the financial crisis ignore contract law. To the extent contract law makes an appearance, the assumption is usually that the contracts at issue should and will be strictly enforced, so there is not much more to say. Contract law, however, is not dead. Nor is it impotent; it has just been forgotten’). 71  Klocek v Gateway, Inc, 104 F Supp 2d 1332 (D Kan 2000) (holding that terms—in this case an arbitration clause received with a product––do not become part of the contract unless the non-­merchant buyer expressly agrees to them). But see Hill v Gateway 2000, Inc, 105 F3d 1147 (7th Cir 1997).

342 

Daniela Caruso

that fairness per se carries diminishing judicial weight.72 The US Supreme Court has spoken with particular clarity against voiding ‘unfair’ arbitration clauses in contract disputes.73

VI.  Fairness Out of Court in US Law and Policy Aware of the reluctance of many jurisdictions, and of the US Supreme Court, to rein in freedom of and from contract, pro-consumer activism has largely abandoned court-centred strategies. The task of pursuing fairness goals in US contract law has been embraced more directly via legislation and executive action. In US academic circles, the aftermath of Lehman Brothers has been hailed by pro-consumer jurists as ‘a time of momentous and transformational change in the world of consumer law’.74 The drastic changes have resulted not from a strengthened judicial policy of fairness in contract cases, but rather from the spectacular success of pro-consumer politics in Washington DC. The recent establishment of a new federal agency, the Consumer Financial Protection Bureau (CFPB), has been noted as a remarkable shift from the reality of pre-crisis America, characterised by little concern for vulnerable market participants and by the relative triumph of autonomy in federal adjudication.75 Interestingly, this shift has downplayed soft principles such as fairness and has instead emphasised hard numbers. Establishing the CFPB required the political left (championed by Massachusetts Senator Elizabeth Warren) to ally with Chicago-trained law and economics scholars so as to ground the case for financial regulation on narrow and fully documented examples of market failure.76 Another, more local example of executive mobilisation towards fairness is to be found in California’s Uber saga. Unfair terms imposed by Uber Technologies, Inc on its drivers could not be successfully challenged in court,77 but could be affected

72  J Feinman, Un-Making Law: The Conservative Campaign to Roll Back the Common Law (Boston, MA, Beacon Press, 2005) (generally deploring the implosion of fairness in court). cf O Ben-Shahar, ‘Fixing Unfair Contracts’ (2011) 63 Stanford Law Review 869 (confirming that US courts interfere with unfairness very little, but arguing that they do so for good reason). 73  Rent-A-Car, W, Inc v Jackson, 561 US 63, 130 SCt 2772, 177 L Ed 2d 403 (2010); AT&T Mobility LLC v Concepcion, 131 SCt 1740, 179 L Ed 2d 742 (2011); CompuCredit Corp v Greenwood, 132 SCt 665, 181 L Ed 2d 586 (2012); Am Exp Co v Italian Colors Rest, 133 SCt. 2304, 186 L Ed. 2d 417 (2013). 74  JA Spanogle et al (eds), Consumer Law: Cases and Materials, 4th edn (St Paul, MN, West Academic Publishing, 2013) v (noting the enactment of ‘ground-breaking federal consumer protection legislation … in the wake of the financial crisis’). 75  ibid (including the 1991 preface, which describes a desolate landscape: unconscionability deflated in court and used more by corporations than individuals, a notably inactive FTC, deregulation and federal preemption of state usury laws, and only little solace in the Baby FTC Acts, meant to compensate for FTC inaction). 76  O Bar-Gill and E Warren, ‘Making Credit Safer’ (2008) 157 University of Pennsylvania Law Review 3. 77  O‘Connor v Uber Technologies, Inc, 58 F Supp 3d 989 (ND Cal 2014).

Fairness at a Time of Perplexity

 343

by the California Labor Commission, which in June 2015 held that drivers are employees, not independent contractors, and as such find protection in statutory employment law.78 More generally, an important strand of legal academia recommends turning to consumer activism so as to reverse the business-friendly treatment of unfair boilerplate terms where it most matters, ie in the forum of public opinion.79 Because the idea of judicial fairness has fallen to the margins of pro-consumer discourse, fairness battles must be fought on broader political grounds. Such changes are momentous and need careful evaluation. At a glance, the decline of fairness as a principle in US contracts adjudication might not seem problematic. After all, a rhetorical emphasis on fairness remains compatible with a narrow reading of the concept, leading to the enforcement of many contract clauses of questionable distributive valence.80 Fairness is obviously a malleable concept.81 In the US Congress, fairness arguments have been used, for example, to reduce rather than expand the possibility of consumer class actions.82 Most worryingly, fairness parlance in court can ‘increase the appearance of … relative justice of the status quo’83 and so prevent the gathering of political consensus around the need for reform.84 It might very well be, therefore, that a social justice agenda in US contract law is better served by activist engagement with legislative and executive bodies than by nostalgic revivals of fairness-based adjudication. Much contract law happens anyway outside of the judicial arena, in private fora for dispute resolution as well as in a growing number of self-regulating communities.85 There are, however, significant problems with the ongoing shift away from judicial fairness. First, in sectors in which there is no legislative will to pursue

78 

Berwick v Uber Technologies, Inc, Case No 11-46739 EK (2015). activism in the form of dissent and voice through Internet and other channels is the heart of the solution.’ S Ghosh, ‘Against Contractual Authoritarianism’ (2014) 44 Southwestern Law Review 239. See generally Faircontracts.org, http://faircontracts.org (Ralph Nader’s project). 80  ‘A bargain is not unconscionable merely because the parties to it are unequal in bargaining position, nor even because the inequality results in allocation of risks to the weaker party.’ Troy Mining Corp v Itmann Coal Co, 176 W Va. 599 (1986) (quoting Restatement (Second) of Contracts § 234 cmt d at 111 (10th Draft No 5, 1970)). 81 ‘Perhaps the concept of good faith takes on so many different meanings in different contexts because at base it is nothing more than a requirement of fairness—a definition so broad as to be virtually meaningless.’ R Hillman, ‘Policing Contract Modifications under the UCC: Good Faith and the Doctrine of Economic Duress’ (1979) 64 Iowa Law Review 849, 877. 82 G Calabresi, ‘Class Actions in the US Experience: The Legal Perspective’ in JG Backhaus, A Cassone and GB Ramello (eds), The Law and Economics of Class Actions in Europe: Lessons from America (Cheltenham, Edward Elgar Publishing, 2012) (discussing the Class Action Fairness Act of 2005). 83  Kennedy (n 57) 2. 84  ‘From the perspective of the law school contracts classroom, it might appear that contract law requires people to be fair to one another. But many judge-made rules of contract law that purport to give weak parties protections against strong parties turn out to be illusory in practice.’ ibid 270 (footnote omitted). 85  F Cafaggi, ‘Self-regulation in European Contract Law’ (2007) 1 European Journal of Legal Studies 1. 79  ‘Consumer

344 

Daniela Caruso

r­ edistributive fixes, the relaxation of fairness as a guiding judicial principle leaves weaker parties entirely out in the cold.86 Employment contracts in the low-wage sector, for instance, are unlikely to be legislatively regulated any time soon, yet are replete with unfair terms that could use more judicial policing in many ­jurisdictions.87 Second, in so far as courts remain repositories and propagators of values in the United States, the decline of fairness as a guiding principle in contract law adjudication is an alarming trend.88 Third, as evidenced by the genesis of the CFPB, empiricism in contract law is on the rise. Scholars traditionally concerned with fairness towards consumers and with protection of weaker parties resort to field experiments more and more often, and they feel compelled to support their objections to untrammelled freedom of contract with tangible evidence of systemic unfairness.89 The shift is evidentiary: the type of empiricism now in vogue assumes that the market is efficient and capable of redressing unfair contractual practices through self-healing, competitive market mechanisms. Only when the opposite is proven, not just by a plethora of anecdotes, and not even by bold examples of contractual overreach,90 but by extensive fieldwork,91 can regulation be tolerated at the margins. In this climate, resorting to ‘considerations of equity and fairness’92 in contract law adjudication comes to sound somewhat old-fashioned, and so does the idea of capturing the economic reality of contracts by focusing

86 R Arnow-Richman, ‘Cubewrap Contracts: The Rise of Delayed Term, Standard Form Employment Agreements’ (2007) 49 Arizona Law Review 637. 87  See, generally, V Moffat, ‘Making Non-Competes Unenforceable’ (2012) 54 Arizona Law Review 939. Professor Rachel Arnow-Richman deems it unlikely that a bill recently introduced in the US Senate (the MOVE Act) will make it through the legislative process in the present political climate. E-mail from Rachel Arnow-Richman, Professor at University of Denver Sturm College of Law, to AALS Contracts List Serve (June 6, 2015) (on file with author). 88  J Resnik, ‘Fairness in Numbers: A Comment on AT&T v Concepcion, Wal-Mart v Dukes, and Turner v Roger’ (2011) 125 Harvard Law Review 78 (‘Courts in democratic social orders are … one of several venues in which the content of law is debated, and other branches of government may, in turn, respond); MJ Radin, Boilerplate: The Fine Print, Vanishing Rights, and the Rule of Law (Princeton, Princeton University Press, 2014). 89  See, eg, I Ayres and A Schwartz, ‘The No-Reading Problem in Consumer Contract Law’ (2014) 66 Stanford Law Review 545. 90  J Resnick, ‘Diffusing Disputes: The Public in the Private of Arbitration, the Private in Courts, and the Erasure of Rights’ (2015) 124 Yale Law Journal 2804 (‘We may change any terms, conditions, rates, fees, expenses, or charges regarding your Services at any time.’—Wireless Provider ‘Customer Agreement’). 91 Consumer Financial Protection Bureau, ‘Consumer Financial Protection Bureau Finds that Arbitration Agreements Limit Relief for Consumers’ (1 August 2015), files.consumerfinance. gov/f/201503_cfpb_factsheet_arbitration-study.pdf (finding that few consumers obtain individual relief through arbitration and the courts; therefore, arbitration agreements limiting class actions greatly diminish the possibility for relief.). 92  Jacob & Youngs, Inc v Kent, 129 NE 889, 890–91 (1921). In this case, considerations of fairness famously led Judge Cardozo to interpret an express condition so as to avoid the ‘unjust’ forfeiture of a significant sum of money. Forfeiture would have resulted from a formalist reading of the contract. A recent survey of doctrinal possibilities aimed to redress the practice of unfair terms in standardized contracts is to be found in N Kim, Wrap Contracts: Foundations and Ramifications (Oxford, Oxford University Press, 2013).

Fairness at a Time of Perplexity

 345

exclusively on what happens in court. The residual cohort of scholars involved in promoting considerations of fairness in US courts’ adjudication finds a formidable obstacle in the new empiricist trend.93 The ALI reporters recently tasked with drafting a Restatement of Consumer Law are dedicated empiricists,94 who have often used economic arguments to prove the irrelevance of existing regulation,95 and have been trained in the conviction that pursuing equitable redistribution via judicial fairness is passé.96 In this climate, fairness projects are likely to remain isolated phenomena, rather than part and parcel of contract law.

VII.  Towards a Fair Division of Risks in Post-Crisis EU Adjudication If compared to the tough stance of the US federal judiciary towards the ongoing plight of American consumers, the attitude of the CJEU in matters of contractual fairness looks remarkably emotive and empathetic.97 The CJEU has been guided by a series of statutory provisions (secondary legislation), some clearly aimed at regulating contracts through private law rules (in primis the UCT directive), others designed to prompt administrative control (such as the UCP directive) but no less helpful towards defining what counts as fair between two parties. Its judgments have grown increasingly sensitive to the drama of financial market collapse and of austerity measures. In Aziz, in particular, the Court began to show real concern for the irreversible, traumatic loss of one’s dwelling in the course of swift execution procedures, in a legal landscape that left little room for suspension orders and at a time of sudden decline in the value of mortgaged property.98 Importantly, Aziz also prompted legislative reform in Spain aimed at enhancing the protection of debtors in the real estate market.99 The general principle most used to do justice to

93 

Radin (n 88); Arnow-Richman (n 86). Ben-Shahar, F Marotta-Wurgler and O Bar-Gill (Reporters), ‘Restatement (Third) of Consumer Contracts’, American Law Institute (last accessed 2 November 2015), www.ali.org/projects/ show/consumer-contracts/. 95 See, eg, O Ben-Shahar and CE. Schneider, More than You Wanted to Know: The Failure of Mandated Disclosure (Princeton, Princeton University Press, 2014). 96  S Shavell and L Kaplow, Fairness versus Welfare (Cambridge, MA, Harvard University Press, 2006). 97  Micklitz (n 50). 98 Case C-415/11 Mohamed Aziz v Caixa d’Estalvis de Catalunya [2013] para 61. See also Case C-169/14 Sánchez Morcillo & Abril García v Banco Bilbao Vizcaya Argentaria SA [2014] EU:C:2014:1388, Order of the President of the Court of 5 June 2014, para 11; Case C-34/13 Kušionová v SMART Capital [2014] para 63 (‘The loss of a family home is not only such as to seriously undermine consumer rights but it also places the family of the consumer concerned in a particularly vulnerable position’). Note that, while in Aziz the Court and AG Kokott avoided direct references to housing as a socioeconomic right, such references are quite explicit in the more recent cases. 99  The Spanish reform has already been itself the object of scrutiny in Luxembourg. See C-482, 484, 485 and 487/13. 94 O

346 

Daniela Caruso

consumers turns out to be the EU law principle of effectiveness used as guidance in interpreting the UCT directive, not the civil law principle of fairness invoked in E Friz. However, boosting the effectiveness of fairness control via general principles is arguably the same as reinforcing the principle of fairness itself. The UCT directive has come to life with a force and prominence that would have been hard to anticipate when it was adopted.100 The resulting case law has been amply praised as a fountain of social justice in times of drought, to the point of prompting parallels between today’s CJEU and the Warren Court101—a remarkable makeover for a court charged with Lochnerism in other contexts.102 This line of judgments has also led several scholars to fear backlash effects,103 to question the propriety of the Court’s foray into political ground,104 or at least to argue that the Court went as far as it could.105 Oliver Gerstenberg, in particular, has carefully recounted the many institutional constraints within which the Court must operate when interpreting the UCT directive for the benefit of national courts. First, the CJEU cannot trespass horizontal boundaries, ie it cannot usurp the role of the EU legislator. Second, the CJEU must limit itself to the task of interpretation, as opposed to applying the law to the facts; applying the law to the facts is a task reserved to the national courts, especially when the text of reference is one of minimum harmonisation. Third, the CJEU must respect the internal integrity of Member States’ private laws, and therefore, must not turn all private law disputes into constitutional questions. Fourth, the CJEU must avoid engaging directly with the politics of solidarity, both because it is only a court and because the EU as a system is not endowed with necessary competences in matters of welfare.106 Yet, it can be argued that the Court is not taking full advantage of the space allowed for by the legislative acquis of consumer protection.107 Recent judgments and opinions indicate a less than full embrace of the private law principle of fairness in Luxembourg. This hesitation is uncalled for. There is ample support, in the

100  N Reich and H-W Micklitz, ‘The Court and Sleeping Beauty: The Revival of the Unfair Contract Terms Directive (UCTD)’ (2014) 51 Common Market Law Review 771. 101  Micklitz (n 50). 102  D Nicol, ‘Europe’s “Lochner” Moment’ (2011) 2011 Public Law 308; Christian Joerges, ‘Justice Within and Between Polities’ Verfassungsblog (10 June 2015); D Caruso, ‘Lochner in Europe: A Comment on Keith Whittington’s “Congress Before the Lochner Court”’ (2005) 85 Boston University Law Review 867. 103  Micklitz (n 50). Backlash effects are already visible. See Kušionová (n 98); Case C-482/12 Macinský v Getfin SRO Financreal SRO [2013] (AG opinion). See F Della Negra, ‘The Uncertain Development of the Case Law on Consumer Protection in Mortgage Enforcement Proceedings: Sánchez Morcillo and Kušionová’ (2015) 52 Common Market Law Review 1009, 1031 (concluding that with Kušionová the Court went too far back from Aziz). 104  Micklitz (n 50). 105  Gerstenberg (n 9) 614–20. 106 ibid. 107  ‘Under EU law, the right to accommodation is a fundamental right guaranteed under Article 7 of the Charter that the referring court must take into consideration when implementing Directive 93/13.’ See Kušionová (n 98) para 65.

Fairness at a Time of Perplexity

 347

regulation of private autonomy stemming from EU legislation, for a general principle of fairness in EU private law. The basic idea that weaker parties should not be hoodwinked, that their inaction should not be exploited, and that every time there is room for interpretation the judge should lean in a direction that prevents the enrichment of parties with higher bargaining power, is an idea firmly built in the architecture of EU private law. This solid base should lend ample credibility to fairness as a principle with teeth. The mandate of fairness in consumer transactions took explicit legislative form exactly because of the realisation that freedom of contract, which to this day is the backbone of the internal market, leads normally to imbalanced terms in B2C standard forms, and needs regulatory control precisely towards the goal of enabling true autonomous choices for both sides—traders and consumers.108 Yet the Court seems eager to remind itself and others that fairness control remains an exception, to be interpreted with care and restraint. This seems to be the case especially in the context of Advocate General opinions, which is where deep judicial rationales can surface.109 It is as if Aziz brought the Court to the edge of the judicial cliff, causing panic and retreat. For instance, even in a remarkably pro-consumer opinion, substantially followed by the Court in Kásler, Advocate General Wahl took pains to drop an uncalled-for reminder that freedom of contract is paramount: 104. While fully aware that this question was not referred directly and specifically to the Court, and that it has not, therefore, been discussed by the parties[,] I think that it is important to stress that … the purpose of the court’s intervention must, as far as possible, be simply to reestablish a degree of equality between the sellers or suppliers and the consumers with whom they conclude contracts[.] 105. It must not lead to the upsetting of the contractual balance through the intervention by a State authority after the contract has been concluded. [… A]ny intervention by a third party, including the State in its legislative function, must be viewed with caution in so far as it could potentially jeopardise the freedom of contract and the free competition which are its corollary.110

Consider another passage, intended by Advocate General Wahl to premise his discussion of mortgaged property repossession in Slovakia (Macinský): 60. As the global financial crisis from which Europe and the rest of the world is slowly recovering shows, a well-functioning and healthy credit system is one of the cornerstones of an open market economy. This may lead legislatures to devise particular security and enforcement schemes which are more or less favourable to the business sector in order to support the credit system. The case under consideration raises the question as to whether the procedure at issue goes too far in favouring traders over consumers.111

108 C-280/13 Barclays Bank v Sara Sanchez Garcia and Alejandro Chacon Barrera [2014] EU:C:2014:279, para 32. 109  M Lasser, Judicial Deliberations: A Comparative Analysis of Transparency and Legitimacy (Oxford, Oxford University Press, 2009). 110  Case C-26/13 Árpád Kásler and Hajnalka Káslerné Rábai v OTP Jelzálogbank Zrt [2013]. 111  Macinský (n 103) (emphasis added).

348 

Daniela Caruso

These are worrisome quotes. Framing the analysis in these terms is not in line with the spirit of the UCT directive, its focus on weakness and vulnerability, and its function as a bulwark against the predictably inequitable effects of sheer market forces. The point is not that vulnerabilities among traders should be disregarded— indeed, the practice of distributive analysis requires consideration of such issues in context.112 But it is hard to see what a healthy credit system has to do with cut-throat debt collection practices. There is general consensus across ideological camps that the global financial crisis did not result from the suppression of private autonomy, but was rather due to excessively permissive regimes. Monitoring the overreach of business is not a hindrance, but rather a condition towards achieving a healthy credit system. The balancing opportunity provided by the fairness test is, in the Advocate General’s opinion, distorted by vulgarized versions of laissez-faire economics. The weighing is flawed, and the distributive stakes are therefore deeply misrepresented. The Advocate General had of course plausible reasons to deem the reference in this case inadmissible (the debtors’ claim, once meritorious, was arguably moot by the time the reference reached Luxembourg), but he would also have had ample opportunity to empathise, if not with the defaulting debtors, at least with the referring court. Instead, the referring court was left wondering whether anything in EU law might point to the injustice of Slovakia’s execution proceedings.113 The Advocate General chose to chide the Slovak court for its lack of clarity, knowledge and organisation, a choice of tone presumably meant to stem the inflow of references in Luxembourg and to explain that efficient debt collection is essential to the economic recovery of Europe. Even bracketing these expressions as dicta and focusing only on instances of empathetic engagement with post-crisis predicaments, it remains unclear whether the contract law emerging in Luxembourg is fully responsive to a principle of fairness—even within an area where the EU legislator has asked the CJEU to set a floor of minimum substantive fairness across the Union. Gerstenberg’s analysis shows that, in a deliberative, procedural sense, the Court’s case law has adequately served the function of representing multiple interests, prompting detailed justifications of outcomes, and learning from a plurality of national experiences. This is, admittedly, one version of fairness. But recall the core question asked by the Spanish court in Aziz: ‘[H]ow is disproportion to be interpreted?’114 Answering this question implies distributive considerations of the type famously embraced

112  See Kukovec (n 5), and more generally Fennell and McAdams (n 15) (illustrating counterintuitive distributive effects of legal rules). 113  Mr and Ms Macinský—both retired—had borrowed €5,000 in April 2011 to face the demands of other creditors. Their lender had quickly transferred the credit to a professional recovery entity, which in turn had charged extravagant default fees (bringing the couple’s debt to over €21,000 by October of the same year) and swiftly called a public auction sale of the Macinskýs’ mortgaged home. Macinský (n 103). 114  Aziz (n 98).

Fairness at a Time of Perplexity

 349

by Justice Brennan in Penn Central. Consumer–trader relations can look very different in the various corners of the Union.115 In its preliminary rulings, the CJEU should guide national judges to unpack the distributive complexities of each dispute, rather than resorting to empty juxtapositions.

VIII.  Law and the Dominance of Economic Dogma The foregoing pages have built upon the assumption that the CJEU has correctly ‘discovered’ a general principle of fairness in the relatively self-contained realm of national private laws and in light of supranational enactments. On this basis, the previous sections have concluded that context-sensitive fairness towards consumers is a welcome proposition, fully ripe for judicial enforcement. Two corollaries accompany this conclusion. First, in speaking against crass abuses of consumer vulnerability, as it did in Aziz and could have done in Macinský, the CJEU duly embraces the idea that fairness and autonomy are fully intertwined.116 The private law notion that emerges from fairness-driven adjudication is distinctively thick, articulate and irreducible.117 Second, the fact that a strikingly thin notion of autonomy continues to surface in judicial opinions goes against the grain of specific legislative choices and against positive law in the nation states. Such representations of ‘disembedded’ autonomy in judicial discourse should not be condoned or praised as expressions of values in deliberative processes and balancing exercises, especially when they may lead to justify aggressive debt-collection and repossession practices. The resilience of out-of-context references to naked autonomy is a disquieting phenomenon that deserves elaboration.118 The EU’s intervention in contract law has brought about an enormous expansion of private autonomy, both across borders (via the judicial activation of the four freedoms) and inside domestic markets (via the reduction of state control over a plethora of services and economic activities);119 on the other hand, and precisely towards the goal of increasing the

115 

Kukovec (n 5). See J Singer, No Freedom without Regulation: The Hidden Lesson of the Subprime Crisis (Boston, MA, Yale University Press, 2015). 117  H Dagan, ‘Autonomy, Pluralism, and Contract Law Theory’ (2013) 76 Law and Contemporary Problems 19; H Dagan and M Heller, ‘Freedom of Contracts’, Columbia Law and Economics Working Paper No 458 (2013). 118  C Crouch, The Strange Non-Death of Neo-Liberalism (Cambridge, Polity Press, 2011); V Schmidt and M Thatcher (eds), Resilient Liberalism in Europe’s Political Economy (Cambridge, Cambridge University Press, 2013). 119 G Comparato, ‘Private Autonomy and Regulation in the EU Case-Law’ in H-W Micklitz, Y Svetiev and G Comparato (eds), ‘European Regulatory Private Law—The Paradigms Tested’, EUI Working Paper Law (2014). 116 

350 

Daniela Caruso

volume of transactions, the EU’s legislator has enacted new common rules.120 Due to curious optical distortions, especially in the wake of recent financial market reforms, EU-level regulation makes a larger impression than autonomyboosting interventions. As a result, with an ironic flip of Scharpf ’s asymmetry paradigm,121 domestic systems come to be romanticised as cradles of autonomy, while the supranational level is identified with regulatory constraint.122 The truth, of course, is otherwise. If contractual autonomy in national private law was ever free of regulatory laches, such halcyon days were certainly over before the inception of the internal market project.123 Autonomy has long been welded to other systemic goals in national systems. In the wake of the 1985 White Paper, many private law scholars considered the harmonisation of private law a carrier of regulatory dilution—not expansion—and mobilised to contain the regressive distributive effects of such developments.124 Fairness won its role of general principle by killing a dragon or two in the EU political and legal arena, and should not be knocked out of place by callous references to tabloid economics. This chapters’s insistence on B2C fairness as positive law with a spine, if not with teeth, is in line with a growing trend towards asserting ‘hard law’ in private disputes and in other facets of post-crisis Europe. Champions of this trend include scholars with ample training in legal realism, fully conscious of law’s pliable nature and definitely uneasy with doctrinal rigidity. The new emphasis on the rule of law as a bulwark of civility in these scholars’ writings is intended to protest the superimposition of an economic paradigm of necessity over legal frameworks of welfare, constitutional arrangements and international obligations. Following CJEU and German Federal Constitutional Court judgments upholding legal measures aimed at saving the euro, scholars have tackled the functionalist relaxation of interpretive canons and lamented the subjugation of law to economic dogma.125 Along the same lines, others have denounced the sheer illegality of measures brought about by austerity politics and decried, for instance, the many unchallenged breaches of human rights caused by the Troika-driven policies of Greek austerity.126 120  ibid (‘[P]rivate autonomy gets on the one hand extended, as regulation is meant to construct and enlarge a market in which all operators have an equal access on a non-discriminatory base, but on the other hand gets subjected to a series of measures which in a traditional private law perspective are qualified as limitations of it’). 121  FW Scharpf, Governing in Europe: Effective and Democratic? (Oxford, Oxford University Press, 1999) ch 2. 122 OO Cherednychenko, ‘Contract Governance in the EU: Conceptualising the Relationship between Investor Protection Regulation and Private Law’ (2015) 21 European Law Journal 500. 123  F Wieacker, A History of Private Law in Europe: with Particular Reference to Germany (Oxford, Oxford University Press, 1996). 124  ‘Social Justice in European Contract Law: A Manifesto’ (2004) 10 European Law Journal. 125  ‘Law must adapt, but must also be constrained by legal reasoning exactly in order to distinguish it from mere political decision-making.’ M Everson, ‘An Exercise in Legal Honesty: Rewriting the Court of Justice and the Bundesverfassungsgericht’ (2015) 21 European Law Journal 474, 478. 126  ‘The way the crises have been governed challenges not only the substantive content of human rights law constitutionally, supranationally and internationally, but notably the very idea that there should be a clear line of responsibility when it comes to the protection of human rights.’ M Salomon, ‘Of Austerity, Human Rights and International Institutions’ (2015) 21 European Law Journal 521.

Fairness at a Time of Perplexity

 351

These critiques resonate loudly in the context of this essay. Advocate General Wahl’s reminder that a ‘healthy credit system is one of the cornerstones of an open market economy’ relies, worryingly, on the rhetorical appeal of economic necessity: the paramount health of the credit system is invoked to rein in the reach of fairness—a principle of law—in the context of aggressive lending practices. Against such judicial stances, and in line with the just described scholarly trend, this chapter has attempted to highlight fairness as the firm line that it should be in accordance with the explicit mandate of secondary EU legislation.127 In myriad ways, the private law of the EU has been and should be subject to critique, both internal, as it is riddled with contradictions, and external, given its predetermined agenda,128 its uneven impact on different regions of Europe,129 and its overly narrow focus.130 But the kernels of substantive justice that it has yielded thus far should not be crushed.

IX.  Fairness and the Perils of Deliberation This chapter partakes of a post-crisis movement, led by a growing circle of scholars, which clings to law as a bulwark against the rhetoric of economic necessity. But it also calls into question another worrisome feature of EU law, namely the mind-boggling multiplication of ‘non-weighted’ narratives in EU discourse, which come to be balanced as if they all held equal importance. In the context of the eurozone crisis, the tale of virtuous Swabian wives, only spending what is already in the pot, has been artfully juxtaposed to alarming reports of southern tragedy, with the rhetorical effect of levelling the two perspectives as if they had an equal claim to distributive justice and as if the difference principle had never entered the universe of political concepts.131 The private law quest for fair results in contracts disputes risks being derailed by similar juxtapositions: evicted consumers on one hand, and a credit system in need of efficiency on the other. The fact that in the weighing of such conflicting needs the EU legislator has been placing emphasis on the protection of situationally weaker parties risks getting lost in the balancing game.

127  See H Collins, The European Civil Code: The Way Forward (Cambridge, Cambridge University Press, 2008); See also U Mattei, ‘Hard Code Now!’ (2002) 2 Global Jurist Frontiers 215 (expressing disenchantment with ‘soft’ principles of civil law). 128  M Bartl, ‘The Way We Do Europe: Subsidiarity and the Substantive Democratic Deficit’ (2015) 20 European Law Journal 23. 129  Kukovec (n 5). 130  Caruso (n 8). 131  D Nicols, ‘Swabian Wives, Suffering Southerners: The Contestability of Justice as exemplified by the Eurozone Crisis’ in Kochenov et al (n 4) 165.

352 

Daniela Caruso

The excessive use of such flat juxtapositions may be attributed to two trends. First, proportionality has fully entered the realm of private law adjudication—a practice by now amply analysed in the literature;132 second, a heartfelt belief in the power of judicial deliberation has gained currency in private law circles. With ­specific regard to the reading of the UCT directive in Luxembourg, ­Gerstenberg portrays the Court as laudably engaged in the desirable process of perfecting, through dialogue with national courts, the meaning of fairness in contract disputes: [The Court’s] experimentalism envisages a circular—recursive and mutually transformative—relationship between general interpretive frameworks and their contextualising application: national courts apply the ‘general criteria’ to a particular term in the circumstances of a case, but through their references to the CJEU for preliminary rulings also, where appropriate, invite revision of those general criteria in the light of their consequences. Harmonisation occurs … through always provisional and rolling endorsements, ultimately by the citizens themselves, of emergent paradigm instances of improper contractual terms.133

This passage clearly echoes the vast body of literature that falls under the headings of conflicts law, justification, and proceduralisation.134 Developed over the course of the past two decades on the basis of authoritative philosophical foundations, this body of scholarship posits that the added value of EU law lies in its procedural coordination of conflicting views. EU law provides a system that will tolerate and even foster pluralism in so far as dissonant voices are able to produce justifications for their arguments. When applied to the realm of private law, this view has led to the idea that the Court could continue the process of harmonisation beyond the point of legislative consensus. The Court may develop organically, by means of examples, the meaning of intrinsically complex principles such as fairness. In this view, the very design of the CJEU may fuel this process. Because the CJEU contains legal actors from different legal traditions, it inherently forms an agglomerate of divergent worldviews.135 The variety of understandings of fairness brought to the bench by such different actors as Advocates General Trstenjak, Kokott and Wahl is apt to yield precisely this type of organic accrual. It is worth noting, however, that hints of perplexity have emerged from within this school of thought. Christian Joerges, after sobering reflections on law in a time of crisis, has more recently conceded that the range of justifications allowed for by the Court is narrow and may actually shut down important voices and

132  D Kennedy, ‘A Transnational Genealogy of Proportionality in Private Law’ in R Brownsword et al (eds), The Foundations of European Private Law (Oxford, Hart Publishing, 2011) 185. 133  Gerstenberg (n 9) 620. 134  J Neyer, ‘Justice and the Right to Justification: Conceptual reflections’ in Kochenov et al (n 4) 211. 135 A Cohen and A Vauchez, ‘Introduction: Law, Lawyers, and Transnational Politics in the Production of Europe’ (2007) 32 Law and Social Inquiry 1.

Fairness at a Time of Perplexity

 353

values.136 To this type of worry one should add, in light of Macinský, the fear that under the guise of pluralism the Court might facilitate the unravelling of hard rules and the unmaking of clear legislative progress. Through Macinský, it is easy to see how the formula of value pluralism can become a victim of its own success, rehashing balancing acts that should have been confined to the legislative arena, and drowning substantive fairness in a sea of justifications.

136 

Joerges (n 102).

354 

20 Discrimination on Grounds of Obesity NIILO JÄÄSKINEN

I. Introduction Obesity has become a major public health problem not only in industrialised countries but also in many developing countries. This is mainly explained by changes in nutrition leading to increased energy intake and decrease in energy consumption due to changes in work and everyday life requiring less motion and physical effort. Moreover, several medical, psychological, social and cultural factors have contributed to this development.1 In fact, in many countries such as the United Kingdom and the United States a majority of the adult population is either overweight or obese. Obesity is usually measured with reference to body mass index (BMI) which is the result of a person’s weight expressed in kilograms divided by square of his/her height in metres (kg/m2). The WHO ranks obesity into three classes by reference to the BMI. Persons with a BMI of 30.00 to 34.99 are Obese class I, persons with a BMI of 35.00 to 39.99 are Obese class II, and persons with a BMI in excess of 40.00 are Obese class III,2 which is sometimes referred to as severe, extreme or morbid obesity. This so-called obesity epidemic3 is not only a challenge to public health policies but also a source of new legal problems. In fact discrimination on grounds of weight or obesity has recently provoked litigation in many jurisdictions around the globe and has become an important problem for labour law or non-­discrimination law

1  For example, according to one study, ‘[w]eight gain is associated with the use of many psychotropic medications, including antidepressants, mood stabilizers, antipsychotic drugs, and may have serious long term consequences’. Ruetsch et al, ‘Psychotropic Drugs Induced Weight Gain: A Review of the Literature Concerning Epidemiological Data, Mechanisms and Management’ [2005] Encéphale, abstract, cited on www.ncbi.nlm.nih.gov/pubmed/16389718. 2  See http://apps.who.int/bmi/index.jsp?introPage=intro_3.html and www.cdc.gov/mmwr/preview/ mmwrhtml/mm5917a9.htm. 3  See, for example, the relevant WHO wep page at www.who.int/nutrition/topics/obesity/en/.

356 

Niilo Jääskinen

in general.4 Concerning EU law, in Fag og Arbejde FOA, acting on behalf of Karsten Kaltoft v The Municipality of Billund,5 the Court of Justice of the EU (CJEU) was asked to rule, for the first time, on the extent to which discrimination that was alleged to have been based on obesity was precluded by EU equal treatment law.

II.  Main Litigation and the Preliminary Questions Mr Karsten Kaltoft, who was represented by his union, the FOA (Fag og Arbejde), in the litigation and at the CJEU, had been employed since 1996 as a childminder by the Municipality of Billund. Childminders are hired to take care of other people’s children in their own homes. Mr Kaltoft had been obese for the whole of his employment. In 2007 his BMI was 54, representing extreme obesity. Due to Mr Kaltoft’s obesity, as part of its health policy the Municipality of Billund provided financial assistance for the period from January 2008 to January 2009 in order for him to attend fitness and physical training sessions. During the course of his employment, Mr Katloft had been referred by medical professionals for a gastric operation that was aimed at reducing the volume of his stomach. The operation, however, could not be completed because of an acute medical incident that arose during it. Mr Kaltoft was dismissed in November 2010. The dismissal took place following an official hearing process applicable to dismissal of public-sector employees. At a meeting held during that process Mr Kaltoft’s obesity was discussed but the reason given in the written notice of dismissal was that the dismissal had been decided on ‘following a specific assessment on the basis of a decline in the number of children’. Obesity was not mentioned, and nor were any reasons given as to why precisely Mr Kaltoft, of the several childminders employed by the Municipality of Billund, was the only one selected for dismissal.6 Mr Kaltoft submitted that he had been unlawfully discriminated against because of his obesity. He instituted proceedings before a Danish court, which sent a preliminary reference to the CJEU. It essentially asked two questions. First, is there a general principle of EU law prohibiting all forms of discrimination in the labour market, which would necessarily have encapsulated obesity? This question was followed by two questions to be answered in the event that the CJEU gave a

4 See generally ML Huggins, ‘Not “Fit” for “Hire”: the United States and France on Weight Discrimination in Employment’ (2015) 38 Fordham International Law Journal 889 and I Solanke in http://eutopialaw.com/2014/12/22/kaltoft-a-step-in-the-wrong-direction-towards-protection-fromweight-discrimination-under-eu-law/. 5  Case C-354/13 Fag og Arbejde FOA, acting on behalf of Karsten Kaltoft v The Municipality of Billund, EU:C:2014:2463. 6  Contrary to some reports in the British media it was clear from the hearing held at the Court of Justice that the Municipality of Billund was not alleging that Mr Kaltoft was failing to fulfil his ­contractual obligations.

Discrimination on Grounds of Obesity

 357

­ ositive answer to the first question. However, because the CJEU answered the first p ­question in the negative, the follow-up questions were left unanswered. Secondly, does Directive 2000/78 exclude discrimination on grounds of obesity?

III.  Is There a General Principle of EU Law Prohibiting Discrimination on Grounds of Obesity in the Labour Market? With regard to this question, it is important to recall that the EU is a system of attributed competences. This is clear from Article 5 TEU. The EU can only act to the extent to which it has an express power to act. The reach of general principles of EU law is equally hemmed in by this fundamental rule. The Amsterdam Treaty, which entered into force on 1 May 1999, added to the EC Treaty Article 13 concerning EU legislative powers to combat discrimination on several grounds enumerated in the provision. Prior to this change, the EC Treaty addressed discrimination against individuals on the grounds of nationality and in the context of equal pay for equal work. Article 13 EC, which later became Article 19(1) TFEU, has a much wider scope. It covers discrimination on grounds of sex, racial and ethnic origin, religion and belief, disability, age, and sexual orientation. Yet this list is not open-ended but confers EU legislative powers only with regard to the grounds of discrimination expressly mentioned in Article 19 TFEU. Council Directive 2000/78/EC establishing a general framework for equal treatment in employment and occupation was adopted on the basis of Article 13 EC.7 Pursuant to Article 1 of the Directive: ‘The purpose of this Directive is to lay down a general framework for combating discrimination on the grounds of religion or belief, disability, age or sexual orientation as regards employment and occupation, with a view to putting into effect in the Member States the principle of equal treatment.’ The scope of Directive 2000/78 cannot, through reference to the general EU law principle of non-discrimination, be extended by analogy beyond the grounds of discrimination listed in Article 1 thereof.8 However, the Directive only encapsulates discrimination in employment and occupation, and does not protect the disabled and others falling within Article 1 from discrimination arising in broader contexts, such as club membership, or market participation as a buyer or seller, or any other non-employment-related field. Aside from the rules precluding direct or indirect discrimination, Article 59 7 

[2000] OJ L303/16. Case C-13/05 Chacón Navas, EU:C:2006:456, para 56. Therefore sickness as such is not a ground for discrimination that is prohibited by Directive 2000/78, para 57. 9  Art 5 states: ‘In order to guarantee compliance with the principle of equal treatment in relation to persons with disabilities, reasonable accommodation shall be provided. This means that employers shall take appropriate measures, where needed in a particular case, to enable a person with a d ­ isability 8 

358 

Niilo Jääskinen

of the Directive imposes on employers an obligation to provide reasonable accommodation to disabled people. Moreover, Article 10 TFEU, states that: ‘In defining and implementing its policies and activities, the Union shall aim to combat discrimination based on … disability.’ Article 21 of the Charter of Fundamental Rights of the European Union (‘the Charter’) prohibits ‘[a]ny discrimination based on any ground such as … disability’, and Article 26 of the Charter states that the ‘Union recognises and respects the right of persons with disabilities to benefit from measures designed to ensure their independence, social and occupational integration and participation in the life of the community’. Obesity is not mentioned as a prohibited ground of discrimination in any provision of the Treaty, the Charter or the Directive itself. However, Article 21 of the Charter an open-ended provision in that it prohibits ‘discrimination based on any ground such as …’ (emphasis added). Therefore, on the basis of the wording of this provision, taken on its own, it might be argued that there is a general principle of non-discrimination in EU law covering grounds not explicitly mentioned in Article 21 of the Charter. Examples of such prohibited grounds of discrimination might lie in physiological conditions such as appearance or size, psychological characteristics such as temperament or character, or social factors such as class or status.10 However, this line of argument is blocked by the rules that are relevant to the material scope of the EU Charter. It is important to bear in mind that Article 6(1) TEU precludes recourse to the Charter to extend ‘in any way the competences of the European Union as defined in the Treaties’,11 while Article 51(2) of the Charter has been interpreted in a similar way.12 These provisions set out the outer-boundary of EU fundamental rights law. According to the explanations to the Charter, Article 21(1) thereof ‘does not alter the extent of powers granted under Article 19 nor the interpretation given to that Article’.13 Under the case law of the CJEU to date it had already been established that the fact that discrimination occurs in a substantive field such as the labour market is an insufficient foundation for concluding that a Member State is ‘implementing’ EU law in the sense of Article 51(2) of the Charter. Indeed, the Court has held that, while EU fundamental rights law encompasses the general principle of non-discrimination, and binds the Member States where the national situation at issue falls within the scope of EU law, ‘it does not follow from this that the scope

to have access to, participate in, or advance in employment, or to undergo training, unless such ­measures would impose a disproportionate burden on the employer. This burden shall not be disproportionate when it is sufficiently remedied by measures existing within the framework of the disability policy of the Member State concerned.’ 10 

See opinion of AG Jääskinen, EU:C:2014:2106, 17. Case C-617/10 Åkerberg Fransson, EU:C:2013:105, para 23 (my emphasis). 12  ibid. See also Case C-370/12 Pringle, EU:C:2012:756, para 179. 13  [2007] OJ C303/17, explanation concerning Art 21. 11 

Discrimination on Grounds of Obesity

 359

of Directive 2000/78 should be extended by analogy beyond the discrimination based on the grounds listed exhaustively in Article 1 thereof ’.14 In its judgment the CJEU concluded that the case file provided to the Court contained nothing to suggest that the situation at issue in the main proceedings, in so far as it related to a dismissal purportedly based on obesity as such, would fall within the scope of EU law.15 Neither EU secondary legislation in general nor Directive 2000/78 lay down a general principle of non-discrimination on grounds of obesity as regards employment and occupation. Consequently, obesity cannot as such be regarded as a ground in addition to those in relation to which Directive 2000/78 prohibits discrimination. Therefore the CJEU’s answer to the first preliminary question was that: ‘EU law must be interpreted as not laying down a general principle of non-discrimination on grounds of obesity as such as regards employment and occupation.’

IV.  Directive 2000/78 and Obesity as Disability Directive 2000/78 does not define disability. An autonomous and uniform interpretation of ‘disability’ has been developed in the CJEU case law. Further, it is important to note that since November 2009, the date on which the EU acceded to the United Nations Convention on the Rights of Persons with Disabilities, the EU meaning of ‘disability’ has to be interpreted in line with the Convention.16 The notion of ‘disability’ for the purposes of Directive 2000/78 has six elements. It must be understood as referring to limitations that result, in particular, from (i) long-term (ii) physical, mental or psychological impairments (iii) which in interaction with various barriers (iv) may hinder (v) the full and effective participation of the person in professional life (vi) on an equal basis with other workers. The CJEU found that the concept of ‘disability’ must be understood as referring not only to the impossibility of exercising a professional activity, but also to a hindrance to the exercise of such an activity.17 Any other interpretation would

14  Case C-13/05 Chacón Navas, EU:C:2006:456, para 56. See also, for example, Case C-5/12 Betriu Montull, EU:C:2013:571, where at para 73 the Court held that, at the time of the facts of the proceedings, there was no prohibition in the EC Treaty or in any other provision of EU law of discrimination between the adoptive father and the biological father in relation to maternity leave. At para 72 the Court held that the situation before it did not fall within the scope of EU law. 15  Paras 35–39. 16 See para 53 of the judgment and Case C-335/11 HK Danmark, EU:C:2013:222, para 32; Case C-363/12 Z, EU:C:2014:159, para 75. This was confirmed recently in Case C-356/12 Glatzel, EU:C:2014:350, para 70, even though the Court reaffirmed in Glatzel at para 69 that the provisions of the UN Convention ‘do not constitute, from the point of view of their content, unconditional and ­sufficiently precise conditions which allow a review of the validity of the measure of EU law in the light of the provisions of that convention’, citing Case C-363/12 Z, EU:C:2014:159, paras 89 and 90. 17  Para 54.

360 

Niilo Jääskinen

be incompatible with the objective of that directive, which aims in particular to enable a person with a disability to have access to or participate in employment.18 I took in my Opinion19 the position that it is sufficient that a long-term condition causes limitations in full and effective participation in professional life in general on equal terms with persons not having that condition. It was necessary to make this point because the Municipality of Billund, Denmark, and the Commission relied on the successful execution by Mr Kaltoft of the job of childminder over the course of 15 years. It was suggested that this precluded him from relying on the Directive. According to the case law, a long-term illness may also under certain conditions fall within the concept of ‘disability’ even if sicknesses as such are not disabilities. Hence, while Directive 2000/78 does not protect workers by prohibiting discrimination on grounds of disability as soon as they develop any type of ­sickness,20 it is established in the case law that ‘if a curable or incurable illness’ entails a limitation corresponding to the definition above, such an illness can be covered by the concept of ‘disability’ within the meaning of Directive 2000/78,21 if medically diagnosed,22 and if the limitation is a long-term one.23 Indeed, Mr Kaltoft argued before the CJEU that the WHO considers obesity to be a chronic and durable illness and that obesity has been considered to be a disability under US law.24 WHO ranks obesity into three classes based on BMI.25 At present, due to various social, psychological and medical reasons, an ever-growing number of people are obese in the sense of having a BMI above 30.00. It is obvious that they cannot be considered as disabled unless their condition is of unusual severity even if they can be considered as having a condition that medically can be considered as a long-term illness. The CJEU also confirmed that obesity as such does not amount to a disability. I took the further position in my Opinion that, in cases where the ­condition of obesity has reached a degree that it, in interaction with attitudinal and ­environmental barriers, as mentioned in the UN Convention, plainly hinders full ­participation in professional life on an equal footing with other employees due to the physical and/or psychological limitations that it entailed, then it could be considered to be a disability. In my opinion this meant that ‘mere’ obesity in the sense

18 

See judgment in Case C-363/12 Z, EU:C:2014:159, para 77 and the case law cited. See opinion of AG Jääskinen (EU:C:2014:2106) paras 46 and 47. C-13/05 Chacón Navas, EU:C:2006:456, para 46. For a discussion of the relationship between sickness and ‘disability’ see the Opinion of AG Geelhoed in Chacón Navas, EU:C:2006:184, points 77–80 and the Opinion of AG Kokott in Case C-335/11 HK Danmark, EU:C:2012:775, points 30–38 and 46. 21  Case C-335/11 HK Danmark, EU:C:2013:222, para 41. 22  ibid, para 47. See also the Opinion of AG Kokott in that case at point 28. 23  ibid, para 41. 24  He referred to the American Disabilities Act of 1990 and the judgment of the United States District, Eastern District of Louisiana, EEOC v Resources for Human Dev, Inc, 827 FSupp 2d 688, 693–94 (ED La 2011). 25  See n 2 above. 19 

20 Case

Discrimination on Grounds of Obesity

 361

of WHO class I obesity was insufficient to fulfil the criteria in the case law on ‘disability’ under Directive 2000/78. I considered that most probably only WHO class III obesity, ie severe, extreme or morbid obesity, would create limitations, such as problems in mobility, endurance and mood, that amounted to a ‘disability’ for the purposes of Directive 2000/78.26 I advised the CJEU to leave this factual assessment to the national referring court, and this suggestion was adopted in the judgment although the judgment did not discuss the pertinence of BMI in this respect. Finally, in my Opinion I noted that the CJEU had already held that it ‘would run counter to the very aim of the directive, which was to implement equal treatment, to define its scope by reference to the origin of the disability’. Therefore, in the context of obesity, it was irrelevant for the purposes of Directive 2000/78 whether the person concerned became obese due to simple excessive energy intake, in relation to energy expended, or whether it could be explained by reference to a psychological or metabolic problem, or as a side effect of medication. The notion of disability under Directive 2000/78 was objective and did not depend on whether it was ‘self-inflicted’ in the sense that the person had contributed causally to the acquisition of the disability. Otherwise, physical disabilities resulting from conscious and negligent risk-taking in traffic or in sports, for example, would be excluded from ‘disability’ in the sense of Article 1 of Directive 2000/78.27 The CJEU confirmed that the concept of ‘disability’ within the meaning of Directive 2000/78 does not depend on the extent to which the person may or may not have contributed to the onset of his disability.28 In Kaltoft the representative of the employer was concerned that admitting obesity in any form as being a disability would lead to intolerable results because alcoholism and drug addiction could then, as serious illnesses, be covered by that notion. I considered this concern to be misplaced.29 It is true that, in medical terms, alcoholism and addiction to psychotropic substances are diseases. This does not, however, mean that an employer would be required to tolerate an employee’s breach of his contractual obligations by reference to these diseases. For example, a dismissal because the employee comes to work intoxicated is not based on the disease of alcoholism or drug addiction as such, but is a breach of the employment contract which the employee could have avoided by abstaining from consuming alcohol or the substance in question. Any employer is entitled to expect such an employee to seek the medical treatment that is necessary for him to be able to properly perform his obligations under the contract of employment. It is worth recalling that Article 5 of Directive 2000/78 merely requires employers to provide ‘reasonable accommodation’ to persons with disabilities.

26 

See Opinion of AG Jääskinen (EU:C:2014:2106) paras 55 and 56. ibid, para 58. 28  ibid, paras 55–56. 29  ibid, para 59. 27 

362 

Niilo Jääskinen

The essence of the CJEU’s answer to the fourth preliminary question was that obesity does not in itself constitute a ‘disability’ within the meaning of Directive 2000/78, because, by its nature, it does not necessarily entail the existence of a ­limitation.30 However, in the event that, under given circumstances, the obesity of the worker concerned entails a limitation which results in particular from physical, mental or psychological impairments that in interaction with various barriers may hinder the full and effective participation of that person in professional life on an equal basis with other workers, and the limitation is a long-term one, obesity can be covered by the concept of ‘disability’ within the meaning of Directive 2000/78. Such would be the case, in particular, if the obesity of the worker hindered his full and effective participation in professional life on an equal basis with other workers on account of reduced mobility or the onset, in that person, of medical conditions preventing him from carrying out his work or causing discomfort when carrying out his professional activity.

V. Conclusion In summary, in EU law obesity is not, as such, a ‘disability’ but may under certain circumstances be considered a disability provided that the criteria set out in the case law are fulfilled. In this latter case the employee is protected against discrimination in accordance with the provisions of Directive 2000/78. It may be questioned whether it is desirable that discrimination on the basis of obesity, ie a person’s BMI when it has reached a certain level, is captured by EU law only when the condition of the person concerned has reached the state of having a disability due to his or her obesity. There are good reasons for protecting employees against discrimination based on any ground that is not appropriate and objectively justified with reference to the requirements of the tasks he or she is required to perform.31 This approach would protect employees against discrimination on grounds of physiological, psychological or cultural qualities such as appearance, temperament or accent. Adoption of such a regulatory strategy would not entail unreasonable burdens to employers because Article 4 of Directive 2000/78 allows difference of treatment where by reason of the nature of particular occupational activities, or of the context in which they are carried out, there are genuine and determining occupational requirements that an employee must fulfil, and the requirements are proportionate. This principle could be applied also in the case

30 

ibid, paras 58–60. An example of such open-ended anti-discrimination clause is in Art 6(2) of the Constitution of Finland: ‘No one shall, without an acceptable reason, be treated differently from other persons on the ground of sex, age, origin, language, religion, conviction, opinion, health, disability or other ­reason that concerns his or her person’ (my emphasis). Cited from www.finlex.fi/en/laki/kaannokset/1999/ en19990731.pdf. 31 

Discrimination on Grounds of Obesity

 363

where the excluded grounds of discrimination were not enumerated in the relevant EU law provisions. Thus, for example, it could still be required that a fighter pilot fits in the cockpit or a truck driver has sufficient eyesight. Nevertheless, the enumeration principle inherent in Article 19 TFEU and Article 1 of Directive 2000/78 are direct consequences of the principle of attributed competences contained in Article 5 TEU. Therefore, a reform in this respect would require a Treaty amendment or at least recourse to Article 352 TFEU.

364 

21 Discrimination on the Grounds of Sexual Orientation and Gender Identity ALINA TRYFONIDOU

I. Introduction The aim of this chapter is to examine the contribution that the general principles of EU law1 have made to the protection of the rights of lesbian, gay, bisexual and trans2 individuals (LGBT)3 and same-sex couples under EU law. The focus, in

1  It should be noted that for ease of reference the umbrella terms ‘EU’ and ‘EU law’ will be used throughout this chapter, even when referring to periods preceding the establishment of the EU. 2  The terminology coined by Professor Stephen Whittle is used in this chapter. Thus, the umbrella term ‘trans’ is used to refer to ‘those people who do not perceive or present their gender identity as the same as that expected of the group of people who were given the equivalent sex designation at birth’. This category of persons includes further sub-categories, the categories relevant to this chapter being ‘transgender persons’ and ‘transsexuals’: ‘transgender’ is used to ‘indicate all those p ­ eople who live, or desire to live, a large part of their adult life in the role and dress of that gender group which would be considered to be in opposition to their sex as designated at birth’, whilst ‘transsexual’ is ‘someone who is intending to undergo, is undergoing or has undergone gender reassignment treatment’. See S Whittle, Respect and Equality: Transsexual and Transgender Rights (London, Routledge, 2002) xxii–xxiii. Unless there is a need to distinguish between the latter two categories, the umbrella term ‘trans’ or ‘trans person(s)’ will be used. It should be noted that as regards medical intervention, the term ‘gender reassignment surgery’ will be used, which is the term used by the ECJ. Other terms which have been used in preference over this are ‘gender confirmation surgery’ and ‘gender confirmation treatment’—see JM Scherpe and P Dunne, ‘The Legal Status of Transsexual and Transgender Persons—Comparative analysis and recommendations’ in JM Scherpe (ed), The Legal Status of Transsexual and Transgender Persons (Antwerp, Intersentia, 2015) 618. For a definition of the terms used at EU level, see Fundamental Rights Agency Report ‘Being Trans in the European Union: Comparative Analysis of EU LGBT Survey Data’ (2014) 14–15, available at https://fra.europa.eu/sites/default/files/ fra-2014-being-trans-eu-comparative_en.pdf. 3  Although—as the LGBT acronym demonstrates—persons of homosexual or bisexual orientation are, often, grouped together with trans persons, different legal issues are involved in the treatment of these groups of persons and, therefore, in reflection of this, the chapter will examine these categories of persons separately, although, where appropriate, a comparison between them and their treatment under EU law will be conducted.

366 

Alina Tryfonidou

­ articular, will be on the general principle of equality and its use as a source of the p EU prohibition of discrimination on the grounds of sexual orientation and gender reassignment. This chapter will follow a chronological approach. It will begin (Section II) with a brief examination of the general principle of equality and the various roles it has played in the process of European integration. This will be followed by a section (Section III) presenting the early—pre-Amsterdam—approach of the EU towards the protection of LGBT persons from discrimination. Section IV will, then, consider the impact that the introduction of (what is now) Article 19 TFEU and the resultant promulgation of the Employment Equality Directive (EED) (Directive 2000/784) has had on the protection of the rights of LGB persons, with specific reference being made to the Court’s initial interpretation of the Directive. In addition, the Court’s rulings on the position of transsexuals during that same period will be analysed. Section V will proceed to explore the post-Lisbon legal framework for the protection of LGBT rights in the EU, and will examine the case law of the Court during this period, where the prohibition of discrimination on the ground of sexual orientation was applied.5 The final section will conclude.

II.  The General Principle of Equality Early on, it was realised that Treaty provisions, and the rules laid down for their implementation, were not sufficiently comprehensive to deal with all the issues that had emerged—and would emerge—in the process of EU integration. Accordingly, it was recognised that the above written instruments should be complemented by general—mostly unwritten—principles of EU law. According to Tridimas, general principles of law are ‘fundamental propositions of law which underlie a legal system and from which concrete rules or outcomes may be derived’.6 Hence, the general principles of EU law have constitutional status, and they underlie the EU legal system. They operate in order to fill-in gaps in the Treaties and EU secondary legislation, as aids to interpretation or as grounds for judicial review, and they bind the EU institutions and—in certain circumstances7—the Member States. 4 Directive 2000/78 establishing a general framework for equal treatment in employment and occupation [2000] OJ L303/16. 5  The position of trans persons will not be considered in this section, for the simple reason that during this period no cases involving this category of persons were referred to the Court. 6  T Tridimas, The General Principles of EU Law, 2nd edn (Oxford, Oxford University Press, 2006) 1. 7  That is, when they implement EU law (Joined Cases 201 and 202/85 Klensch v Secretaire d’Etat a l’Agriculture et a la Viticulture [1986] ECR 3477); when they rely on the Treaty derogations (Case C-260/89 ERT [1991] ECR I-2925) or the mandatory requirements (Case C-368/95 Vereinigte Familiapress Zeitungsverlags- und Vertriebs GmbH v Bauer Verlag [1997] ECR I-3689) in order to justify measures which impose an obstacle to free movement; and even, as ruled in some cases, when national rules ‘fall within the scope of EU law’—see, for instance, Case C-617/10 Åklagaren v Åkerberg Fransson, EU:C:2013105.

Discrimination on the Grounds of Sexual Orientation and Gender Identity  367   Some of them have, even, been held to be applicable also to the actions of private bodies or persons.8 Examples of general principles of EU law are the principles of proportionality, legal certainty and, of course, equality. Focusing, now, on the latter, the Court in the 1970s held that the prohibition of discrimination (based on various grounds) that is enshrined in different Treaty provisions is derived from—and is a specific illustration of—the general principle of equality, which is one of the general principles of EU law.9 The position of equality as a general principle of EU law has been further enhanced following subsequent Treaty amendments, and as a result of the changes made by the Treaty of Lisbon, Article 2 TEU now provides that equality is among the basic values of the EU, and Article 3 TEU now includes combating discrimination among the EU’s aims.10 A number of commentators have explored the various goals that the principle of equality and non-discrimination is expected to achieve in the EU context, and the different roles that the principle has, in fact, played in the EU legal system. Tridimas has noted that equality has a dual function: it operates both as a principle of public law, seeking to protect the individual vis-à-vis EU and national authorities, and as a principle of substantive law, seeking to facilitate integration and the achievement of the EU’s objectives.11 Schiek, on the other hand, has pointed out that: Non-discrimination as a general principle of Community law … rests on different, possibly contradictory approaches to equality: from an economic liberal perspective, non-discrimination as a principle of law secures free market access. … However, from a human rights perspective, non-discrimination has to serve more (and more ambitious) goals than from the internal-market perspective. As human rights, equality and nondiscrimination are not restrained to some technical comparative right, but are instituted to sustain an inclusive society.12

Moreover, in More’s view,13 the general principle of equality has fulfilled three principal roles—a market-unifying role, a regulatory role and a constitutional role. 8  For more on the application of general principles of EU law against individuals see Tridimas (n 6) 47–50. 9  Joined Cases 117/76 and 16/77 Ruckdeschel v Hauptzollamt Hamburg-St Annen [1977] ECR 1753, para 7. The Court had, already, in 1972 referred to equality as a ‘fundamental principle of Community law’, without however stating expressly (as it did in Ruckdeschel) that it is, also, one of the general principles of Community law—see Case 1/72 Frilli v Belgium [1972] ECR 457, para 19. 10  The Court emphasised the importance of the principle of equal treatment as a constitutional principle of EU law in, inter alia, Case C-144/04 Mangold v Rüdiger Helm [2005] ECR I-9981. For an article concerning the constitutionalisation of equality, see C O’Cinneide, ‘The Constitutionalization of Equality within the EU Legal Order: Sexual Orientation as a Testing Ground’ (2015) 22 Maastricht Journal of European and Comparative Law 370. 11  Tridimas (n 6) 7. 12  D Schiek, ‘A New Framework on Equal Treatment of Persons in EC Law? Directives 2000/43/ EC, 2000/78/EC and 2002/???/EC Changing Directive 76/207/EEC in Context’ (2002) 8 European Law Journal 290, 293. 13  G More, ‘The Principle of Equal Treatment: From Market Unifier to Fundamental Right?’ in P Craig and G de Búrca (eds), The Evolution of EU Law (Oxford, Oxford University Press, 1999) 518.

368 

Alina Tryfonidou

McCrudden and Kountouros, on the other hand, have focused on exploring the changes in the role(s) played by equality and non-discrimination over the years: In the European Community, rights to equality (in respect of pay between men and women) and non-discrimination (in respect of nationality) were originally conceived as legal instruments to ensure the establishment and proper functioning of the common market. Subsequent political and legislative developments reflect broader social considerations, leading to the recognition of new rights in a range of areas. … Simultaneously, existing rights, such as the right to equal pay, were being remodelled on the basis of both economic and social considerations.14

In addition, the same authors pointed out that Article 13 EC (current Article 19 TFEU) which—as will be seen subsequently—is the provision that gave competence to the EU to make legislation prohibiting discrimination on the ground of, inter alia, sexual orientation appears to be part of a yet further development in Community law towards recognizing the right to equality and non-discrimination as an ‘autonomous principle’, that is a human right that is of value independently of the economic or social benefits that it may bring.15

Agreeing with the latter point, although in other areas of EU integration the general principle of equality had, originally, been used mostly instrumentally, in furtherance of the economic and/or social aims of the EU, in cases involving LGBT individuals, its role—at least on paper—has been from the beginning mainly that of protecting the fundamental human rights of this segment of the EU population. Accordingly, in this context, the general principle of equality is considered a specific ‘genre’ of another set of general principles of EU law: fundamental (human) rights. The important question, of course—and the one on which I shall focus in this chapter—is whether the way that the prohibition of discrimination on the grounds of sexual orientation and gender reassignment has been applied so far in the ECJ’s jurisprudence is congruent with the role that this prohibition—as a specific expression of the human right to equality—is expected to play.

III.  First Steps Towards the Protection of the Rights of LGBT Persons in the EU: The Period before Amsterdam Although a number of steps to require public authorities to respect and protect the (human) rights of gays and lesbians were taken in some parts of the (Western) 14  C McCrudden and H Kountouros, ‘Human Rights and European Equality Law’ in H Meenan (ed), Equality Law in an Enlarged European Union Understanding the Article 13 Directives (Cambridge, Cambridge University Press, 2007) 112. 15  ibid, 113.

Discrimination on the Grounds of Sexual Orientation and Gender Identity  369   world as long ago as the 1950s and 1960s, the first organisations that had as their exclusive and explicit aim to vocalise the concerns of the gay and lesbian community and to work towards the protection of the rights of this segment of the population were only established in the 1970s, following the Stonewall riots in New York in June 1969.16 The need to recognise and protect the human rights of trans persons was, however, recognised much later, in the 1990s, when the ‘LGBT’ acronym made its first appearance.17 LGBT rights—as a sub-category of human rights—found their place in international human rights protection only in the 1990s, and only as a result of purposive judicial interpretation rather than the amendment of human rights instruments to make explicit reference to them. Thus, the Council of Europe’s ECHR did not— and still does not—include any reference to gay and lesbian rights specifically, or to the rights of trans persons. Nor does the prohibition of discrimination included among its provisions (Article 14 ECHR and Protocol 12 to the ECHR) make reference to discrimination on the ground of sexual orientation or gender identity, although the Strasbourg Court has held that both of the latter forms of discrimination are prohibited by the Convention,18 and a number of other ECHR provisions have been employed either alone or in combination with Article 14 ECHR, in order to ensure that LGBT persons can require the states that are signatories to the ECHR to respect their human rights.19 As regards the EU, as is well known the founding Community Treaties did not contain a reference to fundamental human rights,20 and, unsurprisingly, there was no reference among their provisions to gay and lesbian rights, as they were known at the time that the Treaties were drafted. Yet, and despite this, some tentative steps aiming to protect the rights of this segment of the population were taken in the 1980s, although, until 1999, all initiatives to this effect consisted of the adoption of soft law measures, which, whilst of symbolic value, in practical terms achieved very little.21

16  For an excellent account of the history of the gay rights movement in the US, see L Faderman, The Gay Revolution: The Story of the Struggle (New York, Simon & Schuster, 2015). 17  S Whittle, The Transgender Debate: The Crisis Surrounding Gender Identities (Reading, South Street Press, 2000) 43. 18  Salgueiro da Silva Mouta v Portugal, App 33290/96, Judgment of 21 December 1999, para 28 (sexual orientation); PV v Spain, App 35159/09, Judgment of 30 November 2010 (gender identity). For an article comparing the response of the ECtHR and the ECJ in cases involving LGBT rights, see R Wintemute, ‘In Extending Human Rights, which European Court is Substantively “Braver” and ­Procedurally “Fitter”?’ in S Morano-Foadi and L Vickers (eds), Fundamental Rights in the EU: A Matter for Two Courts (Oxford, Hart Publishing, 2015). 19  eg Arts 3 (freedom from torture and inhuman or degrading treatment), 8 (right to private and family life), 12 (right to marry). 20  G de Búrca, ‘The Language of Rights and European Integration’ in J Shaw and G More (eds), New Legal Dynamics of European Union (Oxford, Oxford University Press, 1995) 36–37. 21  For an analysis of these steps. see F Russell, ‘Sexual Orientation Discrimination and Europe’ (1995) 145 New Law Journal 374 and M Bell and L Waddington, ‘The 1996 Intergovernmental Conference and the Prospects of a Non-Discrimination Treaty Article’ (1996) 25 Industrial Law Journal 320, 331–36.

370 

Alina Tryfonidou

The EU judiciary, on the other hand, was first confronted with the legal hot potato of LGBT rights in the mid-1990s, firstly, in a case involving a transsexual who was dismissed by her employer once and because she began the process of undergoing gender reassignment surgery (P v S and Cornwall), and, secondly, in a case involving the refusal of an employer to extend to same-sex couples the same employment-related benefits granted to opposite-sex couples (Grant). P v S and Cornwall22 involved a male-to-female transsexual who sought to rely on EU law in order to challenge the decision of her employer to dismiss her as a result of her decision to undergo gender reassignment surgery. Given that in the mid-1990s, it was essentially only discrimination on the grounds of sex and nationality that was prohibited under EU law, the claimant argued that the contested dismissal amounted to discrimination based on sex, and since dismissal falls within the scope of ‘working conditions’, it was claimed that the discrimination complained of should be held to amount to a breach of the 1976 Equal Treatment Directive.23 The Court was receptive to these arguments, holding that the contested dismissal amounted to a breach of the 1976 Directive. The Court began its analysis by noting that the Directive ‘is simply the expression, in the relevant field, of the principle of equality, which is one of the fundamental principles of Community law’.24 It then referred to its previous statement from Defrenne,25 that ‘the right not to be discriminated against on grounds of sex is one of the fundamental human rights whose observance the Court has a duty to ensure’,26 and from this it concluded that a broad, purposive, approach must be taken when interpreting the Directive, noting that its scope cannot be confined simply to discrimination based on the fact that a person is of one or other sex. In view of its purpose and the nature of the rights which it seeks to safeguard, the scope of the directive is also such as to apply to discrimination arising, as in this case, from the gender reassignment of the person concerned.27

The Court then explained that in establishing the existence of sex discrimination in such a case, a post-operative transsexual must be compared ‘with persons of the sex to which he or she was deemed to belong before undergoing

22  Case C-13/94 P v S and Cornwall City Council [1996] ECR I-2143. For comments, see H Lardy and A Campbell, ‘Discrimination against Transsexuals in Employment’ (1996) 21 European Law Review 412; P Skidmore, ‘Can Transsexuals Suffer Sex Discrimination’ (1997) 19 Journal of Social Welfare and Family Law 105; L Flynn, ‘Annotation of Case C-13/94, P v S and Cornwall County Council, Judgment of the Full Court of 30 April 1996, [1996] ECR I-2143’ (1997) 34 Common Market Law Review 367; R Wintemute, ‘Recognising New Kinds of Direct Sex Discrimination: Transsexualism, Sexual Orientation and Dress Codes’ (1997) 50 Modern law Review 334, 339–44. 23  Directive 76/207 on the implementation of the principle of equal treatment for men and women as regards access to employment, vocational training and promotion, and working conditions [1976] OJ L39/40. 24  P v S and Cornwall (n 22) para 18. 25  Case 149/77 Defrenne v Sabena (Defrenne II) [1978] ECR 1365, paras 26 and 27. 26  P v S and Cornwall (n 22) para 19. 27  ibid, para 20.

Discrimination on the Grounds of Sexual Orientation and Gender Identity  371   gender ­reassignment’.28 Finally, the Court drew another link between the need to prohibit discrimination on the ground of gender reassignment and the need to protect the human rights of transsexuals: ‘To tolerate such discrimination would be tantamount, as regards such a person, to a failure to respect the dignity and freedom to which he or she is entitled, and which the Court has a duty to safeguard.’29 Accordingly, in this case the Court made it clear that discrimination on the ground of gender reassignment is included within the prohibition of discrimination on the ground of sex and—as such—is prohibited by EU law. The judgment has been characterised as a ‘rare legal victory’ for trans persons.30 This is because—due to the human rights basis of the prohibition of discrimination on the ground of sex—the Court took a broad, purposive, approach to the interpretation of the prohibition on discrimination on the ground of sex, interpreting it, in effect, as being discrimination on the ground of gender: the less favourable treatment suffered by P was due to the fact that her appearance and behaviour did not conform to the stereotypical expectations of the society for a person of her biological sex.31 Hence, as Lardy and Campbell noted when commenting favourably on the case: [T]he phrase ‘on the grounds of sex’ (in Article 5(1) of the [1976] Directive) now carries a much broader meaning than that previously attributed to it. The Court construed the phrase to cover treatment triggered by the applicant’s expression of his or her preferred sexual status.32

Shortly after it delivered its ruling in P v S and Cornwall, the Court was called on to answer the question whether the prohibition of discrimination on the ground of sex also includes discrimination on the ground of sexual orientation or, if not, whether the latter form of discrimination was—at the time—prohibited by EU law. Grant33 involved a female employee of South-West Trains (in the UK) who was refused travel concessions for her (long-term) female partner, although her (male) predecessor in post did receive the same concessions for his (long-term) female partner. Ms Grant claimed that this amounted to (direct) discrimination on the ground of sex and relied on Article 119 EC (now Article 157 TFEU), which

28 

ibid, para 21. ibid, para 22. 30  Whittle (n 17) 55. 31  J McInnes, ‘Annotation of Case C-249/96, Lisa Jacqueline Grant v South West Trains Ltd, Judgment of the Full Court of 17 February 1998, [1998] ECR I-636’ (1999) 36 Common Market Law Review 1043, 1052. 32  Lardy and Campbell (n 22) 415. 33  Case C-249/96 Grant v South-West Trains [1998] ECR I-621. For comments see KA Armstrong, ‘Tales of the Community: Sexual Orientation Discrimination and EC Law’ (1998) 20 Journal of Social Welfare and Family Law 455; McInnes (n 31); E Deards, ‘Discrimination on Grounds of Sexual Orientation’ (1999) 10 King’s College Law Journal 12; M Bell, ‘Shifting Conceptions of Sexual Discrimination at the Court of Justice: from P v S to Grant v SWT’ (1999) 5 European Law Journal 63. 29 

372 

Alina Tryfonidou

requires equal pay for equal work, and the directives on equal treatment of men and women. Moreover, she contended that the refusal constituted discrimination based on sexual orientation, ‘which is included in the concept of “discrimination based on sex”’.34 Indeed, if the Court followed the same (broad) approach it had adopted in P v S and Cornwall, which considered whether the disadvantage suffered by a person is due to the fact that he or she does not conform to the stereotypical assumptions of what is expected of a person of his or her biological sex, discrimination on the ground of sexual orientation would, also, be considered discrimination on the ground of sex.35 The Court, however, disagreed. In relation to the claim that the contested refusal amounted to discrimination on the ground of sex, the Court—contrary to what was suggested by the Advocate General in the case—employed the ‘equal misery’ argument, comparing Ms Grant with a (hypothetical) male employee claiming the same concessions for his (male) partner;36 a methodology that has been widely criticised in the literature, since it uses the wrong comparator. The correct comparator would be a male colleague with a female partner, since in this way, the only circumstance that is changed when the comparison is made is the sex of the claimant, whereas under the comparison drawn by the Court, there was a change of two circumstances (both the sex of the claimant and the sex of the partner).37 The Court also pointed out that ‘in the present state of the law within the Community, stable relationships between two persons of the same sex are not regarded as equivalent to marriages or stable relationships outside marriage between persons of opposite sex’.38 This is a statement which is, clearly, discriminatory against samesex relationships as it draws a distinction between stable relationships outside marriage comprised of two persons of the opposite sex (which are treated in the same way as marriages) and stable relationships outside marriage comprised of two persons of the same sex (which are treated differently from—less well than— marriages). Moreover, it reflects the existence of a strong heteronormativity in the ECJ’s views and assumptions. Finally, the Court examined Ms Grant’s third argument—that discrimination on the ground of sexual orientation is (like discrimination on the ground of gender reassignment) a guise of discrimination on the ground of sex—and concluded that it is not.39 However, possibly wishing to indicate that some protection to the rights of LGB persons would soon be provided under EU law, it pointed out that the Treaty of Amsterdam provided for the insertion of a new legal basis which would give competence to the EU to make legislation prohibiting discrimination on a number of grounds, including sexual orientation.40 34 

Grant (n 33) para 18. McInnes (n 31) 1051. 36  Grant (n 33) para 27. 37  Flynn (n 22) 382; Wintemute (n 22) 347–48; McInnes (n 31) 1049–50. 38  Grant (n 33) para 35 (emphasis added). 39  ibid, paras 37–47. 40  ibid, para 48. 35 

Discrimination on the Grounds of Sexual Orientation and Gender Identity  373   The ECJ’s approach was equally cautious in its subsequent judgment in D,41 which was an appeal from a CFI judgment in a staff case.42 Although the (ECJ) judgment was delivered in 2001—ie after the Treaty of Amsterdam came into force—it is, nonetheless, discussed in this section, since the facts of the case took place in the mid-1990s and it was therefore the old Staff Regulations (before the subsequent amendments that made provision for equal treatment of marriages and registered partnerships43) that were applicable. Moreover, the approach followed by the Court is very similar to the one it adopted in Grant and, hence, the changes brought by Amsterdam did not appear to have any impact on the Court’s reasoning in the case. At issue was the refusal of the Council to grant to one of its officials (D, a Swedish man) a household allowance, because the version of the Staff Regulations that was applicable at the time made such an allowance available only to ‘married officials’. The said official was not married but had entered into a registered partnership with another man in Sweden. D had argued that because Swedish law provided that registered partnerships should have the same legal effects as a marriage, the Staff Regulations should be read as granting the said allowance also to officials who had entered into a same-sex registered partnership, even though they were not married. The Council rejected this argument, noting that the Staff Regulations could not be construed as requiring a registered partnership to be treated as equivalent to marriage. When the issue reached the CFI, the latter held that the Council was right in rejecting D’s application and in ruling that the Staff Regulations that were applicable at the time did not require the grant of the household allowance to a person in D’s situation; the contested rejection did not amount to discrimination on the ground of sex.44 The views of the ECJ were not much different. It pointed out that ‘according to the definition generally accepted by the Member States, the term “marriage” means a union between two persons of the opposite sex’,45 and noted that, although since 1989 an increasing number of Member States had provided legal recognition to same-sex relationships conferring on such unions effects which are the same or comparable to those of marriage,46 such registered partnerships nonetheless remain distinct from marriage.47 For this reason, ‘the Community

41 

Joined Cases C-122 and 125/99 P D and Sweden v Council [2001] ECR I-4319. T-264/97 D v Council [1999] ECR II-1. For comments, see C Denys, ‘Homosexuality: A Non-Issue in Community Law?’ (1999) 24 European Law Review 419; G Scappucci, ‘Court of First Instance Refuses to Recognize Swedish “Registered Partnership” Rights and Duties’ (2000) 6 European Public Law 355. 43 See Council Regulation 723/2004 Amending Staff Regulations of Officials of the European Communities, Annex I, para 3, [2004] OJ L124/1. 44  D v Council (n 42) paras 42–44. 45  D and Sweden v Council (n 41) para 34. 46  ibid, para 35. 47  ibid, para 36. 42 Case

374 

Alina Tryfonidou

judicature cannot interpret the Staff Regulations in such a way that legal situations distinct from marriage are treated in the same way as marriage’.48 Moreover, the Court agreed with the CFI that there was no discrimination on the ground of sex, since ‘it is irrelevant for the purposes of granting the household allowance whether the official is a man or a woman’,49 thus following the ‘equal misery’50 approach it had adopted in Grant. The Court then pointed out that ‘it is clear that it is not the sex of the partner which determines whether the household allowance is granted, but the legal nature of the ties between the official and the partner’,51 in this way ignoring the fact that because of his sexual orientation and the resultant desire to be in a same-sex relationship, the applicant was not capable (at the time) of marrying his partner under Swedish law. The Court also noted the diversity of laws in the Member States as regards the legal recognition of same-sex relationships, and noted further that ‘the situation of an official who has registered a partnership in Sweden cannot be held to be comparable, for the purposes of applying the Staff Regulations, to that of a married official’.52 The EU courts’ approach in Grant and D can be characterised as cautious, at best, and as homophobic, at worse. This is because the message that it sent was that the employers of LGB individuals were entitled to discriminate against them simply because they were in a same-sex relationship, without even needing to point to a justification for this. This implied that LGB persons were not entitled to respect of their dignity under EU law;53 as Lardy and Campbell have noted: If tolerating discrimination against a transsexual person would be a ‘failure to respect the dignity and freedom to which he or she is entitled’, then to permit discrimination against someone on the grounds that he or she is gay or lesbian must surely also represent such a failure of respect.54

A number of reasons may be lying behind the Court’s difference in approach between transsexuals, on the one hand, and LGB persons, on the other. The first one is a pragmatic reason, this being that the financial implications of recognising discrimination based on gender reassignment as a guise of discrimination on the ground of sex would be miniscule, since, as noted by Advocate ­General Tesauro in his Opinion in P v S and Cornwall, ‘transsexuals are in fact not very significant in statistical terms’.55 Conversely, LGB individuals—although

48 

ibid, para 37. ibid, para 46. 50  Denys (n 42) 423. 51  D and Sweden v Council (n 41) para 47. 52  ibid, para 51. 53  For an argument that the ECJ should engage to a greater degree with human dignity in its equality jurisprudence, see O’Cinneide (n 10). 54  Lardy and Campbell (n 22) 417–18. See also McInnes (n 39) 1053. 55  Opinion of AG Tesauro in P v S and Cornwall (n 22) para 20. Of course, it should be noted that this approach takes into account only trans persons who have chosen to undergo gender reassignment surgery. For an argument that the percentage of trans people is not as small as it is generally thought, see Whittle (n 17) 25–32. 49 

Discrimination on the Grounds of Sexual Orientation and Gender Identity  375   a relatively slim minority—form a bigger group than trans persons (and definitely a much bigger group than transsexuals) and, hence, the Court might be feeling that its authority to rule on the matter would be more likely to be questioned by disgruntled employers who would have to foot the bill if the same-sex relationships of their LGB employees were recognised.56 Secondly, as noted by Mark Bell, it is more difficult to achieve political consensus or legislative protection relating to gender identity than in respect of sexual orientation.57 Accordingly, the Court may have felt that judicial action was the only course available to protect the equality rights of trans persons, whereas it was clear that steps for the legislative protection of LGB persons from discrimination would soon be taken following the introduction of (what is now) Article 19 TFEU. Finally, another reason was offered by Flynn for the Court’s more cautious approach towards the protection of the rights of LGB individuals: Transsexuals effectively ask to be treated as the woman (or man) that they consider themselves to be, and whose external physical features they possess after surgery and hormonal treatment. They move from belonging to one sex to the other but do not call into dispute the social roles and expectations imposed on men or women as such. By contrast, for many people lesbians and gay men offer a more fundamental challenge to the social meaning assigned to what it is to be a ‘woman’ or a ‘man’ precisely because they do not wish in any way to be less of a woman or man by reason of their sexual orientation.58

IV.  From Amsterdam to Lisbon The Treaty of Amsterdam came into force in May 1999. This is the Treaty that opened the way for the recognition and protection of LGB rights under EU law. It is, nonetheless, also an important milestone for the protection of equality in the EU more broadly. The Treaty brought equality between men and women within the EU’s objectives, and it introduced for the first time a mainstreaming provision for this aspect of equality, emphasising the need for the EU to take it seriously and to accommodate it in all its policies.59 This demonstrated that equality between men and women would no longer be protected under EU law merely in order to contribute to the achievement of the EU’s economic and social objectives, but it had, now, become a cause that was worthy on its own, as an aspect of human rights protection. This seems to have been also endorsed judicially since, 56 

For a similar argument, see Flynn (n 22) 367; Bell (n 33) 74–75. M Bell, ‘Gender Identity and Sexual Orientation: Alternative Pathways in EU Equality Law’ (2012) 20 American Journal of Comparative Law 127, 133. 58  Flynn (n 22) 381. 59  Currently found in Art 8 TFEU. For comments see F Beveridge, ‘Building Against the Past: The Impact of Mainstreaming on EU Gender Law and Policy’ (2007) 32 European Law Review 193. 57 

376 

Alina Tryfonidou

shortly after this Treaty came into force, the Court noted that the economic aim pursued by (what is now) Article 157 TFEU is secondary to the social aim pursued by the same provision, which constitutes the expression of a fundamental human right.60

A.  The Protection of Transsexuals from Discrimination The increased importance attached by the Treaty of Amsterdam to the principle of equality between men and women unavoidably benefited transsexuals too. As seen in the previous section, discrimination on the ground of gender reassignment had already—prior to the coming into force of this Treaty—been accepted by the Court as an aspect of discrimination on the ground of sex. However, as will be seen in this section, the protection offered by the EU to transsexuals had been further enhanced after Amsterdam, when the Court was given the opportunity to clarify further its position towards the protection of the rights of this segment of the EU population. In the two cases that emerged, the Court was not confronted with the actions of the employer of a transsexual person which were the direct result of the latter undergoing gender reassignment surgery (as was the case in P v S and ­Cornwall), but—rather—with the (negative) consequences that the failure of national law to recognise transsexuals in their post-operative sex had on their ability to claim certain work-related benefits or entitlements. In KB61 the question was whether the refusal of the UK authorities to award a widower’s pension to the post-operative female-to-male transsexual partner (R) of a female nurse (KB) who was a member of the NHS Pension Scheme amounted to discrimination on the ground of sex, and thus was in breach of (then) Article 141 EC and the Equal Pay Directive.62 The problem was that although R had had gender reassignment surgery, English law at the time did not allow him to amend his birth certificate to reflect this and, hence—and because in the eyes of the law he was still a woman and English law at the time did not allow two persons of the same sex to marry—he and KB could not get married. The Court noted that in a situation such as that before the national court, there is inequality of treatment which … does not relate to the award of a widower’s pension but to a necessary precondition for the grant of such a pension: namely, the capacity to marry.63

From this, it concluded that Legislation, such as that at issue in the main proceedings, which, in breach of the ECHR, prevents a couple such as K.B. and R. from fulfilling the marriage requirement

60 

Case C-50/96 Schröder [2000] ECR I-743, para 57. Case C-117/01 KB v NHS Pensions Agency and Secretary of State for Health [2004] ECR I-541. Directive 75/117 on the approximation of the laws of the Member States relating to the application of the principle of equal pay for men and women [1975] OJ L45/19. 63  KB (n 61) para 30. 61 

62 

Discrimination on the Grounds of Sexual Orientation and Gender Identity  377   which must be met for one of them to be able to benefit from part of the pay of the other must be regarded as being, in principle, incompatible with the requirements of Article 141 EC.64

Richards65 involved a male-to-female post-operative transsexual who argued that the UK’s refusal to grant her a retirement pension at the age of 60 (which was the age that women could receive such a pension under English law) on the ground that—in the eyes of the law—she was (still) a man, amounted to a breach of the prohibition of discrimination on the ground of sex. The ECJ held that this was, indeed, the case and that the contested refusal was a breach of Directive 79/7, which prohibited discrimination on the ground of sex in the area of social ­security.66 After confirming that ‘the right not to be discriminated against on grounds of sex is one of the fundamental human rights the observance of which the Court has a duty to ensure’,67 the Court proceeded to interpret Directive 79/7 broadly. Accordingly, and following the same reasoning that it applied when interpreting Directive 76/207 in P v S and Cornwall, the Court noted that: The scope of Directive 79/7 cannot thus be confined simply to discrimination based on the fact that a person is of one or other sex. In view of its purpose and the nature of the rights which it seeks to safeguard, the scope of that directive is also such as to apply to discrimination arising from the gender reassignment of the person concerned.68

The Court then proceeded to explain that the contested refusal was discriminatory on the ground of sex because ‘[t]he unequal treatment at issue in the main proceedings is based on Ms Richards’ inability to have the new gender which she acquired following surgery recognised with a view to the application of the Pensions Act 1995’,69 and: Unlike women whose gender is not the result of gender reassignment surgery and who may receive a retirement pension at the age of 60, Ms Richards is not able to fulfil one of the conditions of eligibility for that pension, in this case that relating to retirement age.70

The above two cases are important, because they made it clear that discrimination on the ground of gender reassignment is not, only, included within the prohibition of discrimination on the ground of sex, in the obvious instances where the discrimination complained of is the direct result of a transsexual having undergone gender reassignment surgery. Rather, the prohibition goes further, and prohibits more subtle instances of discrimination, which are, simply, the result of the failure of the law to cater for the particular situation of transsexual persons and to

64 

ibid, para 34. Case C-423/04 Richards v Secretary of State for Work and Pensions [2006] ECR I-3585. 66  Directive 79/7 on the progressive implementation of the principle of equal treatment for men and women in matters of social security [1979] OJ L6/24. 67  Richards (n 65) para 23. 68  ibid, para 24. 69  ibid, para 28. 70  ibid, para 29. 65 

378 

Alina Tryfonidou

recognise them in their post-operative sex. Bell has, rightly, noted that ‘[i]n retrospect, the gender reassignment cases provide examples of how the Court can move beyond a narrow reading of the law by taking into account contextual principles of human rights’.71 Nonetheless, the Court has not so far been given the opportunity to rule in a case involving a trans person who has not undergone—and does not intend to undergo—gender reassignment surgery. The human rights basis of the equality principle would, nonetheless, clearly require the extension of this principle to the latter and the prohibition of discrimination to be read more broadly, as being a prohibition of discrimination on the ground of gender identity.72

B. The Protection of LGB Individuals and Same-Sex Couples As seen in the previous section, until the coming into force of the Treaty of Amsterdam, LGB individuals did not derive from EU law any protection from discrimination based on their sexual orientation. Moreover, unlike discrimination on the ground of gender reassignment, discrimination on the ground of sexual orientation was held by the Court not to be an aspect of discrimination on the ground of sex. This gap in protection of LGB individuals was, however, filled by the Treaty of Amsterdam, which introduced (what used to be) Article 13 EC,73 which is the (slightly amended) current Article 19 TFEU, which gives the EU competence to make legislation combating discrimination based on a number of grounds, including sexual orientation. It should be underlined that this is merely a competence-giving provision and, thus, the prohibition of discrimination on the ground of sexual orientation, did not, automatically, emerge from its insertion into—what was at the time—the EC Treaty, but was only introduced when, in the subsequent year, the EU legislature took action and promulgated the Employment Equality Directive (EED)—Directive 2000/78.74

71  M Bell, ‘The Principle of Equal Treatment: Widening and Deepening’ in P Craig and G de Búrca (eds), The Evolution of EU Law (Oxford, Oxford University Press, 2011) 617. 72  Fundamental Rights Agency Report (n 2) 19 and 49. At the moment, discrimination on the ground of gender identity is mentioned in two instruments: Recitals 9, 17 and 56 of the Victims’ Directive (Directive 2012/29/EU establishing minimum standards on the rights, support and protection of victims of crime, and replacing Council Framework Decision 2001/220/JHA, [2002] OJ L315/57), and Recital 30 and Art 10(1)(d) of Directive 2011/95/EU on standards for the qualification of thirdcountry nationals or stateless persons as beneficiaries of international protection, for a uniform status for refugees or for persons eligible for subsidiary protection, and for the content of the protection granted (recast), [2011] OJ L337/9. 73  For commentary on the steps that led to the introduction of this provision by the Treaty of Amsterdam, see Bell and Waddington (n 21). For comments on Art 13 EC in general, see L Flynn, ‘The Implications of Article 13 EC—After Amsterdam, Will Some Forms of Discrimination be More Equal than Others?’ (1999) 36 Common Market Law Review 1127; M Bell, ‘The New Article 13 EC Treaty: A Sound Basis for European Anti-Discrimination Law?’ (1999) 6 Maastricht Journal of European and Comparative Law 5. 74  Directive 2000/78 (n 4).

Discrimination on the Grounds of Sexual Orientation and Gender Identity  379   The Directive—still in force—prohibits (direct and indirect) discrimination on the grounds of religion or belief, disability, age or sexual orientation, in the areas of employment, vocational training and membership of a professional organisation.75 Moreover, it provides that instruction to discriminate and harassment shall be deemed to be discrimination contrary to its provisions. Its Preamble demonstrates that its basis is the need to protect equality as an aspect of human rights protection,76 however, its limited material scope shows that in reality it is not so much a general human rights instrument but, rather, a measure which forms part of the EU’s overall economic and social policy. Procedurally, the Directive seeks to facilitate complaints-based enforcement by individuals and for this purpose in its Article 10 it provides for a shift in the burden of proof from the complainant to the respondent on a prima facie showing of discrimination, whilst it also protects individual litigants from victimisation.77 Associations (including NGOs) can engage in legal proceedings either on behalf or in support of the complainant,78 and the Directive provides that sanctions have to be effective, proportionate and dissuasive.79 Although clearly important (in that it is the first EU law instrument that established a binding prohibition of discrimination on the ground of sexual orientation), the Directive has been criticised for not going far enough, especially when compared with the instruments prohibiting discrimination on the grounds of sex and racial or ethnic origin. In particular, the protection offered by the Directive appears to be inferior to that provided by (its sister) Directive 2000/43,80 which was adopted on the same basis—Article 13 EC—and which was promulgated just months before it. This is because its material scope of application is much more limited (it only applies to employment, vocational training and membership of a professional organisation, whereas Directive 2000/43 applies, in addition, to social protection (including social security and healthcare), social advantages, education, and access to and supply of goods and services which are available to the public, including housing), and by allowing a relatively large amount of exceptions, it seems to be protecting the categories of persons that fall within its scope to a lesser extent than those that can rely on Directive 2000/43, which allows more limited scope for exceptions. Moreover, Directive 2000/43 imposes a duty on Member States to designate a body for the promotion of equal treatment of all persons without discrimination on the ground of racial or ethnic origin,81 whereas no such duty is imposed by Directive 2000/78.

75 

ibid, Art 3. ibid, Recitals 1–5. 77  ibid, Art 11. 78  ibid, Art 9(2). 79  ibid, Art 17. 80  Directive 2000/43 implementing the principle of equal treatment between persons irrespective of racial or ethnic origin [2000] OJ L180/22. 81  ibid, Art 13. 76 

380 

Alina Tryfonidou

The first judgment on the prohibition of discrimination on the ground of sexual orientation under Directive 2000/78 was delivered eight years after its coming into force. This was in the case of Maruko,82 where the reference emerged from proceedings between Mr Maruko and the German Theatre Pension Institution (Versorgungsanstalt der deutschen Bühnen, VddB), relating to the refusal by the latter to recognise Mr Maruko’s entitlement to a widower’s pension as part of the survivor’s benefits provided for under the compulsory occupational pension scheme, of which his deceased registered life partner had been a member. The VddB’s refusal was based on the ground that its regulations only provided for such an entitlement for surviving spouses, excluding surviving registered life partners. When considering whether the contested refusal amounted to discrimination on the ground of sexual orientation contrary to the 2000 Directive, the Court pointed out that: [F]rom 2001 … the Federal Republic of Germany altered its legal system to allow persons of the same sex to live in a union of mutual support and assistance which is formally constituted for life. Having chosen not to permit those persons to enter into marriage, which remains reserved solely to persons of different sex, that Member State created for persons of the same sex a separate regime, the life partnership, the conditions of which have been gradually made equivalent to those applicable to marriage.83

The Court, then, summarised the views of the referring court and, without providing its own conclusion as to whether registered partnerships are treated as equivalent to marriage under German law, it pointed out that: If the referring court decides that surviving spouses and surviving life partners are in a comparable situation so far as concerns that survivor’s benefit, legislation such as that at issue in the main proceedings must, as a consequence, be considered to constitute direct discrimination on grounds of sexual orientation, within the meaning of Articles 1 and 2(2)(a) of Directive 2000/78.84

Thus, the Court left it to the referring court to determine whether, for the purposes of the claimed benefit, life partnerships and marriages were in a comparable situation. Accordingly, what the Court stated in Maruko is that if a Member State considers— for a certain purpose (eg the grant of a survivor’s pension)—same-sex registered partnerships as equivalent to marriage, it must treat the former in the same way as it treats the latter. Moreover, and taking a pragmatic approach, the Court ruled that in such a case, the discrimination that is practised is direct discrimination on the ground of sexual orientation: since, on the facts of the case, marriage was

82 Case C-267/06 Maruko v Versorgungsanstalt der deutschen Bühnen [2008] ECR I-1757. For ­ comments, see C Tobler and K Waaldijk, ‘Annotation of Case C-267/06, Tadao Maruko v. ­Versorgungsandstalt der deutschen Bühnen, Judgment of the Grand Chamber of the Court of Justice of 1 April 2008, not yet reported’ (2009) 46 Common Market Law Review 723. 83  Maruko (n 82) para 67. 84  Ibid, para 72.

Discrimination on the Grounds of Sexual Orientation and Gender Identity  381   not available to same-sex couples, legislation which limits the availability of a ­certain entitlement to married couples is taken to amount to (direct) discrimination on the ground of sexual orientation, if the said entitlement is refused to a couple that is not married (because marriage is not open to it) but has entered into a registered partnership;85 a finding that ‘renders Maruko remarkable’,86 and demonstrates that during this phase, the Court (partly) overruled its judgment in D and Sweden v Council, where it had noted that there was no discrimination on the ground of sex because ‘it is clear that it is not the sex of the partner which determines whether the household allowance is granted, but the legal nature of the ties between the official and the partner’.87 The judgment gave cause for both praise and criticism. Starting with the former, the Court should be commended for taking a pragmatic approach when determining what type of discrimination is at issue in such a case. As noted above, the Court ruled that in a situation such as that at issue in Maruko, the discrimination that is practised is direct discrimination on the ground of sexual orientation. In subsequent case law which confirmed this (and which we shall see later), it was made clear that this approach was here to stay.88 This seems to be a (rather isolated) example of an instance where the Court in its sexual orientation case law drew inspiration from a case involving transsexuals: the approach in Maruko seems to have the same basis as the approach that was followed in KB89 where the Court held that if a couple cannot marry because in the eyes of the law (which does not recognise a transsexual in his or her post-operative sex) is comprised of two persons of the same sex, the refusal to such a couple of entitlements that are available to married couples amounts to discrimination on the ground of sex. On the other hand, the part in Maruko where it was stated that only if the Member State concerned considers for a specific purpose same-sex registered partners as being in a comparable situation with opposite-sex married persons should there be equality of treatment between the two, should be criticised because the Court was, in effect, washing its hands of the matter by leaving same-sex couples to the mercy of the Member States. As Toggenburg has rightly pointed out: The Court’s approach in Maruko has two major weaknesses. Firstly, it provides no protection against discrimination where it is most needed, namely in national systems where homosexual relationships find no legal recognition. Secondly, the definition and identification of the point at which EU law steps in is entirely left to the Member States.90

85 

For comments see Tobler and Waaldijk (n 82) 739–40. Eriksson, ‘European Court of Justice: Broadening the Scope of European Nondiscrimination Law’ (2009) 7 International Journal of Constitutional Law 731, 742. 87  D and Sweden v Council (n 41) para 47. 88  Case C-147/08 Römer v Freie und Hansestadt Hamburg [2011] ECR I-3591 and Case C-267/12 Hay v Crédit agricole mutuel de Charente-Maritime et des Deux-Sèvres ECLI:EU:C:2013:823. 89  KB (n 61). 90  GN von Toggenburg, ‘“LGBT” Go Luxembourg: On the Stance of Lesbian Gay Bisexual and Transgender Rights Before the European Court of Justice’ (2008) 5 European Law Reporter 174. 86  A

382 

Alina Tryfonidou

Similarly Möschel has noted that: [A]s long as legislation does not explicitly treat marriage and life partnerships as equivalent, homosexual life partners will be forced to litigate their way through courts with the comparison occurring on a case-by-case basis for each benefit granted in connection with marriage.91

As a general conclusion for this chronological period, therefore, it should be noted that some progress in the position of LGB persons under EU law had, indeed, been achieved, whilst further progress in the protection of transsexuals from discrimination had been made. Hence, the Court confirmed its P v S and Cornwall finding that discrimination on the ground of gender reassignment is included within the scope of sex discrimination and it further clarified that this does not merely include situations where the less favourable treatment complained of is the direct result of a person having undergone (or having decided to undergo) gender reassignment surgery but, also, where it emerges more indirectly, as a result of the failure of a Member State to make provision in its laws for recognising a transsexual person in his or her post-operative sex. As regards LGB persons, by promulgating Directive 2000/78, the EU legislature introduced for the first time in EU law a prohibition of discrimination on the ground of sexual orientation which, as we saw, is interpreted broadly (to cover harassment and instruction to discriminate) and the Directive is designed in such a way as to take into account specific issues that are encountered in discrimination cases (eg victimisation). Moreover, in its Maruko judgment which appears to have—partly—overruled D and Sweden v Council, the Court found that a difference in treatment between (opposite-sex) married couples and (same-sex) registered partners, in Member States that do not allow same-sex couples to marry, is direct discrimination on the ground of sexual orientation and, hence, it has made it more difficult to justify than would have been the case if—as suggested by Member States—this was found to amount to indirect discrimination on the ground of sexual orientation.92 Yet, some gaps in the protection of LGBT persons from discrimination undoubtedly remained, despite the above advances in their protection from discrimination. As regards trans persons, the cases heard by the Court during this period still involved transsexual persons (ie persons who had chosen to undergo a gender reassignment surgery), and the Court in its case law made reference to discrimination on the ground of gender reassignment, rather than more broadly to discrimination on the ground of gender identity. Accordingly, it had not been clarified

91 M Möschel, ‘Germany’s Life Partnerships: Separate and Unequal?’ (2009–10) 16 Columbia Journal of European Law 37, 62. 92  It can only be justified under Art 2(5) of Directive 2000/78 (n 4) instead of, also, under the objective justifications provided for in Art 2(b)(i) of the Directive.

Discrimination on the Grounds of Sexual Orientation and Gender Identity  383   whether the latter—broader—form of discrimination which would be able to cover all trans persons is included within the scope of EU law. In relation to LGB persons, a significant gap in their protection from discrimination remained, which was due to the limited material scope of Directive 2000/78, which only covers the areas of employment and vocational training. Moreover, the interpretation of the Directive by the ECJ in its Maruko judgment had been— partly—disappointing, because it had left same-sex couples in a precarious position: whether they could claim equal treatment with married couples, would depend not only on whether their Member State allowed them to enter into a registered partnership (since the case only offered protection to same-sex couples that had formalised their relationship) but, also, on whether their Member State considered registered partnerships equivalent to marriage. This demonstrated the EU’s reticence to intervene with Member State choices regarding not merely the legal recognition of same-sex relationships (which is, clearly, a matter than falls within national competence93), but also the legal effects of entering into a legal status provided under national legislation.94

V.  The Post-Lisbon Legal Framework for the Protection of LGB Rights from Discrimination This section will focus on the protection that EU law currently offers to LGB persons from discrimination, following the changes made by the Treaty of Lisbon. The section will begin by documenting the changes brought by this Treaty which have an impact on the protection of the rights of LGB persons, and will, then, proceed to consider the cases decided by the ECJ in this area from 2009 onwards. It will conclude by examining whether the post-Lisbon legal framework—and the ECJ’s approach in recent case law—offers a satisfactory protection from discrimination to LGB individuals, reflecting the fact that the prohibition of discrimination on this ground is a fundamental (human) right. Since no advances in the law were made during this period, as regards the protection of trans persons from discrimination, the position of this group of persons will not be examined in this section.

93 

Para 76 of the Opinion of AG Jääskinen in Römer (n 88). should be noted that the EU institutions (especially the EU legislature) have been hesitant in making it entirely clear that for the purposes of family reunification of migrant Union citizens, same-sex relationships should be treated in exactly the same way as opposite-sex relationships—see A Tryfonidou, ‘EU Free Movement Law and the Legal Recognition of Same-Sex Relationships: The Case for Mutual Recognition’ (2015) 21 Columbia Journal of European Law 195. 94  It

384 

Alina Tryfonidou

A.  The Post-Lisbon Legal Framework The Treaty of Lisbon came into force in December 2009. It amended the EU and EC Treaties, and changed the name of the latter to ‘TFEU’. In addition, it made the EU Charter of Fundamental Rights (EUCFR) binding, by providing in Article 6(1) TEU (as amended) that this document now has the same legal value as the Treaties. The coming into force of the Treaty of Lisbon has been immensely important for the protection of LGB rights for a number of reasons. Firstly, the Treaty added a new mainstreaming provision—Article 10 TFEU— which provides that: ‘In defining and implementing its policies and activities, the Union shall aim to combat discrimination based on sex, racial or ethnic origin, religion or belief, disability, age or sexual orientation.’ This is hugely important for, inter alia, the protection of LGB rights, since it proactively ensures—or, at least, requires EU institutions to ensure—that all EU policies are free from discrimination on the ground of sexual orientation and in this way it suitably complements the reactive protection from this sort of discrimination which is offered by secondary legislation (ie Directive 2000/78) and by the Court of Justice. Secondly, as noted above, the EUCFR became binding. This means that its provisions bind the EU institutions, and individuals can bring actions against Member States relying on the Charter in situations that fall within the scope of EU law.95 As regards LGB individuals, this is important because the Charter includes a provision (Article 20) which states that ‘[e]veryone is equal before the law’ and, more significantly from a practical point of view, its Article 21 provides that: Any discrimination based on any ground such as sex, race, colour, ethnic or social origin, genetic features, language, religion or belief, political or any other opinion, membership of a national minority, property, birth, disability, age or sexual orientation shall be prohibited.

Hence, apart from the areas of employment, vocational training and membership of a professional organisation, to which Directive 2000/78 applies, the prohibition of discrimination on the ground of sexual orientation is now applicable in all other areas of human life, provided that the situation falls within the scope of the

95  Art 51(1) of the Charter provides that ‘[t]he provisions of this Charter are addressed to the institutions, bodies, offices and agencies of the Union with due regard for the principle of subsidiarity and to the Member States only when they are implementing Union law.’ Although the Explanations attached to the Charter and the Court’s most recent case law interpreting Art 51, adopt a broad reading of this provision, stating, respectively, that the Charter is binding on the Member States ‘when they act in the scope of Union law’, and that ‘[t]he applicability of European Union law entails applicability of the fundamental rights guaranteed by the Charter’ (Case C-617/10 Fransson, ECLI:EU:C:2013:105, para 21), it has not been made clear, yet, when a situation falls within the scope of EU law for this purpose, and in some of its latest case law, the Court appears to have adopted a rather restrictive approach to this question—see, eg, Case C-45/12 Hadj Ahmed, ECLI:EU:C:2013:390 and Case C-198/13 Hernández, ECLI:EU:C:2014:2055.

Discrimination on the Grounds of Sexual Orientation and Gender Identity  385   Charter. Of course, whether a situation does fall within the scope of the Charter is not clear, with the Court so far giving confusing signals.96 Moreover, the Charter also includes a number of other rights that can be of help to LGB persons—most significantly, the right to human dignity, the right to private and family life, the right to be free from torture or inhuman or degrading treatment—however, this is not, really, an innovation since, even before the coming into force of the Charter, these rights were considered to be fundamental (human) rights which form part of the general principles of EU law and, as such, could be relied on by LGB persons in situations that fell within the scope of EU law.

B.  Case Law in the Post-Lisbon era During this period the Court was called on, again, to interpret Directive 2000/78, but, for the first time, it also had to consider (i) the rights of LGB persons in situations which did not involve discrimination; and (ii) the prohibition of discrimination on the ground of sexual orientation in situations which fell outside the material scope of Directive 2000/78 but within the scope of the EUCFR. Since this chapter focuses on the general principle of equality and the prohibition of discrimination against LGBT persons, the case law involving the rights of LGB persons in situations where there was no discrimination will not be discussed.97

(i)  Cases involving Directive 2000/78 Römer98 was the first case during this period that concerned the interpretation of Directive 2000/78. At issue was the refusal of the German authorities to adopt for former employees who had entered into a German life partnership with their same-sex partners the same method of calculating the supplementary pension to which they were entitled, as that which was used for former employees who were married to their (opposite-sex) partner. The Court in its judgment followed exactly the same reasoning as that adopted in Maruko. Hence, it concluded that the Directive precludes the use of such a different method of calculating the supplementary pension if, firstly, ‘in the Member State concerned, marriage is reserved to persons of different gender and exists alongside a registered life partnership … which is reserved to persons of the same gender’ and there is direct discrimination on the ground of sexual orientation because, under national law, that life partner is in a legal and factual situation comparable to that of a married person as regards that pension. It is for the referring court to assess the comparability,

96 

See the cases cited in the previous footnote. Case F-86/09 W v European Commission, ECLI:EU:F:2010:125; Joined Cases 199–201/12 Minister voor Immigratie en Asiel v X and Y and Z v Minister voor Immigratie en Asiel, ECLI:EU:C:2013:720; Joined Cases C-148-150/13 A, B and C v Staatssecretaris van Veiligheid en Justitie, ECLI:EU:C:2014:2406. 98  Römer (n 88). 97 See

386 

Alina Tryfonidou

focusing on the respective rights and obligations of spouses and persons in a registered life partnership, as they are governed within the corresponding institutions, which are relevant taking account of the purpose of and the conditions for the grant of the benefit in question.

The only difference from Maruko was that the Court provided further guidelines to the national Court for when conducting its comparability assessment, but this did not make any real change to the overall protection afforded to LGB persons. The next case was Asociaţia Accept,99 which, again, involved Directive 2000/78, and which emerged from proceedings between Accept (a Romanian NGO that defends and promotes LGBT rights) and the Romanian Council for Combating Discrimination, concerning the latter’s decision partially dismissing a complaint lodged following public statements made by a person (Mr Becali) who presented himself as having—and was considered by public opinion to play—a leading role in a Romanian professional football club (FC Steaua), ruling out the recruitment by that club of a footballer that was thought to be gay. Accept’s claim was that the principle of equal treatment had been breached in recruitment matters. The Court in its judgment firstly pointed out that in order for direct discrimination—within the meaning of Directive 2000/78—to be established, there is no need for there to be an identifiable complainant who claims to have been the victim of such discrimination;100 it suffices, like in this case, if the action is brought by an NGO without identifying the person who has been harmed by the action with regards to which the complaint has been made. The Court then explained that even if the statements which amount to discrimination come from a person who does not have legal capacity to bind or represent the club in recruitment matters, this does not matter, and the fact that the club has not distanced itself from the statements concerned is a factor which the referring court may take into account when making its judgment.101 The Court then recalled that the sanctions that Article 17 of Directive 2000/78 requires to be laid down in national law when there is a finding of discrimination, must be effective, proportionate and dissuasive, regardless of whether there is an identifiable victim.102 For this reason, a ‘purely symbolic sanction cannot be regarded as being compatible with the correct and effective implementation of Directive 2000/78’,103 and Article 17 of the Directive precludes national rules by virtue of which it is possible only to impose a warning where a finding of discrimination on grounds of sexual orientation is made after the expiry of a limitation period of six months from the date on which

99 Case C-81/12 Asociaţia Accept v Consiliul Naţional pentru Combaterea Discriminării, ECLI:EU:C:2013:275. For an excellent analysis of the case, see U Belavusau ‘A Penalty Card for Homophobia from EU Non-Discrimination Law: Comment on Asociaţia Accept (C-81/12)’ (2015) 21 Columbia Journal of European Law 329. 100  Asociaţia Accept (n 99) para 36. 101  ibid, paras 46–53. 102  ibid, para 62. 103  ibid, para 64.

Discrimination on the Grounds of Sexual Orientation and Gender Identity  387   the facts occurred where, under those rules, such discrimination is not sanctioned under substantive and procedural conditions that render the sanction effective, proportionate and dissuasive.104 Finally, the Court made it clear that discrimination by assumption is also prohibited by the Directive: ie for discrimination on the ground of sexual orientation to be proved, it suffices if someone discriminates against a person because they think that he or she is gay, irrespective of whether the latter is actually gay or not. This is, clearly, a judgment which must be applauded, as the Court demonstrated that it is willing to adopt a broad, pragmatic, approach when applying the prohibition of discrimination on the ground of sexual orientation. The Court seems to have tried to achieve not merely formal equality but also substantive equality, by seeking to ensure that even potential obstacles to access to the employment market which are liable to occur as a result of the existence of discriminatory measures or practices are prohibited. In other words, the Court’s analysis appears to be no longer confined to a consideration of whether a specific person has been discriminated against when compared with another person similarly situated at a particular instance, but, more broadly, it considers whether a certain practice or action of a person or body is such as to create a discriminatory climate against a segment of the population which shares a characteristic which is a prohibited ground of discrimination under the Directive. In addition, and more specifically, the Court’s approach in Asociaţia Accept demonstrates a desire to be sensitive to the specific issues that emerge in situations involving LGB individuals, especially in Member States where homophobia is still (highly) prevalent. In particular, the fact that actions can be brought by NGOs in situations where there is no identifiable victim is especially important because it ensures that LGB individuals who are not ‘outed’ are not faced with the difficult dilemma of either refraining from bringing an action in order to be able to continue to hide their sexuality or to bring an action and in this way ‘come out’, at a time that they may not be ready or in circumstances that will make them suffer negative consequences. The next case that was considered during this period involved, again, an interpretation of Directive 2000/78, but its facts were quite similar to Maruko and Römer, in that it involved the rights of a same-sex couple, as opposed to those of LGB individuals. This was the case of Hay,105 where at issue was the refusal of a French bank to award one of its employees, Mr Hay, days of special leave and a bonus granted to staff who marry, following the conclusion by that employee of a civil solidarity pact with his male partner. Unlike in its judgments in Maruko and Römer where the Court merely provided guidelines as to how the comparability assessment between opposite-sex married couples and same-sex registered partners should be conducted, in this case it held that same-sex couples that entered into a Pacte Civil de Solidarité (PACS) were in a comparable position

104  105 

ibid, para 73. Hay (n 88).

388 

Alina Tryfonidou

with ­opposite-sex married couples, as regards benefits in terms of pay or working conditions.106 The Court then confirmed what it had held in Maruko and Römer, which is that discrimination against same-sex couples who have entered into a form of registered partnership (but not marriage, which is not available to them) amounts to direct discrimination on the ground of sexual orientation.107 In relation to this, the Court provided two further clarifications. Firstly, it noted that: The fact that the PACS, unlike the registered life partnership at issue in the cases which gave rise to the judgments in Maruko and Römer, is not restricted only to homosexual couples is irrelevant and, in particular, does not change the nature of the discrimination against homosexual couples who, unlike heterosexual couples, could not, on the date of the facts in the main proceedings, legally enter into marriage.108

Secondly, it noted that: The difference in treatment based on the employees’ marital status and not expressly on their sexual orientation is still direct discrimination because only persons of different sexes may marry and homosexual employees are therefore unable to meet the condition required for obtaining the benefit claimed.109

Therefore, Hay—in line with Asociaţia Accept—can be considered an example of a case which favoured LGB persons and same-sex couples, by interpreting the prohibition of discrimination on the ground of sexual orientation broadly, whilst appearing less hesitant to interfere with the actions of Member States—or, even, private employers—which are, clearly, discriminatory. The Court appears to be willing, now, itself to conduct the comparability assessment and to determine whether for a particular purpose, under national law, same-sex registered partners are similarly situated with opposite-sex married couples, something that will, undoubtedly, require Member States to extend to same-sex couples who have formalised their relationships entitlements which they might have otherwise preferred to bestow only on married couples. Accordingly, as a general conclusion, it can be said that although Directive 2000/78 does leave some (important) gaps in the protection of LGB persons from discrimination—its most significant weakness being its narrow material scope which is limited to employment-related issues—the way it has been interpreted by the Court in its recent (post-Lisbon) case law demonstrates that it can, indeed, play an important role in the improvement of the position of LGB persons under EU law by respecting their fundamental (human) right to equality.110

106 

ibid, paras 36–37. ibid, para 41. ibid, para 43. 109  ibid, para 44. 110  At the time of writing, a new preliminary reference in a case involving a claim that a rule under the Irish occupational scheme is in violation of, inter alia, the prohibition of discrimination on the ground of sexual orientation under Directive 2000/78 is pending before the ECJ—see Case C-443/15 Parris v Trinity College Dublin, Higher Education Authority, Department of Public Expenditure and Reform, Department of Education and Skills (pending). 107  108 

Discrimination on the Grounds of Sexual Orientation and Gender Identity  389  

(ii) A General Principle (and Fundamental Right) of Non-Discrimination on the Ground of Sexual Orientation? Advocate General Jääskinen in his Opinion in Römer pointed out that since the Court in Mangold111 had ruled that the prohibition of discrimination on the ground of age is a general principle of EU law, there was no reason why the same could not be ruled for the prohibition of discrimination on the ground of sexual orientation.112 The Court in its judgment in the same case was, nonetheless, more reticent in its approach as the Court’s words towards the end of its judgment provide a hint that it considers that there is a general principle of non-discrimination on the grounds prohibited by Directive 2000/78, but it stops short of explicitly saying so: [T]he Court has held that the Directive does not itself lay down the principle of equal treatment in the field of employment and occupation, which derives from various international instruments and from the constitutional traditions common to the Member States, but has the sole purpose of laying down, in that field, a general framework for combating discrimination on various grounds.113

Similarly in Hay, the Court noted that: [T]he purpose of Directive 2000/78 is to combat, as regards employment and occupation, certain types of discrimination, including discrimination on the ground of sexual orientation, with a view to putting into effect in the Member States the principle of equal treatment.114

In any event, the fact that the prohibition of discrimination on the ground of sexual orientation is a fundamental human right is made clear by Article 21 of the Charter, which makes explicit reference to sexual orientation as one of the grounds on which it is prohibited to base a difference in treatment. Accordingly, from a practical point of view, it does not matter that it has not been made entirely clear whether there is a general principle of non-discrimination on the ground of sexual orientation, as the EUCFR does provide for such a prohibition, unless the scope of the Charter (as laid down in its Article 51) is interpreted more restrictively than that of the general principles of EU law. The Court has, recently, had the chance to rule on Article 21 of the Charter in a case which involved discrimination on the ground of sexual orientation. This was in the case of Léger,115 which concerned the compatibility with EU law of the 111 

Mangold (n 10). Paras 129–31 of the Opinion of AG Jääskinen in Römer (n 88). For a similar view in the literature see E Howard, ‘EU Equality Law: Three Recent Developments’ (2011) 17 European Law Journal 785, 802. 113  Römer (n 88) para 59. 114  Hay (n 88) para 26. 115  Case C-528/13 Léger v Ministre des Affaires sociales, de la Santé et des Droits des femmes and Etablissement français du sang, ECLI:EU:C:2015:288. For comments, see A Tryfonidou, ‘The Léger Ruling as Another Example of the ECJ’s Disappointingly Reticent Approach to the Protection of the Rights of LGB Persons under EU Law’ (2016) 41 European Law Review 91. 112 

390 

Alina Tryfonidou

French blanket ban on blood donation by men who have had sex with other men (MSM). It was argued that this amounted to discrimination on the ground of sexual orientation since it is, essentially, gay and bisexual men that comprise this category of persons. The Court held that the Charter was applicable because the contested French legislation was implementing EU secondary legislation on the matter, and, thus, it was necessary to consider whether a permanent ban on blood donation by the MSM population is compatible with the fundamental rights protected under that instrument.116 The Court noted, in particular, that the French legislation must respect inter alia Article 21(1) [of the Charter] … according to which any discrimination based on sexual orientation must be prohibited. Article 21(1) is a particular expression of the principle of equal treatment, which is a general principle of EU law enshrined in Article 20 of the Charter.117

The Court found that the contested ban ‘may discriminate against homosexuals on grounds of sexual orientation’,118 since the legislation challenged ‘determines the deferral from blood donation on the basis to the homosexuality of the male donors who, on account of the fact that they have had homosexual sexual relations, are treated less favourably than male heterosexual persons’.119 It was, however, subsequently noted by the Court that the measure may be justified if it satisfies the conditions laid down by Article 52 of the Charter and, although it left it to the national court to make the final decision regarding this, the Court provided detailed guidelines for this purpose, pointing out, in particular, that the national court must examine whether there are measures which can achieve the same aim (a high level of human health protection) without requiring the imposition of a lifetime ban on blood donation by the MSM population.120 As explained elsewhere,121 the Court’s judgment is disappointing. This is due to the fact that although the Court found that the measure ‘may’ amount to discrimination on the ground of sexual orientation contrary to Article 21 of the Charter, it proceeded to consider whether it is justified, under Article 52 of the same instrument, although it should have ruled that the measure could under no circumstances be justified. Although there are, indeed, some gay and bisexual men that engage in risky sexual behaviour by engaging in unprotected sex with multiple partners, the same is the case for some heterosexual men and for some heterosexual, bisexual or lesbian women; accordingly, there is no good reason to single out gay and bisexual men as the only category of persons that must automatically and en bloc be subjected to a lifetime ban on blood donation. Moreover, the

116 

Léger (n 115) paras 45–47. ibid, para 48. 118  ibid, para 50. 119  ibid, para 49. 120  ibid, paras 51–69. 121  Tryfonidou (n 115). 117 

Discrimination on the Grounds of Sexual Orientation and Gender Identity  391   extent of the ban (it applies to all men who have had (even once in their life) sex with other men) and its duration (lifetime exclusion), coupled with evidence that medical advances have made it possible easily to detect a virus such as HIV after a certain period of time since it has been contracted, means that a lifetime exclusion of the MSM population from blood donation cannot be justified on medical grounds. Accordingly, the contested ban is unnecessary for ensuring the protection of public health and thus cannot be justified on medical grounds. Moreover, with its hands-off approach, leaving it to the national court to decide whether the measure is justified or not, the Court has, in effect, condoned a national measure which is based on deep-seated homophobia and on stereotypical misconceptions about the sexual behaviour of gay and bisexual men, who are caricatured as highly promiscuous individuals who engage in unprotected sex with multiple partners. The correct approach would, rather, be for the Court to rule that the measure can, under no circumstances, be justified. It is, of course, understood, that the main reason behind the Court’s reticent approach in the judgment is likely to have been its desire not to be seen as interfering in a matter which is sensitive from the point of view of the Member States, given that the contested measure touched on matters of public health and public morality, in relation to which a wide margin of appreciation is often left to the Member States. Yet, in situations where there is a clear breach of the fundamental (human) rights of a segment of the EU population—as was the case in Léger— wishing to keep the Member States satisfied by avoiding to interfere with their freedom in regulating sensitive matters, is not acceptable.

C. The (Near) Future—The Proposal for an Equality Directive As noted earlier, one of the main gaps left by Directive 2000/78 is its limited material scope which only covers employment-related issues. These gaps in protection will, nonetheless, be filled if the proposed Equality Directive122 becomes law. In particular, the proposed Directive will complement Directive 2000/78 by extending the prohibition of discrimination on the grounds covered by it beyond the employment field to all areas covered by Directive 2000/43. Moreover, it will make provision for a duty to be imposed on Member States to designate a body for the promotion of equal treatment irrespective of religion or belief, disability, age or sexual orientation, which is currently missing from Directive 2000/78, whilst it is provided in Directive 2000/43. However, whilst since 2008 there has been extensive discussion regarding the content of the proposed Directive (which, in some cases, has led to amendments to the original proposal), some Member States (especially Germany) are still strongly opposed to it.

122 Proposal for a Council Directive Implementing the Principle of Equal Treatment between ­Persons Irrespective of Religion or Belief, Disability, Age or Sexual Orientation COM (2008) 426.

392 

Alina Tryfonidou

One may ask why adopting this proposal would be important for LGB persons from a practical point of view; after all, discrimination on the ground of sexual orientation is, now, already prohibited for areas outside employment and vocational training by Article 21 of the Charter and, possibly, as a general principle of EU law, in situations that fall within the scope of EU law. However, since it is not entirely clear how the latter requirement should be interpreted (especially as regards the Charter), this leaves the possibility for a rather restrictive approach to its interpretation which could, for instance, allow the prohibition of discrimination on the ground of sexual orientation outside employment and vocational training to be applied only in situations involving a Member State implementing a Directive, as was the case in Léger. By adopting the proposed Directive, however, the EU will impose an obligation on all Member States to prohibit discrimination on the ground of, inter alia, sexual orientation outside employment and vocational training and, hence, it will ensure the uniform123 protection of LGB persons from discrimination in all Member States. Moreover, by requiring the promulgation of national legislation which will prohibit discrimination on the ground of sexual orientation in a wide range of areas, LGB individuals will be protected from discrimination in these areas, even in situations that are entirely unrelated to—and thus fall outside the scope of—EU law. Finally, since the proposed Directive—like Directive 2000/78—provides that the prohibition of discrimination will have to apply both in the public and the private domain, this will mean that LGB individuals will be able to rely on the implementing national legislation even against private individuals. Accordingly, it is important to ensure that the proposed Equality Directive becomes law soon, because it will significantly improve the position of LGB persons under EU law as regards their protection from discrimination and will reflect the nature of the prohibition of discrimination on the ground of sexual orientation as a genuine human right. Given that Commissioner Frans Timmermans— the portfolio of whom includes the protection of fundamental rights—has made the passing of the Directive one of his priorities, we do have reason to feel optimistic about the possibility of the proposed Directive becoming law in the next five years.124

123  As Mark Bell has noted: ‘A mapping study for the Commission published in 2006 found that although there was a wide range of legislation in the Member States on discrimination outside employment, this was often variable in its material scope and it was not always consistent in the range of discrimination grounds covered.’ M Bell, ‘Advancing EU Anti-Discrimination Law: The European Commission’s 2008 Proposal for a New Directive’ (2009) 3 Equal Rights Review 7, 16. 124  The recent European Parliament Resolution of 8 September 2015 on the situation of fundamental rights in the European Union (2013–14) (2014/2254(INI)) states that the European Parliament ‘[s]trongly deplores the fact that the Council has still not adopted the 2008 proposal for a directive on implementing the principle of equal treatment between persons irrespective of religion or belief, disability, age or sexual orientation; welcomes the prioritization of this directive by the Commission; reiterates its call to the Council to adopt the proposal as soon as possible’ (para 44 of the Resolution). Moreover, in December 2015, the Commission identified the 2008 proposal as one of the pending instruments that must become law (by 2019) in order to enhance the protection of LGB persons from

Discrimination on the Grounds of Sexual Orientation and Gender Identity  393  

VI. Conclusion The aim of this chapter was to consider the development of the principle of nondiscrimination against LGBT persons under EU law. It has been seen that although the general EU principle of equality had, originally, been used mostly instrumentally, in furtherance of the economic and social aims of the EU, in cases involving LGBT individuals, its role has been, ab initio, that of protecting the fundamental human rights of this segment of the EU population. Accordingly—and as reflected, in particular, in the Preamble to Directive 2000/78—the prohibition of discrimination on the ground of, inter alia, sexual orientation has been a specific enunciation of the fundamental human right to equality protected under EU law and under a number of international instruments. Similarly, the Court in P v S and Cornwall relied on its previous finding that the prohibition of discrimination on the ground of sex is, also, a fundamental human right under EU law, in order to base its decision to extend this prohibition to include discrimination on the ground of gender reassignment. However, the analysis of the legal instruments prohibiting discrimination on the ground of sexual orientation and the Court’s case law interpreting them demonstrates that although, on paper, the prohibition of discrimination on the ground of sexual orientation is a specific application of the fundamental (human) right to equality, in practice, LGB individuals do not, yet, enjoy a fundamental (human) right to be free from discrimination based on their sexual orientation. This is due to the confluence of a number of factors. Firstly, Directive 2000/78, which, at the moment, is the only instrument that requires Member States to make legislation to prohibit discrimination on the ground of sexual orientation in their own legal system, is limited to employmentrelated issues. This is, nonetheless, incongruent with a view that this prohibition is an aspect of the fundamental human right to equality: how can it be said that LGB persons have the fundamental (human) right to be free from discrimination on the ground of sexual orientation, when this is protected only in the area of employment? The solution to this appears to be the proposal—which is currently pending—for a new Equality Directive which, as we have seen, will require the Member States to make legislation prohibiting discrimination on the ground of sexual orientation in a wider range of areas of human life, in this way truly reflecting the fact that LGB persons enjoy under EU law the fundamental human right to be free from discrimination on the ground of their sexual orientation. Secondly, although the EUCFR, with its prohibition of discrimination on, inter alia, the ground of sexual orientation, makes an explicit statement that freedom from discrimination on this ground is a fundamental (human) right, nonetheless,

discrimination—see ‘List of actions by the Commission to advance LGBTI equality’ available at http:// ec.europa.eu/justice/discrimination/files/lgbti_actionlist_en.pdf.

394 

Alina Tryfonidou

the way that this prohibition has been applied by the Court to date—which has appeared quite hesitant in finding a breach of it in a case which involved, clearly, an unjustified breach of the principle—shows that, in practice, the prohibition is not treated as a reflection of a human right, since it can be side-stepped in situations where the Court fears that it may be accused of interfering with sensitive matters that are close to Member States’ hearts. Accordingly, although there has been some progress, in recent years, in the recognition and protection of the fundamental rights of LGB persons and, in particular, in their protection from discrimination, this should not obscure the fact that the protection afforded by EU law to this segment of the EU population is not yet satisfactory. On the other hand, by including discrimination on the ground of gender reassignment within the prohibition of discrimination on the ground of sex, transsexual persons have fared better in terms of the protection from discrimination they receive under EU law, since they can take advantage of the full panoply of EU sex equality legislation and the prohibition of discrimination on the ground of sex under the Treaties and the Charter. However, the Court has—to date— only protected trans persons who have undergone a gender reassignment surgery, and, thus, it is unclear whether under EU law trans persons who have not undergone—and do not intend to undergo—gender reassignment surgery receive any protection. Moreover, the complete absence of a reference to trans persons and/or to discrimination on the ground of gender identity in the Treaties and the vast majority of secondary legislation is problematic from both a practical and a symbolic perspective, as it perpetuates their invisibility and creates the possibility that the specific concerns of this segment of the EU population will not be taken into account, especially in situations which require substantive (as opposed to formal) equality.125

125  The European Parliament Committee on Civil Liberties, Justice and Home Affairs has recently prepared a Report on the EU Roadmap Against Homophobia and Discrimination on Grounds of Sexual Orientation and Gender Identity (2013/2183(INI)) whereby it was suggested that the various legal instruments which seek to promote equality and to prohibit discrimination should make explicit reference to trans persons and discrimination on the ground of gender identity.

Concluding Remarks

396 

22 Judging General Principles DAVID EDWARD It seems a long time ago since we took part in the conference that led to this book. The outcome of the Brexit referendum has brought a certain unreality to discussion of legal principles that lawyers in the United Kingdom may soon feel free to ignore. ‘We want our country back’ was the cry, and British judges may soon feel free to echo the sentiments of the scholarly (though Etonian) Scottish judge Lord Hailes, when faced with an argument based on English law at the trial of Deacon Brodie:1 By the articles of the Union, our own laws and forms of procedure are secured to us, and we have as little connection with those of England as with the laws of Japan, being as little bound to obey them.2

There is already a loss of interest amongst British students in the study of EU law and, amongst students from other European countries, in the opportunities that are offered by our law schools. Why, then, should we continue to be interested in the legal system of the European Union? The realist will point out that, for at least the next two years, the UK will remain a Member State to which EU law applies and that, even after final Brexit, British companies will have to abide by the rules of the single market if they wish to trade there and will, on any view, be subject to the competition jurisdiction of the European Commission.3 The academic may also point out that a legal system that has attracted so much interest amongst lawyers in this country is still worth studying. Professor Henry Schermers used to say that, whereas French lawyers came to European law through public international law and German lawyers did so through constitutional law, British lawyers came to it through comparative law and were interested as much in its relevance for private rights as for those of states and institutions. This can be seen in the kind of issues that have been referred to the Court of Justice by British courts—most notably, perhaps, in the Factortame saga.4 1  A worthy Edinburgh citizen who varied his daytime occupation as cabinet-maker and locksmith with a little night-time burglary. 2  W Roughead, Trial of Deacon Brodie (Glasgow & Edinburgh, Hodge & Co, 1906) 130. 3  Case 48/69 Imperial Chemical Industries Ltd v Commission (‘Dyestuffs’), ECLI:EU:C:1972:70. 4  Case C-213/89 The Queen v Secretary of State for Transport ex parte Factortame Ltd & others, ECLI:EU:C:1990:257, and Joined Cases C-46/93 and C-48/93 Brasserie du Pêcheur SA v Bundesrepublik Deutschland and The Queen v Secretary of State for Transport ex parte Factortame Ltd and others, ECLI:EU:C:1996:79.

398 

David Edward

While the details of these cases may be peculiar to EU law, many of the underlying problems are not. They are common to all attempts to regulate rights and obligations in a complex legal environment that transcends the boundaries of any single national system. I came to EEC law, first, through its effects on the legal profession, then as an advocate, and only later as an academic and judge. So I tend to approach the study of the Court’s case law from what is essentially the viewpoint of a practitioner. Other contributors have discussed the problem of general principles from a variety of points of view. I will look at it from the point of view of a former judge of the Court of Justice bred in the common law tradition, addressing the following points: —— —— —— —— —— —— ——

What are the essential features of the judicial function? How and where are EU judges to find the law? General principles as a source of law. The problem of legitimacy. The search for principles in EU law. The problem of translation, and Van Gend en Loos. Proportionality.

I.  What Are the Essential Features of the Judicial Function? The life of a judge, even in the European courts, is more pedestrian than some of the contributions to this book might suggest. In his Maccabaean Lecture on ‘The Search for Principle’ Lord Goff of Chieveley contrasted the respective roles of the judge and the jurist: The primary function of judges is not the formulation of legal principles. Their main task, more workaday, more humdrum, is to try cases. … For jurists, on the other hand, the formulation of legal principles is one of their main functions.5

The judge has to decide cases, ideally with the minimum of delay, for ‘the courts are neither a debating club nor an advisory bureau’.6 The special characteristics of judicial work can, I think, be summed up in six points.

5  Maccabaean Lecture in Jurisprudence (1983) lxix Proceedings of the British Academy 169ff, accessible at www.britac.ac.uk/pubs/proc/files/69p169.pdf. 6  Lord Justice Clerk Thomson in Macnaughton v Macnaughton’s Trustees 1953 SC 387, 392. Compare Case C-149/82 Robards v Insurance Officer, ECLI:EU:C:1983:26: ‘The task assigned to the Court by Article 177 of the EEC Treaty is not that of delivering opinions on general or hypothetical questions but of assisting in the administration of justice in the Member States.’

Judging General Principles

 399

First, judges cannot choose the cases they have to judge. It is up to the parties and those who advise them to decide what cases come to court. Judges who long to pronounce on their favourite topic, to condemn some legislative abomination or to expose a judicial or academic heresy, may never have the opportunity to do so, save in the evanescent context of the lecture hall or perhaps the Festschrift. For them, the day for writing the judgment that makes legal history may never come. Second, a judge cannot refuse to take a case because it is difficult, disagreeable or unpopular. In many cases, we know from the outset that, whatever the outcome, there will be outrage from some quarter. Someone will be ‘devastated’ and the judge will be pilloried. Even in the more rarified atmosphere of EU case law, the Daily Telegraph and the academic commentator lie ready to pounce. Third, once seised of a case, unless it is settled or withdrawn, the judge must give judgment. The French Civil Code (the Code Napoléon) is very explicit on this point. Article 4 provides that: The judge who refuses to judge on the pretext of the silence, the obscurity or the inadequacy of the law shall be prosecuted for denial of justice.7

The judge must ‘find’ the law somewhere, and success will depend on the extent to which the legal system offers a well-stocked storehouse of material where the law can be found. Fourth, the process of judging must be conducted within a more or less tightly defined legal and procedural framework. As Dicey put it: ‘The primary duty of a judge is to act in accordance with the strict rules of law.’8 In the case of the Luxembourg courts, the procedural straitjacket is tight. By contrast, the Strasbourg Court is master of its own Rules of Procedure. Within the general scheme set out in the ECHR,9 the Plenary Court lays down the Court’s Rules of Procedure.10 The extent of this power can be judged by a study of the Rules.11 A request by the Court of Justice to be given comparable procedural autonomy when Treaty revision was under consideration was, to my personal knowledge, refused (or simply ignored). This severely curtails the scope for innovation and sometimes common sense. Fifth, the duty to judge carries with it the duty to reason. But there are cultural differences as to how judgments should be reasoned. The discursive style of common law judgments and the Opinions of Advocates General is in stark contrast to the

7  ‘Le juge qui refusera de juger, sous prétexte du silence, de l’obscurité ou de l’insuffisance de la loi, pourra être poursuivi comme coupable de déni de justice.’ 8 Dicey, The Law of the Constitution, 8th edn, Introduction, xxxix. 9  Arts 19–51 ECHR. 10  Art 25(d) ECHR. 11  Rules of Procedure, last amended 1 January 2016 (available at www.echr.coe.int/Documents/ Rules_Court_ENG.pdf).

400 

David Edward

laconic and impersonal style of judgments in many (perhaps most) other European countries and largely followed in the Luxembourg courts. When I presented my first draft judgment for consideration by the Chamber, one of my colleagues said: ‘That is a good opinion, now we must make it aseptic.’ The latitude allowed, for rather particular reasons, to Judge Joliet as Rapporteur in the Ideal Standard case was unusual.12 The requirement (unique to Luxembourg) to produce judgments in more than twenty languages adds a further complication, since the legal terminology of one language may have no direct equivalent in another. Even in Strasbourg, where there are only two official languages, there are notable examples of potential misunderstanding.13 I discuss the problem of translation further below. Sixth, having delivered judgment, the judge cannot, except in the most exceptional circumstances, recall, revise, modify or seek to explain the judgment. The judge can only say, with Pontius Pilate, ‘What I have written I have written’,14 and leave it at that. This creates particular difficulties for judges of the Court of Justice when they are called upon to explain or justify judgments with which, personally, they disagree. In case of disagreement the majority view prevails and the Judges are bound by their solemn undertaking to preserve the secrets of the Court’s deliberations. Cumulatively, these features of the judicial function differentiate the judicial power from the legislative power and the executive power. Legislators, ministers and administrators enjoy (albeit within limits) a range of choice, and a discretion to act or not to act, that judges do not. The corollary of the judges’ power is the strictness of the rules that govern what they do and how they must do it.

II.  How and Where Are EU Judges to Find the Law? How and where are the judges in Luxembourg to find the law that they and, under the reference procedure, the national judges are to apply? The direction from the Treaty-makers is that the Court ‘shall ensure that in the interpretation and application of the Treaties the law is observed’.15 The use of the word law in English conceals the crucial distinction in (for example) Latin, French and German between lex, loi or Gesetz on the one hand, and ius, droit or Recht on

12  Case C-9/93 IHT Internationale Heiztechnik GmbH and Uwe Danzinger v Ideal-Standard GmbH and Wabco Standard GmbH, ECLI:EU:C:1994:261. 13  See, eg, the judgment in Delcourt v Belgium, 2689/65, [1970] ECHR 1 (17 January 1970), concerning the presence of the Avocat général at the deliberations of the Cour de Cassation. The French text refers to ‘un membre du ministère public’, while the English text refers to a ‘member of the Procureur général’s department’—hardly an appropriate description of the Avocat général, even if true in a sense. 14  John 18.22. 15  Now Art 19(1) TEU, originally in Arts 31 ECSC, 164 EEC and 136 EAEC (emphasis added).

Judging General Principles

 401

the other: the distinction between what has been enacted by the legislator and, more generally and in a more abstract sense, what is ‘right’ and forms part of the whole corpus of law. The Treaty-makers’ direction could have been limited to requiring the Court to ensure compliance with the provisions of the Treaty, as interpreted according to the established canons of the law of treaties—possibly by a provision such as Article 38 of the Statute of the International Court of Justice with its reference to ‘the general principles of law recognized by civilized nations’.16 But the Treaty-makers’ direction was not so limited: the Court is instructed to draw on the whole corpus of law and legal science as the context within which the Treaty is to be interpreted and applied. Viewed in that way, the Treaty provision that the non-contractual liability of the Union is to be assessed ‘in accordance with the general principles common to the laws of the Member States’17 is not a lex specialis, but simply makes explicit in a particular context what is already part of the general scheme of the Treaty.18

III.  General Principles as a Source of Law The idea that the law includes general principles has a long history. For example, we find in Justinian’s Digest the general presumption in favour of freedom; the principle that there can be no obligation to do the impossible; and the equitable principle (iure naturae aequum est) that no-one should be enriched to the detriment of another.19 Indeed, principles of law, neatly encapsulated in Latin maxims, used to be part of the lawyer’s stock in trade. The distinction between reasoning from principle or from precedent is often cited as the basic difference between the civil law and common law systems. But the common lawyer will often seek to extract a principle from a series of precedents. In a rather similar way, the French magistrat, who may not create law,20 will seek to extract (dégager) a rule (norme) from the existing body of law. By contrast,

16 

See Professor Redgwell’s chapter on ‘General Principles of International Law’ in this volume. Art 340 TFEU, originally Arts 215(2) EEC and 188(2) EAEC. The ECSC Treaty dealt with non-contractual liability in different terms (see Art 40 ECSC). As far as my researches have taken me, the words of the current Treaties appear for the first time in the draft of Part Six of the EEC Treaty, which was submitted by a Working Group including Pierre Pescatore, later a Judge of the Court, just over a month before the Treaty was signed—see Avant-Projet de dispositions générales préparé par le Groupe de Travail Constituée de Messieurs Devadder et Pescatore, Bruxelles, le 13 février 1957, in R Schulze and T Hoeren (eds), Dokumente zum Europäischen Recht (Berlin, Springer Verlag, 1999) Band 1, Gründungsverträge, 1178–84. I can find no explanation for the choice of words or any evidence that it was discussed before incorporation in the Treaty. 19  D.50.17.122, 185 and 206. 20  Art 5 of the French Code civil provides that ‘Il est défendu aux juges de prononcer par voie de disposition générale et réglementaire sur les causes qui leur sont soumises.’ 17  18 

402 

David Edward

since there is no Code of administrative law, the French Conseil d’Etat has developed a body of principes généraux du droit.21 In France, recourse to general principles has been criticised as a covert method of judicial legislation in violation of the principle laid down in Article 5 of the French Code civil22—a criticism that is reflected in some of the contributions to this book. But the Conseil constitutionnel has affirmed that general principles are a legitimate source of law—for example in asserting the obligation of the administration to ‘respect the general principles of law and notably the rights of defence’23—leading Professor Georges Vedel to say that: [L]e Conseil constitutionnel reçoit les principes généraux du droit en tant que normes de niveau législatif, constituant une source autonome du droit administratif, distinct du droit écrit, manifestant le pouvoir normative du juge.24

In German administrative law, the fundamental rights guaranteed by Article 3 of the Basic Law are binding on the administration, as are the general principles of proportionality, equality and respect for legitimate expectations.25 In the context of the ECSC Treaty, the question arose whether the High Authority was bound by Article 14 of the German Basic Law guaranteeing private property. Advocate General Lagrange said: It is not for the Court, whose function is to judge the legality of the authorizations, to apply, or at least to do so directly, rules of national law, even constitutional rules, in force in one or other of the Member States (judgment of 4 February 1959 in Stork v High Authority). It may only allow itself to be influenced by such rules in so far as, where appropriate, it may see in them the expression of a general principle of law which may be taken into consideration in applying the Treaty. While it may certainly be admitted that the protection of the right to property, including the remedies which must be available against any infringement of that right, such as expropriation, is a rule of law common to the six countries, it is certain beyond doubt that the present case is not one of such a kind. There is no infringement of property rights, even understood widely, of which the High Authority may have been guilty.

In its judgment, the Court declared, without further discussion, that: ‘Community law, as it arises under the ECSC Treaty, does not contain any general principle, express or otherwise, guaranteeing the maintenance of vested rights.’26 21 For a brief description, see http://fr.jurispedia.org/index.php/Principes_g%C3%A9n%C3% A9raux_du_droit_(fr). 22  See n 7 above. 23  Decision No 90-287 DC (ECLI:FR:CC:1991:90.287.DC) para 28. 24  For this citation, and a more general discussion, see D Menna, La théorie des principes généraux du droit à l’épreuve de la jurisprudence constitutionnelle in Le droit administrative en mutation (Paris, PUF, 1993) 201ff (accessible online at www.u-picardie.fr/curapp-revues/root/31/domenico_menna. pdf_4a07e0101bb41/domenico_menna.pdf 25  See, for a detailed discussion, G Nolte, ‘General Principles of German and European Administrative Law—A Comparison in Historical Perspective’ (1994) 57 MLR 191ff. 26  Joined Cases 36, 37, 38/59 and 40/59 Geitling v High Authority, ECLI:EU:C:1960:36, 438–9 and 450 (emphasis added).

Judging General Principles

 403

In Transocean Marine Paint, Advocate General Warner addressed the question directly in relation to the principle of the right to be heard (audi alteram partem). Having reviewed the law of the Member States, he concluded: [T]hat review … of the laws of the Member States must, I think, on balance, lead to the conclusion that the right to be heard forms part of those rights which ‘the law’ referred to in Article 164 of the Treaty upholds, and of which, accordingly, it is the duty of this Court to ensure the observance.

The Court’s practice of reasoning from general principles of law can thus be seen to have been authorised by the Treaties and to have a long and respectable jurisprudential ancestry. Nevertheless, as some of the contributions to this book suggest, resort to general principles may raise a problem of legitimacy.

IV.  The Problem of Legitimacy The issue of legitimacy arises when the appeal to general principles in a particular case can be thought to amount to judicial legislation, usurping the role of the legislator. But even here, there is room for argument. One of Dicey’s lectures in Law & Public Opinion is, surprisingly, entitled ‘Judicial Legislation’.27 He begins with the claim that: As all lawyers are aware, a large part and, as many would add, the best part of the law of England is judge-made law.

Some pages later, he says: Judicial legislation aims to a far greater extent than do enactments passed by Parliament, at the maintenance of the logic or the symmetry of the law. The main employment of a Court is the application of well-known legal principles to the solution of given cases and the deduction from these principles of their fair logical result. Men trained in and for this kind of employment acquire a logical conscience; they come to care greatly—in some cases excessively—for consistency.

The search for the ‘fair logical result’ was taken up by Lord Goff in his lecture mentioned above: If I were asked what is the most potent influence upon a court in formulating a statement of legal principle, I would answer that in the generality of instances it is the desired result in the particular case before the court. But let me not be misunderstood. When we talk about the desired result, or the merits, of any particular case, we can do so at more than one level. There is the crude, purely factual level—the plaintiff is a poor widow who has lost her money, and such like. At another level there is the gut reaction, often

27  AV Dicey, Lectures on the Relation between Law and Public Opinion in England during the Nineteenth Century (London, Macmillan, 1926) Lecture XI, 361ff.

404 

David Edward

most influential. But there is a more sophisticated, lawyerly level, which consists of the perception of the just solution in legal terms, satisfying both gut and intellect. It is in the formulation, if necessary the adaptation, of legal principle to embrace that just solution that we can see not only the beneficial influence of facts upon the law, but also the useful impact of practical experience upon the work of practising lawyers in the development of legal principles. [T]he prime influence upon jurists is not so much facts as ideas; and just as fragmentation presents a danger for practising lawyers, who tend to adopt an unsystematic approach to their work, so jurists are subject to danger from preconceived ideas, and may regard too inhospitably a judicial decision which does not accord with their own preconceptions.28

Justice Benjamin Cardozo put the point in this way: We go forward with our logic, with our analogies, with our philosophies, till we reach a certain point. At first, we have no trouble with the paths; they follow the same lines. Then they begin to diverge, and we must make a choice between them. History or custom or social utility or some compelling sentiment of justice or sometimes perhaps a semiintuitive apprehension of the pervading spirit of our law must come to the rescue of the anxious judge, and tell him where to go.29

Lord Goff warned against two pitfalls in the judge’s search for principle: The first I call the temptation of elegance. This is a temptation which can attract us all, simply because a solution, if elegant, automatically carries a degree of credibility; and yet the law has to reflect life in all its untidy complexity, and we have constantly to be on our guard against stating principles in terms which do not allow for the possibility of qualifications or exceptions as yet unperceived. The second is the fallacy of the instant, complete solution. It is understandable that judges and jurists should from time to time believe that they see the complete answer to a particular problem; indeed, without a measure of self-confidence, no judge is competent to perform his duties. But it must never be forgotten, not only that all law is in a continuous state of development, but also that too strong a conviction of the correctness of one’s own analysis may blind one to its imperfections. Humility is perhaps too much to ask of judges; but a reasonable degree of modesty, or at least of diffidence, should be part of the judicial job specification.

V.  The Search for Principles in EU Law In a sense, the EU Court of Justice is over-supplied with sources from which to distil general principles of law. As well as the Treaties with their multifarious statements of values, aims and objectives, there are the national constitutions, the

28  29 

See n 5 above. BN Cardozo, The Nature of the Judicial Process (New Haven, CT, Yale University Press, 1921) 43.

Judging General Principles

 405

jurisprudence of national courts, the writings of jurists and commentators, and a range of international conventions and covenants.30 From these sources, the Court has distilled a variety of ‘general principles’. The principle of unjust enrichment, set out in the Digest,31 underlies a large and growing corpus of decisions on the obligation to repay taxes and other levies unlawfully exacted (la répétition de l’indu).32 Lord Goff, in his Maccabaean lecture, discusses the influence of the same principle on the previously firm rule of English law as to the irrecoverability of money paid under mistake of law. In the same way, the presumption in favour of freedom enunciated in the Digest33 seems to me to underlie the principle of proportionality discussed below. Other principles invoked by the Court are not general principles of law in a universal sense, but rather ‘principles of EU law’, developed in the particular context of that law. For example, the much-criticised judgment in Mangold34 should, I think, be read as deducing the principle of non-discrimination on grounds of age, as a principle of EU law, from the language of the EU legislature itself in the Recitals and Article 1 of the Directive under consideration.35 Read in this way, the judgment, far from being an example of improper or over-reaching judicial legislation, is one of the judiciary interpreting and applying the expressed intention of the legislature. That leads me to remark that some criticism of the Court’s judgments fails to take full account of the subtleties of language with which the Court is beset.

VI.  The Problem of Language A very simple example of this problem was illustrated by one of the speakers at the conference who criticised the expression ‘It must be observed that …’ which introduces so many paragraphs in the English texts. This was said to show the arrogance of intellectual certainty. In reality, the words are a literal translation of the French ‘Il y a lieu d’observer que …’, which the French judge, Jean-Pierre Puissochet, deplored as totally unnecessary since it amounts to little more than a clearing of the throat before passing to a new point. Nowadays, these and similar phrases have largely disappeared from the English texts. (The judgment in Mangold is a good example.) A more fundamental misunderstanding due to translation can be illustrated by commentaries on the judgment in Van Gend en Loos,36 which Professor Weatherill, 30 

See Case C-144/04 Mangold v Hain, ECLI:EU:2005:420, paras 74–75.

32 

Case 199/82 Amministrazione delle Finanze dello Stato v San Georgio, ECLI:EU:C:1983:318.

31 D.50.17.206. 33 D.50.17.122. 34 

n 30 above. Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in employment and occupation [2000] OJ L303, 16–22. 36  Case 26/62 NV Algemene Transport- en Expeditie Onderneming van Gend en Loos v Netherlands Inland Revenue Administration, ECLI:EU:C:1963:1. 35 

406 

David Edward

in his chapter in this book, characterises as demonstrating ‘astonishing chutzpah’ and ‘sheer audacity’. Let us begin by noting that the question with which the Court was faced was not one of its own invention. The mandatory nature of Article 12 EEC had been settled in the Gingerbread case earlier in the year: It follows … from the clarity, certainty and unrestricted scope of Articles 9 and 12, from the general scheme of their provisions and of the treaty as a whole, that the prohibition of new customs duties, linked with the principles of the free movement of products, constitutes an essential rule and that in consequence any exception, which moreover is to be narrowly interpreted, must be clearly stipulated.37

The question in Van Gend en Loos was whether the ‘clear, certain and unrestricted’ provisions of Article 12 could be ‘directly’ enforced by affected individuals in the national courts. There was no doubt that provisions of a treaty could be ‘selfexecuting’ and have ‘direct’ or ‘immediate’ (unmediated) effect.38 The issue was whether provisions of Article 12 were so. This had become a live constitutional issue in the Netherlands. The Court of Justice, to which the issue had been referred by a Dutch court, was bound to decide it one way or the other, chutzpah apart. Unfortunately, the English text of the judgment obscures the careful reasoning of the judgment. At that time, the format of the Court’s judgments in French marked the passage from one point to another in this way: Attendu que ….

Que ….



Que ….

Attendu que ….

Que ….



Que ….

In the French text of Van Gend en Loos, the passage under ‘B—Quant au fond’ begins: Attendu que la Tariefcommissie pose en premier lieu la question de savoir si l’article 12 du traité a un effet immédiat en droit interne, dans le sens que les ressortissants des États membres pourraient faire valoir sur le base de cet article des droits que le juge national doit sauvegarder;

37 

Joined Cases 2/62 & 3/62 Commission v Luxembourg and Belgium, ECLI:EU:C:1962:45. The terminology was far from settled. See my contribution to the Festchrift for Judge Mancini: ‘Direct Effect, the Separation of Powers and the Judicial Enforcement of Obligations’ in Scritti in onore di Giuseppe Federico Mancini (Milan, Giuffrè, 1998). See also ‘Direct Effect: Myth, Mess or Mystery’ in JM Prinssen and A Schrauwen (eds), Direct Effect: Rethinking a Classic of EC Legal Doctrine (Amsterdam, Europa Law Publishing, 2002). 38 

Judging General Principles

 407

Attendu que pour savoir si les dispositions d’un traité international ont une telle portée il faut en envisager l’esprit, l’économie et les termes.39

There then follow three passages, each introduced by ‘Attendu que…’, dealing respectively with l’objectif du traité, l’économie du traité and le texte de l’article 12. These are followed by passages dealing with the need for national implementing legislation and with the argument that enforcement of Treaty obligations was a matter for the Commission under Articles 169 and 170 EEC. The process of reasoning—the same as in the Gingerbread case quoted above— is lost in the English translation which does not mark the passage from one point to another, so that the judgment appears to consist of a random succession of ex cathedra pronouncements.40 Properly understood, so far from being a judicial coup d’état, the judgment in Van Gend en Loos was a carefully reasoned response to a question posed by the Dutch court, which in turn reflected a live debate in the Netherlands. Anyway, as we now know, the outcome was decided by a narrow majority of four to three—a bad start for a putsch.41

VII. Proportionality The principle of proportionality occupied a good deal of the discussion at the conference and consequently of the contributions to this book. As a general principle of law, I believe it stems from the presumption in favour of freedom which, as noted above, is found in the Digest.42 Essentially, the question is how far the authorities of the state are entitled to restrict, or interfere with, the personal autonomy of the individual. As Paul Craig illustrates in this volume, the same concern was developed by common law judges over the centuries and it is wrong

39  The correctness of this approach was confirmed in Art 31(1) of the Vienna Convention on the Law of Treaties (1969). 40  Professor Weatherill describes the Court’s reference to the Economic and Social Committee, ‘a body pitifully devoid of profile or legitimacy’, as ‘shallow, feeble and opportunistic’. This reflects a peculiarly British view of society. The idea that there should be an advisory body representing the ‘social partners’ goes back in France to the creation of the Conseil national économique by the government of Edouard Herriot in 1924. If it be true that the EESC is now ‘pitifully devoid of profile or legitimacy’, this is due to the plethora of think-tanks, lobbyists, special advisers and researchers who throng the corridors of Brussels. In response to a proposal to abolish the EESC, its President remarked that: ‘It’s very strange that Liberals, who also ask for transparency and for the development of society, would try to discuss the idea of cancelling the only body that is for people who are not politicians.’ 41  ‘Interview with Pierre Pescatore: The Early Judgments of the Court of Justice (1962–1966)’ accessible at www.cvce.eu/en/histoire-orale/unit-content/-/unit/7b44d3a7-ea26-432f-a826-185ff0cc353c/ f5555be1-2043-4be9-93f9-5e32523c2906. See also S Gori, ‘Souvenirs d’un survivant’ in Van Gend en Loos 1963–2013 (Court of Justice, 2013) 29ff. 42 D.50.17.122.

408 

David Edward

to suggest that the underlying idea was unknown to British law before our accession to the EEC. I would suggest that the contribution of EU law, as developed by the Court of Justice on the basis of German administrative law, has not been to formulate a new conception of proportionality as such, but rather to develop a step-by-step process of enquiry in order to impose what might be called a ‘discipline of reasoning’. For example, in Gebhard, the Court said: National measures liable to hinder or make less attractive the exercise of fundamental freedoms guaranteed by the Treaty must fulfil four conditions: they must be applied in a non-discriminatory manner; they must be justified by imperative requirements in the general interest; they must be suitable for securing the attainment of the objective which they pursue; and they must not go beyond what is necessary in order to attain it.43

Similar statements of the required steps of enquiry can be seen in the judgment of the UK Supreme Court in Christian Institute v Lord Advocate. In that case, an Act of the Scottish Parliament44 was challenged on the ground, amongst others, that it was incompatible with Article 8 of the European Convention on Human Rights. (The Act provides for the appointment, for every child, of a ‘named person’ who would be responsible in various ways for that child’s well-being.) The Court said: In our view these challenges raise the following four questions: (i) what are the interests which art 8 of ECHR protects in this context, (ii) whether and in what respects the operation of the Act interferes with the art. 8 rights of parents or of children and young people, (iii) whether the interference is in accordance with the law, and (iv) whether that interference is proportionate, having regard to the legitimate aim pursued.

Coming to the fourth question, the Court said: It is now the standard approach of this court to address the following four questions when it considers the question of proportionality: (i)

whether the objective is sufficiently important to justify the limitation of a protected right, (ii) whether the measure is rationally connected to the objective, (iii) whether a less intrusive measure could have been used without unacceptably compromising the achievement of that objective, and (iv) whether, balancing the severity of the measure’s effects on the rights of the persons to whom it applies against the importance of the objective, to the extent that the measure will contribute to its achievement, the former outweighs the latter (ie whether the impact of the rights infringement is disproportionate to the likely benefit of the impugned measure).45

I find this a good illustration of Lord Goff ’s distinction between the role of the judge (to try cases) and that of the jurist (to formulate legal principles). The 43  Case 55/94 Gebhard v Consiglio dell’Ordine degli Avvocati e Procuratori di Milano, ECLI:EU:C: 1995:411, point 6 of the dispositif. 44  Children and Young People (Scotland) Act 2014. 45  Christian Institute v Lord Advocate [2016] UKSC 51, paras 70 and 90.

Judging General Principles

 409

step-by-step discipline of enquiry and reasoning helps the judge to ask the right questions in the right order. It is more difficult for the jurist to formulate the underlying principle, but as Lord Goff said: I see the law not so much as Maitland’s seamless web, but as a mosaic, and a mosaic that is kaleidoscopic in the sense that it is in a constant state of change in minute particulars. The legislature apart, it is the judges who manufacture the tiny pieces of which the mosaic is formed, influenced very largely by their informed and experienced reactions to the facts of cases. The jurists assess the quality of each piece so produced; they consider its place in the whole, and its likely effect in stimulating the production of new pieces, and the readjustment of others. In this their approach is certainly broader, perhaps more fundamental, and also more philosophical than that of the judges.

Judging proportionality is rather like judging negligence. We know in broad terms what is the principle to be applied, but as Lord Macmillan said in Donoghue v Stevenson: In the daily contacts of social and business life human beings are thrown into, or place themselves in, an infinite variety of relations with their fellows; and the law can refer only to the standards of the reasonable man in order to determine whether any particular relation gives rise to a duty to take care as between those who stand in that relation to each other. The grounds of action may be as various and manifold as human errancy; and the conception of legal responsibility may develop in adaptation to altering social conditions and standards. The criterion of judgment must adjust and adapt itself to the changing circumstances of life. The categories of negligence are never closed.46

The essential point, surely, is that what Lord Bingham called ‘the business of judging’ is an intensely human activity. What one judge thinks is the right, or even self-evident, general principle to be applied may appear quite the opposite to another. There is no infallible yardstick. In Luxembourg, the business of judging leads to a single judgment reached after deliberation between the judges in camera. It deprives us of the illuminating dissenting judgment, though also, with all due respect, of much unnecessary verbiage. I can only say that, in my own experience, the Luxembourg system conduces to mutual respect, a willingness to learn from the experience and knowledge of others and to understand the nuances of cultural difference, and, ultimately (I hope), in Lord Goff ’s words, to ‘a reasonable degree of modesty, or at least of diffidence’.

46 

Donoghue v Stevenson [1932] UKHL 100, 24; [1932] AC 562; 1932 SC (HL) 31, 70.

410 

INDEX

adjudication, 57–59, 152, 192 fairness, 331–32, 335–39 EU, 345–49 US contract law, 339–45 anti-discrimination law: age, 265 equal treatment, 258–59 general principle of non-discrimination, 258 obesity, 265 secondary legislation, 258–59 sickness, 265 treaty law, 258 Aquinas, Thomas, 277, 280–81 arbitration proceedings, 263–63 unfair arbitration clauses, 341–42 Aristotle, 74, 277, 280–81 autonomy, see private autonomy Berlin, Isaiah, 271–73, 278, 289, 294, 296 Brexit, 47, 397–98 Calvin, John, 279, 280–84, 289 will theory, 284–85 capacity, 26 private autonomy, 256 Catholic Counter Reformation, 278–79 Charter of Fundamental Rights (EUCFR), 2, 27–28, 267–68 binding nature, 266, 384 coherence of general principles, 80–83, 86–90 discrimination based on age, 266 discrimination on ground of sexual orientation, 385, 389, 393–94 general principles, 266–67 good administration, 86–90 harmonisation and re-regulation, 261 Mangold case and, 266 proportionality trends: impact of EUCFR, 248 protection of the weaker party, 257 comparative law, 3 CJEU method, 111, 120, 125, 128 competence allocation: legality and, 71–72 coherence of general principles, 41–43, 44–45, 47 creative coherence, 79–80, 102–03

functions: creative coherence, 79–80 normative coherence, 80–83 transnational coherence, 84–86 good administration and: EUCFR, 86–88 Art. 41, 88–90 inter-legality, 92–93 jurisprudential coherence, 90–91 national procedural autonomy, 91–92 transnational coherence, 84–86 human rights and, 47–48, 77–78 ideological coherence, 93, 102–03 Opinion 2/13, 93–102 normative coherence: applicative process of general principles, as, 80–83 ECHR, 83 EUCFR, 80–83 Opinion 2/13, 80 Opinion 2/13: ideological coherence, 93–102 normative coherence, 80 policy-based coherence, 78 substantive coherence, 78–79 transnational coherence: Akerberg case, 84 inter-legality of EU law, 84 principle of good administration, 84–86 spillover to national law, 84–86 comitology, 47 constitutionalisation of EU law, 30–31, 35, 55, 67–68, 118 constitutional legality, 73–74, 75 consumer contracts: cognitive and behavioural biases, 320 information available, 313–15, 319–20, 326 protection of the weaker party, 256, 261–62 substantive fairness and, 340–41 contract law, 304–05 20th century developments, 293 ethics and social law, 294 freedom of contract, 291–92 Enlightenment theories, 292–93 negative freedom, 300 shift to personal autonomy, 300–02

412 

Index

private autonomy and, 255–56, 274–76 early modern debates, 278–79 Stair, 279–80 conception of liberty, 285–89 engagement, 290–91 late scholastics and, 280–84 will, desire, resolution and engagement distinguished, 290 Spanish scholastics, 281–82 tension thesis, 298–99 personal autonomy ideal compared, 301–02 will theory, 276–78 Calvinist will theory, 284–85 northern natural law will theories, 279–80 see also protection of weaker party; tension thesis contractual obligations: bilateral nature, 297–98 breach, 361 voluntary nature, 297 Costa v ENEL, 22, 37 interpretation of EU treaties and, 32–33 legality, 67 object and purpose, 32–33 primacy principle, 25–29 substance of rights, 67 Counter Reformation, 278–79 creation of general principles: case law, 42 Defrenne case direct effect principle, 24–25, 49 legality, 67, 70 rule of law, 70 Defrenne II case, 70 deriving general principles: analogies with domestic law, 15–16 examples, 17–18 Digital Rights Ireland case, 28, 138, 174–76 direct effect principle, 21–22, 32, 37–38, 62, 66–67, 69, 78 Cassis de Dijon case, 24 Defrenne case, 24–25, 49 direct effect against private parties, 24–25 discrimination on grounds of nationality, 23–24 free movement of goods, 24 horizontal direct effect, 24–25, 49 implementation of Directives, 24–25 legality and, 49–51 preliminary reference procedure, 22–23 Reyners v Belgian State, 23–24 Van Gend en Loos case, 22–23 discrimination: obesity, grounds of, 355–63 sexual orientation and gender identity, grounds of, 365–94 see also obesity; sexual orientation and gender identity

Domat, Jean, 276–77 duty to give reasons, 173–74 equality: dual function: public law principle, 367 substantive law principles, 367 general principle of equality, 366–68 goals of general principle, 367 see also obesity; sexual orientation and gender identity EU customary law, 43–44, 45–47, 114–15 EU fundamental rights and, 126–28 functions of customary law, 121–22 democratic legitimacy, 122 division of powers in EU law, 122–23 jurisprudential nature of EU law, 123 supranationality of EU law, 122 general principles compared, 105–06, 125 general principles, relationship between, 107–10 hidden nature of, 106–07 limited role of customary law, 123–24 sources of customary law, 118 CJEU law making, 120–21 primary law, 118–19 qua unwritten law, 120 unwritten sources of law, 119–21 written sources of law, 118–19 types of EU customary law, 115 constitutional customary law, 115–16 historical examples, 116–17 material or institutional customary law, 115–16 subjects of customary law, 115 see also EU fundamental rights; EU law EU fundamental rights: application and interpretation of EU fundamental rights, 127 EU customary law and, 126–28 function of EU fundamental rights, 127 human rights competence, 127 Omega case, 126 sourced in domestic law, 126–27 see also Charter of Fundamental Rights EU law, 2–3 CJEU, 2–3 customary law compared, 105–06, 125 direct effect principle, 21–25 EUCFR, 2 functions of general principles: coherence, 114 gap-filling role, 114 international law, interaction with, 3 primacy principle, 25–29 proportionality, 167, 176–77 compensatory scrutiny, 171–74 Digital Rights Ireland case, 175–76 duty to give reasons, 173–74

Index four-step test, 168 investigation of facts, 171–73 judicial restraint, 168–71 manifest errors, 169–71 Schecke case, 174–75 margin of appreciation, 139 sources of general principle, 111–13 Treaty of Lisbon, 2 Treaty of Paris, 2 Treaty of Rome, 2 Treaty on the Functioning of the European Union, 2 types of general principle, 110–11 see also direct effect principle; EU customary law; EU fundamental rights; primacy principle European Convention on Human Rights (ECHR): application, 181 CJEU distrust of, 101–02 creative coherence, 80 LGBT rights, 369 normative coherence, 81–83 Opinion 2/13, 31–32, 78, 94–96, 101–02 proportionality: Denmark, 243, 245, 247–48, 250–51 France, 204–15 intensity of review, 209–15 interpretation, 204–08 rule of law and, 135–36, 141, 180–83 European Court of Human Rights (ECtHR): assessments, 189 balancing procedural values and substantive outcomes, 188–89 intensity of scrutiny, 185–86 interpretation of national law, 186–87 proportionality and margin of appreciation, 184–88 widening margin of appreciation, 188–89 subsidiarity principle, 185–86 European Court of Justice (CJEU), see direct effect principle; individual cases; primacy principle Fag og Arbejde FOA, acting on behalf of Karsten Kaltoft v The Municipality of Billund, 82, 265 arguments, 362–63 equal treatment in employment, 357–59 whether general principle prohibiting discrimination on grounds of obesity, 357–59 whether obesity a disability, 359–62 background, 356–57 fairness, 329–30 autonomy and, 349–51 CJEU, 330–31, 332 Aziz case, 345

 413

E Friz case, 332–34 Unfair Contract Terms Directive and, 345–47 distributive justice, as, 334–36 general principle of fairness, 330–31 judicial fairness, 336–39 move away from, 343–45 Penn Central case, 335–36, 348–49 Unfair Contract Terms Directive, 345–47 US contract law and, 339–42 law and policy, 342–45 France, 220 CJEU interpretation: impact on administrative judges, 199–203 lack of clarity, 201–03 Conseil d’Etat case law, 195–96 application of proportionality, 196, 197–99 non-explicit use of proportionality, 196, 197 Constitutional Council, 196 Cour de Cassation, 196 declaration of state of emergency, 215–18 ECHR, impact of, 204–15 EU law, impact of, 199–203 limited nature, 203 intensity of review, 208 declaration of state of emergency, 215–18 ECHR, impact of, 209–15 manifest error, 208–09 sanctions on prisoners, 211–15 sanctions, 209–11 judges, 218–20 manifest error, 208–09 origins, 195–96 restriction of economic liberties, 199–203 sanctions, 209–11 sanctions on prisoners, 211–15 freedom: Berlin, 271, 272, 273–74 freedom from constraint, 271 Grotius, 271, 272 Hobbes, 271 Mill, 272 negative freedom, 271–72 Berlin, 272 Grotius, 272 Pufendorf, 272 positive freedom, 270–71 Berlin, 273 Hume, 273 Pufendorf, 271, 272 Smith, 272 general principles concept, 1 comparative law, 3 defining ‘general principles’, 9–11 obligations, 10–11 rules and principles, 10–11

414 

Index

EU law, 2–3 functions of general principles, 114 sources of general principle, 111–13 types of general principle, 110–11 international law, 5–6, 18–19 Art. 38 of ICJ Statute, 5–9 deriving general principles, 15–18 function of Art. 38 general principles, 7–9 identification and application of general principles, 8 judicial practice, 11–13 case law, 13–15 positivism and natural law compared, 8–9 role of the Court, 8 subjectivity and, 8–9 interpretation of EU treaties and, 32–37 obligations, 10–11 rules and principles, 10–11 see also deriving general principles good administration principle, 44, 47, 78 coherence and: EUCFR, 86–88 Art. 41, 88–90 inter-legality, 92–93 jurisprudential coherence, 90–91 national procedural autonomy, 91–92 transnational coherence, 84–86 see also mutual trust principle Grant v South West Trains, 371–72, 374 Grotius, Hugo, 271, 272, 276, 279–81, 282–84, 286, 289, 291 Hallstein, W: legality of EU law, 56 harmonisation, 46 protection of the weaker party, 257–58 re-regulation, 257–58, 259–60 EUCFR, 261 TFEU, 261 Hobbes, Thomas, 271–72, 279, 283–84, 286, 289 Human Rights Act 1998 (UK), 135–36 fourth limb of proportionality test, 191–92 judicial interpretation, 191 margin of appreciation, 190 proportionality stricto sensu, 191–92 proportionality, 189–93 Supreme Court approach to ECtHR case law, 190–91, 192–93 Hume, David, 273, 292 impartial spectator, 295–96 impartiality principle, 232, 240 indeterminacy: rules and principles compared, 10–11, 59 strong indeterminacy, 48–49, 60–62 weak indeterminacy, 48–49, 59–60

information, availability of, 313–15, 319 consumer contracts, 313–15, 319–20, 326 perfect information assumption, 312–13 private law and, 325–26 International Centre for Settlement of Investment Disputes (ICSID), 6 International Court of Justice: Art. 38 of ICJ Statute, 5–6, 401 function of Art. 38 general principles, 7–9 identification and application of general principles, 8 judicial practice, 11–13 case law, 13–15 positivism and natural law compared, 8–9 role of the Court, 8 subjectivity and, 8–9 Committee of Jurists, 8–9 defining ‘general principles’, 9–11 obligations, 10–11 rules and principles, 10–11 positivism and natural law compared, 8–9 role of the Court, 8 international law: EU law, interaction with, 3 general principles and, 5–6 national law, interaction with, 3 sources, 5–6 Statute of International Court of Justice, 5–6, 7–9 Italy, 221–22, 240–41 Constitutional Court, 235–39 balancing rights and interests, 237–38 EU law, impact of, 231–32 increased application of proportionality, 232, 233–34 impartiality principle, 232–33 origins of proportionality principle, 222 administrative law, 223–24 fairness, 225 German influence, 222–23 necessity, 225 reasonableness compared, 223–26 proportionality in stricto sensu, 239 reasonableness and, 223–26, 232 regional administrative courts, 233–34 reverse discrimination, 238–39 unreasonableness and, 235–36 judges: duty to reason, 399–400 essential features of role, 398–400 general principles, 401–403, 404–05 language, 405–07 legitimacy, 403 fair logical result, 403–04 proportionality, 407–09 sources of law for judges, 400–401 France, 402

Index general principles as a source of law, 401–403 Germany, 402 principle and precedent, 401–403 judicial deliberation: fairness and, 351–53 judicial restraint, 138, 142–43, 168–71, 340 judicial review, 73, 146–48 bankruptcy, 152–53 criminal law, 150 defence of the realm, 150–51 excessive burden, 154–57 improvement of localities, 152 legality principle, 153 reasonable benefit, 154–57 regulatory intervention, condition of, 157–58 regulatory powers, 149 Rooke case, 154–55 social purposes, 149–50 statutory interpretation and, 153–54 statutory provisions and, 149–53 tolls and charges, 149 see also proportionality (UK) Kant, Immanuel, 270, 273, 295–96 language, 400, 405–07 legal innovation, 69–70 legal positivism, 8–9, 107, 293 legality, 53, 56 a-legality, 63–64 case law, 49–51 competence allocation issues, 71–72 constitutional legality, 73–74 constitutionalisation of EU law, 67–68 Costa case, 67 decision-making, 62–64 definition, 53–54 EU law, 54–55 Defrenne case, 67 Defrenne II case, 70 degree of legality, 57 direct effect and, 49–51 emergency powers and, 72–73 general principle, as a, 43, 44, 53–75 judicial policy making, 49 judicial review, 73, 153 legal certainty and, 48 McCulloch v Maryland, 74–75 precedents, 57–58 predictability and, 48 preliminary reference procedure and, 49, 55, 57–58 appellate system compared, 58 clarification and unification, 58, 60 incentive effect, 59 strong indeterminacy, 60–62 treaty interpretation and, 64–67

 415

Lessius, Leonard, 281, 289 Luther, Martin, 278 Lutherian theology, 282–83 Mangold case, 36, 45, 264–65, 266, 389 language, 405 over-constitutionalisation of EU law, 35 manifest errors, 169–70, 174–75 reasonable doubt, 169–70 review of proportionality, 171, 198, 208–09 margin of appreciation, 139, 391 proportionality and, 180 Strasbourg proportionality, 184–89 UK proportionality, 190 Mill, John Stuart, 272 Molina, Luis, 280–81 mutual trust principle, 43, 44 Opinion 2/13 and, 94–100, 103 see also good administration principle natural law, 8–9, 282–84 Calvinist will theories, 284–85 EU law as, 67–68 northern natural law will theories, 279–80, 292 Stair, 282–84, 286, 288 will theory and, 276–78 northern natural lawyers, 276, 277, 292 will theories, 279–80 see also Grotius, Hugo obesity, 355–56 see also discrimination on grounds of obesity Onate, Pedro, 281 Opinion 2/13, 31–32, 33, 250–51 conflict between ECtHR and CJEU, 94–95, 101–02 constitutional speech act, 94 ideological coherence, 93–102, 103 mutual trust principle, 95–102 normative coherence, 80 ‘structured network of principles’, 47, 77–78, 80–81 P v S and Cornwall, 370–71, 374–75 Permanent Court of International Justice (PCIJ), see International Court of Justice preliminary reference procedure, 22–23, 37, 42, 44, 55, 57–58, 61–62 appellate system compared, 58 benefits of, 57–59 clarification and unification, 58, 60 contract disputes, 338 discrimination on ground of obesity, 356–57 incentive effect, 59 legality and, 48–49 primacy principle, 25, 29–31, 37–38, 50, 94, 121 Costa v ENEL, 26–27

416 

Index

Digital Rights Ireland case, 28 fairness, 335 human and fundamental rights protection, 26–28 Internationale Handelsgesellschaft case, 26–27 national autonomy, conflict with, 34–37 protection of property rights, 27 reciprocity principle, 26 Stefano Melloni case, 27 supremacy and primacy compared, 28–29 principle of conferral, 73, 127, 290 private autonomy, 294–96 allocating resources, 313 behavioural psychology and economics, 313, 315 inequalities, 318–21 lack of full commitment, 313–14 lack of information, 313–15 transaction costs, 313 capacity, 256 challenges to exercise of private autonomy: macro level, 316–17 micro level, 316 contract law and, 255–56, 274–76 economic dimensions: trade and cooperation, 307–08 moral autonomy, 270 personal autonomy, 270 political autonomy, 270 protection of the weaker party and, 255–56 history of ideas, 269–70 intellectual history, 269 societal dimension, 307–08 unfair terms in consumer contracts, 256 see also freedom private law, 18, 293–94 consumer protection, 330 fairness, 330–31, 345–49 inequality and, 307, 318–20, 321–23 information available, 325–27 private autonomy, 308, 317 public law and, 165–66 see also private autonomy proportionality (Denmark), 243, 250–51 ECHR proportionality compared, 247–48 EU proportionality compared, 245–47 foreseeability, 244 labour law, 244 lack of codification of principle, 245 legal certainty, 244–45 rule of law, 244–45 Supreme Court, 244–45 trends: dialogue between national and international courts, 249–50 ECHR, increased role, 248 EUCFR, impact, 248

proportionality (France), 220 CJEU interpretation impact on administrative judges, 199–203 lack of clarity, 201–03 Conseil d’Etat case law, 195–96 application of proportionality, 196, 197–99 non-explicit use, 196, 197 Constitutional Council, 196 Cour de Cassation, 196 ECHR, impact of, 204–08 intensity of review, 208–15 sanctions, 209–11 sanctions on prisoners, 211–15 EU law, impact of, 199–203 limited nature, 203 intensity of review, 208 declaration of state of emergency, 215–18 ECHR, impact of, 209–15 manifest error, 208–09 sanctions on prisoners, 211–15 sanctions, 209–11 judges, 218–20 manifest error, 208–09 origins, 195–96 restriction of economic liberties, 199–203 see also France proportionality (generally), 133–34, 142–43, 352, 407–09 breach of, 42 CJEU, 228–31 control of outcomes, 137–38 EU law, 167, 176–77, 226–28 compensatory scrutiny, 171–74 Digital Rights Ireland case, 175–76 duty to give reasons, 173–74 four-step test, 168 investigation of facts, 171–73 judicial restraint, 168–71 manifest errors, 169–71 Schecke case, 174–75 versions of proportionality test, 228–29 incoherence, 228–29 margin of appreciation, 139, 180 origins: Danish law, 135 ECHR, 134–36 ECtHR, 134 English law, 134–35 French law, 135, 136 German law, 134 Italian law, 136 reasonableness and proportionality, 138–39 rights review and, 141–42 rule of law values, 179–84 sanctions and, 141–42 scope of application, 141–42 structured test, 140–41

Index proportionality (Italy), 221–22, 240–41 Constitutional Court, 235 balancing rights and interests, 237–38 proportionality in stricto sensu, 239 reverse discrimination case, 238–39 unreasonableness and, 235–36 EU law, impact of, 231–32 increased application of proportionality, 232, 233–34 impartiality principle, 232–33 origins, 222 administrative law, 223–24 fairness, 225 German influence, 222–23 necessity, 225 reasonableness compared, 223–26 reasonableness and, 223–26, 232 regional administrative courts, 233–34 proportionality (UK), 145–46 administrative law, 146–48 free market principles, 148 government regulation, 148 judicial review, 146–48 bankruptcy, 152–53 criminal law, 150 defence of the realm, 150–51 excessive burden, 154–57 improvement of localities, 152 legality principle, 153 reasonable benefit, 154–57 regulatory intervention, condition of, 157–58 regulatory powers, 149 Rooke case, 154–55 social purposes, 149–50 statutory interpretation and, 153–54 statutory provisions and, 149–53 tolls and charges, 149 principle of good administration, 147 proportionability and, 158–60, 162–65 democratic mandates and, 160–62 judicial proportionability, 160 legislative proportionability, 160 proportionable burden, 159 reasonableness and proportionality, 148 protection of the weaker party: EU law, 256–57 EUCFR, 257 inequality, 323 inequality in transactions, 323–25 laissez-faire approach, 302–04 market structure and contract terms, 322 paternalism and, 302–04 private autonomy and, 255–56 corrective measures, 318–21 private law, 321 adequate information and, 325–26 horizontal remedial measures, 321–22

 417

state neutrality and, 303 Treaty on the Functioning of the European Union, 257–58 unfair terms in consumer contracts, 256, 261–64 see also anti-discrimination law; contract law; private autonomy; unfair contract terms Protestantism, 278, 279, 280–81 see also Stair public law, 110, 134 equality, 367 Europeanisation of public law, 135 proportionality and, 134, 149, 152, 221 Pufendorf, Samuel von, 271, 272, 276–77, 279–80, 282–84, 286, 289 rule of law, 33, 51, 54, 56, 94, 114 Defrenne II, 70 proportionality and, 160, 179–84 rules and principles compared, 10–11, 59 strong indeterminacy, 60 weak indeterminacy, 59–60 Schecke case, 138, 174–76 sexual orientation and gender identity, 365–66, 393–94 Equality Directive, proposals for, 391–92 EU judiciary and, 370 Grant v South West Trains, 371–72, 374 other case law, 372–73 P v S and Cornwall, 370–71, 374–75 pre-Amsterdam Treaty, 368–75 Treaty of Amsterdam: protection of LGB individuals, 378–83 protection of same-sex couples, 378–83 protection of transsexuals, 376–78 recognition of LGB rights, 375–76 Treaty of Lisbon, 383–85 Associatia Accept case, 386–87, 388 fundamental right of non-discrimination on the ground of sexual orientation, 389–91 Hay case, 388, 389 Leger case, 389–90 Maruko case, 385–86, 387 Romer case, 385, 387 Smith, Adam, 148, 272, 292–93, 295–96 sources of law, 45–46, 107, 126, 128–29 EU customary law, 114–24 general principles, 110–14 general principles qua customary law, 124–28 Spanish scholastics, 276–77, 280–81, 292 Stair (James Dalrymple, Viscount Stair), 279–80 conception of liberty, 285–89 engagement, 290–91

418  late scholastics and, 280–84 will, desire, resolution and engagement distinguished, 290 supremacy principle, see primacy principle tension thesis, 298–99 demise of tension thesis, 300 see also contract law Thomasius, Christian, 276–77 trade, 307–08, 327 benefits, 310 comparative advantage, 309–10 contracts, 311 perfect information assumption, 312–13 gains from trade: macro level, 309–11 micro level, 311–13 inequalities, effect of, 323–25 market structure and contract terms, 322 non-competitive sectors, 310 productivity, 309, 310 Treaty of Lisbon, 2, 27, 36, 80, 91, 101–02, 118, 257, 259, 267 LGBT rights, 375–86 mutual trust, 96 Treaty of Paris, 2 Treaty of Rome, 2, 21–22, 33–34, 60, 78 Treaty on the Functioning of the European Union, 2 anti-discrimination law, 258–59 obesity/disability, 357–58, 363 sexual orientation, 371, 375, 378, 384 Art. 114, 257–58 direct effect principle, 22–25 duty to give reasons, 173 good administration, 91–92 harmonisation, 259–60, 261 preliminary ruling mechanism, 241 proportionality and, 229 protection of the weaker party, 257–58 UK, 145–46 administrative law, 146–48 free market principles, 148 government regulation, 148 judicial review, 146–48

Index bankruptcy, 152–53 criminal law, 150 defence of the realm, 150–51 excessive burden, 154–57 improvement of localities, 152 legality principle, 153 reasonable benefit, 154–57 regulatory intervention, condition of, 157–58 regulatory powers, 149 Rooke case, 154–55 social purposes, 149–50 statutory interpretation and, 153–54 statutory provisions and, 149–53 tolls and charges, 149 principle of good administration, 147 proportionability and, 158–60, 162–65 democratic mandates and, 160–62 judicial proportionability, 160 legislative proportionability, 160 proportionable burden, 159 reasonableness and proportionality, 148 see also judicial review; proportionality (UK) unfair contract terms: arbitration agreements, 262–63 Directives, 261–64, 332 ‘doorstep selling’, 264 harmonisation of rules, 261 non-binding nature of unfair terms, 261 protection of the weaker party, 256, 261–63 Van Gend en Loos case, 21 direct effect principle, 21–25 interpretation of EU treaties and, 32–33 language, 405–07 Vienna Convention on the Law of Treaties, 32–33, 65, 109 Wesenbecius, Mattheus, 277 will theory, 275–76 Calvinist will theory, 284–85 natural law tradition and, 276–78 northern natural law will theories, 279–80 see also contract law Wolff, Christian, 276–77